''''1CAL ARCHAEOL.OG.

TRANSAOTIONS
OF THE

AMERICAN PHILOSOPHICAL SOCIETY

HELD AT PHILADELPHIA

FOR PROMOTING USEFUL KNOWLEDGE

•

NEW SERIES-VOLUME 57, PAR'l' 8

1967

CAPE GELIDON YA: A BRONZE AGE SHIPWRECK

GEORGE F. BASS
UniverS1:ly M'useusn, University oj Pennsylvania

With the Collaboration of

PETER THROCKMORTON, JOAN DU PLAT TAYLOR, J. B. HENNESSY, ALAN R. SHULMAN,

and HANS·GUNTER BUCHHOLZ

THE AMERICAN PHILOSOPHICAL SOCIETY

INDEPENDENCE SQUARE
PHILADELPHIA

December, 1967 ,"

\
!

NAUTICAL ARCHAEOLOGY
 Copyright © 1967 by The American Philosophical Society
Printed in U. S. A.

"

I Library of Congress Catalog

Card Number 67-28643
l
 2

NAUTICAL ARCHAEOlOGY

PREFACE

This book forms the final report of the underwater The formation of such a novel excavation, involving
excavation which I directed at Cape Gelidonya, Turkey, divers and equipment fr0111 five countries, took an un-
in 1960, for the University Museum of the University of usual amount of paperwork. That Mr. Throckmorton's
Pennsylvania. I am indebted to many people for mak- work was continued on a large scale was due especially
ing the excavation and the publication possible, and a to the foresight and planning of Mr. John Huston of
brief mention of them here is but small measure of my the Council of Underwater Archaeology in America,
gratitude. ancl Miss Joan du Plat Taylor in England. In Turkey,
To Dr. Froelich Rainey, Director of the University we were grateful for the technical advice and assistance
Museum, and to the Board of Managers of the Museum, given by Mr. Daniel Siglin ancl the late Mr. Kenneth
lowe thanks not only for their making the excavation Sprague. 1\1y excellent course in diving was under the
possible, but for enabling me to remain at the Museum instruction of 1\1r. David Stith, who has since become
while preparing this report, and for a travel grant which president of the Underwater Society of America.
allowed me to visit the museums of Syria, Lebanon, The excavation would never have been a success
Greece, and Cyprus in order that I "might study related without the ingenuity and hard work, under 1110st ad-
• material at first hanel. verse conditions, of all the members of the staff: Peter
The excavation, itself, was further made possible by Throckmorton, technical adviser and photographer;
a number of foundations and individuals who bad the Joan du Plat Taylor, in charge of preservation and
vision to see the value of the work at a time when records; Frederic Dumas, chief diver: Claude Duthuit
"underwater archaeology" was too often considered only and Waldemar Illing, divers; Herb Greer, underwater
adventure: the American Philosophical Society, Mr. photographer; Honor Frost, Eric]. Ryan, and Yiiksel
Nixon Griffis, Mr. Iohn Huston of the Council of Egdemir, underwater draftsmen; Terry Ball, object
Underwater A rchaeology, and the Lucius N. Littauer draftsman; Peter Dorrell, object photographer; and
Foundation. The preservation and recording of finds Ann Bass, in charge of cleaning and cataloguing the
"vas undertaken by the Institute of Archaeology of finds during the last third of the season. Hakki
London University, with funds from the British Acad- Gultekin and Lutfi Tugrul represented the Turkish
emy ancl the Craven Fund. Our diving equipment was Antiquities Department, and were a constant source
acquired through the generosity of the U. S. Divers of aid and advice. All of the above, with the excep-
Company in America, and La Spirotechnique in France. tions of Miss Taylor, Ball, Dorrell, and Tugrul, worked
The British School of Archaeology in Athens also lent on the wreck under water. Visitors who also dived and
us all of its diving equipment, but, unfortunately, this worked on the wreck were Mustafa Kapkiu, Rasi111
did not reach us during the season because of customs Divanli, Roland J. Lacroix, Nixon Griffis, Gernolf
difficulties. Our high-pressure air compressor was Martens, and Luis Marden. At the conclusion of the
made available to us at a greatly reduced price by Bauer season, Gottfried Gruben came to Bodrum in order to
Kompressoren of IVr unich. Photography was made pos- draw the only preserved section of the hull, as it was
sible by loans of the latest underwater camera by the being disassembled. John Dereki, captain of H aj£
Nilcou Corporation of New York, and of a Polaroid Baba of Beirut, visited us for a day and took a series
Land Camera with film by the Polaroid Corporation of of depth readings with his electronic sounding devices.
Cambridge, Massachusetts. An underwater case for Following the excavation, I have conversed and cor-
the Polaroid camera was designed and constructed by responded with so many experts in various fields that
the French Navy's Undersea Research Group. Special I cannot hope to give adequate thanks to alL Among
polyethylene bags, for preserving perishable finds, were those who have provided me with helpful suggestions
supplied by Anglo-American Plastics, Ltd., of London, and information concerning previously unpublished ma-
and A ra1dite for treating wood was a gift of ClBA terial were Dr. William F. Albright, Dr. Hans-Cunter
(A.R.L.) of Cambridge, England. For preserving Buchholz, Dr. Hector Catling, Dr. V. d'A, Desborough,
cloth, we were given a supply of Gelvatol 1-30 by the Dr. Porphyries Dikaios, Dr. Marie Farnsworth, Dr.
Shawinigan Resins Corporation of Springfield; Massa- Cyrus Gordon, Dr. Sara A. Immerwahr, Dr. Machteld
chusetts, but our cloth proved to be a mirage. For Mell1l1k, Dr. Hugo Miihlestein, Dr. George My]ol1as,
various illnesses and poisonous fish stings we had drugs Mr. David Owen, :I\{r. Eric Parkinson, Dr. N. Platen,
from the Wellcome Foundation, Ltd., of London, and Professor Claude F. A. Schaeffer, Dr. Alan 13. Schul-
anti-histamine creams from Scientific Pharmacals, Ltd., man, and Dr. Arthur Steinberg. At the University of
of Cambridge. Our large, fine dinghy was lent by Pennsylvania, I have been able to receive welcomed
Baskin Soku1lu of the Turk Balik Aclamlar Kiiliibii. advice from my professors: Lloyd Daly, G. Hoger
The underwater metal detector was brought during the Edwards, Michael Jameson, Miss Ellen Kohler, ancl
last few days of the season by Luis Marden of the Rodney S. Young. I also received much information
N arional Geographic staff. about the casting of copper ingots from th" £j,,:kfMi,,):al
)

,~A(~I
,
I'
I
 2

4 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMeR. PHIL. SOC.

Division of H. Kramer and Company, who also kindly Oxford; David O'Connor of the Egyptian Section of the
analyzed several metal samples for me. University Museurn ; I-I. Vl. Hodges, Lecturer in the
A major part of the report consists of illustrative ma- Conservation of Archaeological Materials at the In-
terial drawn by 13al1,Miss Frost, Ryan, and Egdemir dur- stitute of Archaeology, London University; Frederic
ing the course of the excavation, and by Laurence joline Dumas, colleague of J acques- Yves Cousteau and for-
and Miss Susan Womer since that time. For permis- merly a civilian research member of the French Under-
sion to reproduce illustrations, I wish also to thank the sea Research Group; A. Vif. D. Larkutn, Department of
Griffith Institute, Oxford (figs. 52, 80), the Metro- Agriculture, Oxford; A. Eric Parkinson, Chemist of the
politan Museum of Art (figs. 69, 70, 73, 74, 75, 76, 77, University Museum ; and Elisabeth Crowfoot, the well-
8J, 83, 88), and the Oriental Institute of the University known authority on ancient basketry and malting.
of Chicago (ngs. 89, 90). My own parts of the report, only slightly revised here,
The task of putting this material in order fell to Miss served 3S Illy doctoral dissertation at the University of
Susan, Womer, who prepared all of the plates, maps, Pennsylvania. I must also bear the responsibility for
and diagrams in the work. Her patience and excellent the translation of Dr. Buchholz's chapter on the cylinder
j uclgment allowed me to leave the entire matter DEillus- seal from German 111to English, although parts of it •
trations in her hands, and for this I am most grateful. were translated by Waldemar IUing of the Gelidonya
1 especially want to thank those who have contributed staff and Eberhard Wolckenhaar of the graduate school
chapters or parts of chapters to this report: Peter of the University of Pennsylvania; Dr. Buchholz has
Throckmorton, now a Research Associate of the Uni- approved of the final version as published here.
versity Museum , Joan du Plat Taylor of the Institute Finally, there are three people for whom I must give
of ~\rchaeol?gy, London University; J. B. Hennessy, a special word of appreciation:
~\sslstant DIrector of the British School of Archaeology 1\11". Peter Throckmorton, whose title of technical
Jl1 Jerusalem; Alan R. Schulman, then of the Univer- director belittles the guiding role he played in all aspects
sity Museum, and now at Queens College, New York; of the expedition, from the discovery of the wreck to
H ans-Gunter Buchholz, Deutsches Archaologisches In- the conception, organization, and execution of the
stitut, Berlin; Elizabeth K. Ralph, Associate Director excavation.
of the University Museum's A pplied Science Center for My professor, Rodney S. Young, who had the con-
Archaeology, and Director of that center's Carbon-14 fidence in me to suggest that I direct the excavation
Laboratory; A. C. Western, Conservator, Ashmolean while I was still a graduate student, and whose wisdom
1\1"LlSeUI11,Oxford; Mark C. Hall, Research Chemist at and encouragement since that time have been invaluable.
the University Museum : Robert I-I. Brill, Administra- 1\105t of all, however. my thanks must go to my wife,
tor, ~cientific Research, The Corning Museum of Glass, for whose assistance in every aspect of this study, from
Corning, New York; Edward V. Sayre of tile Brook- the re-examination of all the material in Bodrum
haven National Laboratory in Upton, Long Island, New through the final preparation of the manuscript, I cannot
York; Franz R. Dykstra, Vice President of E. J. Lavine adequately express my thanks.
and Company, Philadelphia; A. Millet of the Research University Museum, Philadelphia
Laboratory for Archaeology and the History of Art, GEORGE F. BASS

It

t
(
 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK

GEORGE F. BASS

With the collaboration of PETER THROCKMORTON, JOAN DU PLAT TAYLOR, J. B. HENNESSY,

ALAN R. SCHULMAN, and HANS-GuNTER BUCHHOLZ

CONTENTS

• PAGE
Abbreviations . 6
Bibliography ..............................................•... 7
r. The Discovery. PETER 'T'lTIWCK:'.'10RTON . .. , . 14
If The Excavation. GEORGE F. BASS and PETER Tnl~oCK)JORTOl\ 21
III Condition and Treatment of Finds. J. DU PLAT TAYLOR . 40
IV. The Ship and its Lading. GEORGEF. BAss . . 44
V. The Ingots. GEORGE F. BASS ..............................•... 52
VI. The Bronzes. GEORGE F. BASS . 84
VII. The Pottery. J. B. HENNESSY and J. DU FLAT TAYLOR . 122
VIII. The Stone Objects. J. DU FLAT TAYLOR . 126
IX. Miscellaneous Finds. GEORGE F. BASS . 131
X. The Weights. GEORGE F. BASS . _ . 135
XI. The Scarabs. ALAN R. Sell CLi'llAN . 143
XII. The Cylinder Seal. HANS-GUNTER BUCHHOLZ . 148
XIII. Basketry and Malting. J. DU PLAT TAyLOR . 160
XIV. Conclusions. GEORGE F. BASS . .................•..... 163
Appendix:
1. Carbon-14 Dates for Wood. E. K. RALPH... . ... 168
2. Identification of Wood. A. C. WESTERN........ .. 168
3. Report on Copper Ingots by Spectrographic Analysis. MARK
C. HAN . 169
4. Chemical Analysis and Jsotope Ratio Examination of Lead (1.22).
ROBERT JI. BRILL. . ....... ......... ....... 170
Sa. Analysis of Bead. ROBERT H. BRILL. . . . . . . . . . . . . . . . . . . . . 170
5b. Analysis of Bead. EDWARDV. SAYRE... 171
6. Analysis of Tin Sample. F. R. DYKSTRA. . . . . . . . 171
7. Spectrographic Analyses of Pottery. A MILLET . . . . . .. 172
8. Model Ingots in Egyptian Foundation Deposits. DAYID O'CONNOR 172
Index 175

5
 2

ABBREVIATIONS

Arcbiiolopiscker Anseiqer, IDB The Interpreter's Dictionary of the Bible.

AA
AASOR Annual of the American Schools of Oriental ILN lllnstroted London News.
Research. Lst. Porsct-, Lstanbuter Forscluuujew, herausgegeben von
ActaA Acta Archaeofogica. dC1' Abteihwg l stanbui des deutschew arctuia-
AegOr H. Kantor, "The Aegean and the Orient in loqischew Lnstituts.
the Second Millennium B.c.," AlA 51 (1947). IstMilt M-ittei///Ilgen des deutschew orchiioloqischew
AlA American Journal of Archaeology. Lwstituts, Ableihmg istanbul.
AJNtt11L American l ournci of Nwmisniatics. lAOS Journal of the Amercctui Oriental Society.
ANEP James B. Pritchard, The Ancient Near East ICS Journal of Cuneiform Studies.
m Pictures Rela/ing to the Old Testament JdI lahrbuch des k. deu.tschen archiiolooischew
(Princeton, 1954). Instituts.
ANET James B. Pritchard, Ancient Near Eastem lEA J oernal
of Eg'Jlptian Archaeology.
Texts Relating to the Old Testomens (Prince- lHS Lowmat of Hellenic Stvdies.
ton, 1955). lNES l onrnal
of Near Eastern Studies.
Ant] Antiquaries' Journal. KMK2 D. Eimmen, Die Kretiscli-niykenische K'IIltur
AO Der Alte Orient, herousqeqebew von der Vor- (2nd ed., Berlin, 1924). •
derasiatischen. Gesellschaft. KMS H. Biesantz, Kretisch-snykenische Siegelbilder
AOF Arckio fii'r Orientforsctnnvq. (Marburg, 1954).
AP and P Ancient Peoples and Places Series. Marb\JVPr Marburger Winckelmann-Programrn.
ArcheI Archeoloqia Ctassica. MDIl( Mitteilungen des deuiscliew Lnstituts fii,r
ArchEph Archaiofogike Ephemeris. iigyptische Altertwmskwnde, Kairo.
AS O. Weber, Altoriental-ische Sieqelbitder I 1md MDOG JI1-itteilullgcn del' deuiscliew Orient-Gesell-
II; Der Alte Orient 17/18 (Leipzig, 1920). schaft.
ASAE Aswoles dw Service des ant·iqwites de l'Egypte. Mem. Miss. Memoires, Mission archeologique francaise au
ASAtene AmtHGrio della R. SC11Ora, Archcotooica di Caire.
Atene. Afinoica Festschrift Sundwall (Bedin, 1958).
AthMitt Nfi/teihmgen des delltschen Archiiologischen MMA Metropolitan Museum of Art.
lustit'llts, Athel1'ische Abteilllllg. .MMR~ M. Nilsson, The .Minoan-JVfyceliaean Religion
BASOR Bulletin of the American Schools of Oriental (2nd ed., Lund, 1950).
Research. AIP A. Ft1l"umal"k,The iV!:jlcenaean Pottery (Stock-
BCJ-I Buffetill de correspolldance hc/Unique. holm, 1941).
BdA Eaffel/ino d'Arte. NatCeo Natiollal Geographic Magazine.
BilJ{/lrch The Biblical Archaeologist. Newell H. H. von der Osten, Ancient Oriental Sears
BibO Bibliotheca arientahs. in the Collect-ion of Mr. Edward T. Newel!
Bibl. Nationale L. J, Delaporte, Catalogue des cylindres ori- (OIP 22, Chicago, 1934).
entaux de la Bibriotheque Nationale (Paris, OIP Publications of the Oriental Institute, Uni-
1910). versity of Chicago.
BMMA Bulletin of tlte M etropolilan MHsemn of Art, Otymp-ia, 1V Adolf Furtwangler, Die Bronzen Wid die
New Ym-k. i~brige}/. kleinerm Funde VOlt Olympia (Ber-
BritSchArch British School of Archaeology in Egypt. lin, 1890).
BSA British School at Athens, Annual. OLZ Orientalische Lileratu'rzeitung
CAH Cambridge Ancient History. OpusArch Opllsw.la Arcltaeofog-ica.
Catling C31priot Bronsework in the Mycenaean World OpusAth Opl/scula Alhcniensia.
(Oxlord, 1964). PEFQ Palest·ine E.1;ploration Fund Quarterly Slate-
CG Catalogue gbleml des ant-iq'witis egyptiennes 'ment (1869-1936, continued as PEQ).
du Musee du Caire. Pili Sir Arthur Evans, The Palace of Minos (4
CPP ]. G. Duncan, Corpus of Dated Palestinian Y.,London, 1921-1935).
Pottery (London, 1930). Poche A. Schmidt, Die SW1Jtmhmg des Henn Baron
CRAJ Comptrs rendlts de t'Acadhnie des inscriptions G~tillaUime Poche, CamHI von der Czekoslo-
et belles lettres. vakei, Aleppo (Privately published, no date).
CretS V. E. G. Kenna, Cretan Seals
(Oxford, 1960). PPS P1'oceedings of the Prehistoric Society.
CS H. Frankfort, Cylinder Seals
(London, 1939). PraktAkAth Praktika tes Akademias Athenon.
Dar.-Sag. Charles Darembel-g and Edmond Saglio, Dic- Praktika Praktika les en Athenais Archaiologikes He-
tiomwire des a_ntiqltitJs grecques et romoines tairias.
(Pads, 1919). PW POil/}ls Realcncyc!opiidie del' classichen Alter-
De Clercq Catalog-lie methodique et raisonllc de 10 col- tUJHswissenscha[ t.
lect-ion De Clercq I (Paris, 1890). PZ Priihistorische Zeitsc1lrift.
Deltton Arcliaiologiko1l- deltion. QDAP Quarterly of the Department of Alltiquities
Dcshayes Les Olltits de bronze, de !'Indus au Dall1~be in Palestine.
(IV6 an lI6 miltenaire) (Paris, 1960). RA Revlte archeologique.
Ergon To Ergo-II, les Archaiorogikes J-Jetair-las. RAssJ'r ReVile d'Ass'JlJ"iologr:e.
Et. Gret. Ehldes cretoises puMiees SOItS 10. d'irection de RBibl Re~)lI.e bibl-iqlle.
i' Ecole Fran-wise d' Athenes. RevEg'yptol ReVile eg}'ptologiqlle, Paris.
Expedition BI/./letin of tlie University Museum of the RevLig Revue dietl/des ligures.
Unive1'Sity of Pennsylvania,. RHR RevHe de l'his/oire des religions.
FKS F. Matz, Die Fr-iihkret-ischen Siegel (Berlin, RV Reallc.1;ikolt der Vorgeschichte (published by
1928) . M. Ebe,·t, 192+-1932).

6
&

VOL. 57, PT. 8, 1967J B1BLIOGRAPHY 7

S8 (Followed by name of acad-
Sit:::lIl1gberic1lte Letters used in catalogue numbers:
emy, abbreviated, e.g., SBMiinch).
SeE The Swedish Cypr'lts E:J;pedit-iolt, 1927-1931 B Bronze
(4 Y., Stockholm, 1934-1962). BI Bun ingot
v, Aulock H. H. von der Osten, Altorientalische Siege/- BM Basketry and matting
steine der Sammlung Han Silvius VOlt Au/ack In Ingot
(Uppsala, 1957). L Lead
VRS A. Moortgat, Vorderasiatische Rollsiegel p Pottery
(Berlin, 1940). Sc Scarab
VTM S. Xanthoudides, The Vaulted Tombs of Mes-
S1 Slab ingot
sara (London, 1924).
Ward-Morgan W. H. 'Ward, Cvtinders and other Nea', East- St Stone
ern Seals in the Libmry of J. Pierpont Mor- W Weight
galt (New York, 1909). Wd W'ood
ZAssyr Zeitschriit !1'ir Assyrialogie.
ZDPV Zeitsclirijt des deutschen. Patiistina- V ereins. Dimensions (always given in meters) :

.Map and Catalogue Abbreviations: Pres . Preserved

• G Gulley area D. Diameter
S Sandy area H. Height
P Platform area L. Length
M Miscellaneous area Th. Thickness
E Extension area W. \Vidth

BlBLIOGRAPHY OF WORKS C!TED

ABERG, NILS. 1933. Bronseeeitliche fJnd [riiheiseneeitliclie -- 1963. "The Promise o l Underwater Archaeology." Ameri-
Chronologie: IV, Griechentand (Stockholm). can Scholar 32, 2.
AGRICOLA, GEORGIUS. De Re Metal/ica (trans. by Herbert C. -- 1963. "Mycenaean and Protogeometric Tombs in the Hali-
and Lou H. Hoover, New York, 1950; reprinted from The karnassos Peninsula." AJ A 67.
Mining Magazine, London, 1912). First published in 1556. -- 1965. "The Asheratv: A Submarine for Archaeology."
AKURGAL, E. 1946. Remarques stvlistiques sur les reliefs de A'rchaeology 18.
Malatya (Istanbul). -- 1966. Archaeology Under Water (AP and P; London,
AL, B. S. A. 1942. De mannelyke en de vrouwelijke Godheid New York).
van de Boomcultus in de -minoische Godsdiewst (Amsterdam). BI~AUFORT, FRANCIS. 1818. Karanunwi, a Brief Description of
ALBRIGHT, WlJ.LIAM F. 1932. "Joint Excavation at Tell Beit the South Coast of Asia Minor and of the Remains of An-
Mirsim." AlA 36. tiqnity (2nd ed., London).
-- 1935. "Palestine in the Earliest Historical Period." Jour. BELOCH, J. 1964. Griechiscke Geschichte 1, 2 (2nd ed., Strass-
Palest. Oriental Soc. 15. burg) .
-- 1936-1937. "Tell Beit Mirsim 11." AASOR 17. BENOIT, FERNAND. 1961. L'Epove du Grand Cong/olle Ii 1I1ar-
-- 1941. "New Light on the Early History of Phoenician seille (Go1fia suppl. 14, Paris).
Colonization." BASOR 83 (Oct.). -- 1962. "Nouvelles epaves de Provence (III)." Ga.llia 20,
1941-1943. "Tell Beit Mirsirn IlL" AASOR 21-22. fasc. 1.
-- 1949. The Archaeology of Palestine (Harmondsworth). BENSON, J. L. J960. "Bronze Tripods from Kourion." Greek,
-- 1950. "Some Oriental Glosses on the Homeric Problem." Roman, and Bvcontine Studies 3, 1 (winter).
AlA 54. -- 1961-1963. "Coarse Ware Stirrup Jars of the Aegean."
-- 1961. "The Role of the Canaanites in the History of Civi- Berytns 14.
lization." The Bible and the Ancient Near East (essays in BENTON, S. 1964. "No Tin from Kirrha in Phokis." Antiqu·ity
honor of William Foxwell Albright, G. Ernest Wright, ed., 38.
New York). BERAN, THoMAs. 1959-1960. "Hethitische Rollsiegel del' Gross-
ALDRED, C. 1963. "The Parentage of King Siptab." lEA 49. reichszeit II." lstMitt 9/10.
ALEXIOU, S. 1964. "Neue Wagendarstellung aus Kreta." AA -- 1962. "Bericht tiber die Ausgrabungen in Bogazkoy im
1964. Jahre 1961." Ticrk Arkeuloji Derqisi 11, 2.
ASTOUR, MICHAEL C. 1964. "Second Millennium B.c. Cypriot BERRIMAN, A. E. 1955. "Some Marked Weights in the Petrie
and Cretan Onomastica Reconsidered." lAOS 84. Collectioll." lEA 41.
BmsANTz, H. 1954. Kretisch-mykenische Siege/bilder (Mar-
BARNETT, RICHARD D. 1956. "Ancient Oriental Influences on burg).
Archaic Greece." The Aegean and the Near East (studies BITTEL, K. 1939-1941. "Bemerkungen tiber einige in Kleinasien
presented to Hetty Goldman, S. Weinberg, ed., New York). gefullclene Siegel." AOF 13.
BASS, G. F. 1961. "A Bronze Age Shipwreck." Expedition BITTEL, K., and A. HERMANN. 1934. "Grabungsbericht Her-
3,2. mopolis 1933." MDIIC 5.
-- 1961. "The Cape Gelidonya Wreck: Preliminary Report." BLAKE\YAY, A. 1933. "Prolegomena to the Study of Greek
AlA 65. Commerce with Italy, Sicily and France in the Eighth and
-- 1962. "Underwater Excavations at Yassi Ada: A Byzan- Seventh Centuries B.c." BSA 33.
tine Shipwreck." AA 1962, BLEGEN, C. W. 1937. Prosymlla. The I-lelladic Settlement pre-
-- 1963. "Underwater Archaeology: Key to History's Ware- ceding the A'rgi'lle H e1'ae1l1'1!(Cambridge).
house." NatCeo 124, 1. BLEGEN, C. W., et al. 1950-1958. T1'o)' (4 v., Princeton).
 8 CAPE GELIDONYA: A BRONZE AGE SH1PWRECK [TRANS. A~IEI{. run.. SOC.

BLINKENBERG, C. 1931. Lintios, [ouilles et recherches 1902- -- 1951. Notes on the Prehistoric Metallurgy of Copper and
1914 I (Ber-lin). Bronec in the Old World (Pitt Rivers Museum, Occasional
BCISS, F. ]. 1898. A Afound of Many Cities or Tell-el-He.~y Papers on Technology 4, Oxford).
Exctnxued (2nd ed., London). COJ.1PARETTE, T. LoUIS. 1918. "Aes Signatum.' Al NlI11t 52.
BOARDMAN, J. 1961. The Cretan Collection in Oxford: the CONTENAU, G. 1922. La G!},pfiqlte svro-luttite (Paris).
Dictaetni Cave alld Lrow Age Crete (Oxford). -- 1949. La Ciuilisotiow phCnidenne (Paris).
-- 1963. Island Gellis. A Study of Greek Seals in tile Geo- COOK, J. :.\'1., and ]. BOARDMAN. 1954. "Archaeology in Greece,
metric alld Early Archaic Periods (Society for Promotion 1953." lHS 74.
of Hellenic Studies, Suppl. 10, London). CROS, GASTON, et ol. 1910. Nowuelles [ouiltes de Tello (Paris).
BONNET, H. 1952. Reatlexikow tier iigyptischm Retiqions-
geschiclifc (Ber-lin}. DANIEL, JOHN FRANKLIN. 1941. "Prolegomena to the Cypro-
BOSANQL'ET, R. C. 1901-1902. "Excavations at Palaikastro L" Minoan Script." AlA 45.
BSA 8. DAVIDSON, GLADYS R. 1952. Corinth XII, The Minor Objects
BOSSERT, E.-M. 1954. "ZUI" Datierung del' Graber von Arkesine (Princeton) .
auf Amorgos." Festschrift jiir Peter Goessler (Stuttgart). DAVIES, NINA Of; G., and ALAN H. GARDINER. 1915. The Tomb
-- 1960. "Die gestempeltcn Verzierungen auf Iriihbronze- of Amenenitist, no. 82 (The Theban Tomb Series, 1st
zeitlichen Getassen del' Agfiis." JdI 75. Memoir, Egyptian Exploration Fund, London).
BOSSERT, II. 'I'u. 1930. Geschiclne des Kuustqetuerbes III DAVIES, NINA DE G., and ALAN H. GARDINER. 1926. The Tomb

--
(Berlin) .
1942. Altonototicn: Kunst WId Handwerk: ilt Kteinasicn
of H1/.J' (Egypt Exploration Society, Theban Tombs Series, •
4th Memoir, London).
von del' Anjii/Lgm bis :::um volligen Aufgellen in del' DAVIES, KINA DE G., and NORMAN DE G. DAVIES. 1933. The
griecllisclle Kultur; (Berlin). Tombs of Menkheperrasonb, Amewmose, a.nd Another (Egypt
BRl-:A, L. BERNABO. 1957. Sicily before the Greeks (AP and Exploration Society, Theban Tombs Series, 5th Memoir,
P, London). London) .
-- 1957. "Recenti scavi a Poliochni nell'isola di Lemnos." DAVIES, NDIA DE G., and NORl\IAN DE G. DAVIES. 1940. "The
BdA 42. Tomb of Amenmose (no. 89) at Thebes." lEA 26.
BlI.Ol\'lEHEAD, C. N. 1954. "Mining and Quarrying." A History DAVIES, NORMAN DE G. 1902. The Rock Tombs of Deir el
of Tecllllology T (Oxford). Gebrtnoi II, T01nb of Ztuc oiui Tombs of the Northern Gr01lp
BUCHHOLZ, HANS-GUKTER. 1958. "Der Kupferhandel des (Egypt Exploration Fund, Archaeological Survey of Egypt,
zweiten vorchristlichen Iabrtaosends im Spiegel del' Schn lt- 12th Memoir, London).
forschung." ill1'lIoica (Berlin). -- 1903. The Rock Tombs of El A,ita-rna I, The Tomb of
-- 1959. "Keftiubarren und Erzhandel im zweiten vorchrist- Jl[e1"'y;ra (Archaeological Survey of Egypt, 13th '.Memoir,
lichen ]ahrtausend." PZ 37. London) .
-- 1963. "Dreifussschalen des agaischen Kulturkreises und -- 1905. The Rock Tombs of El Amorlla II, The Tombs of
ihre Beziehungen zum Osten." Jdl 78. Pallehessy O1tdMeryra [[ (Archaeological Survey of Egypt,
BURTON-BROWN, T. 1955. The Coming of hon to Greece 14th Memoir, London).
CWincle, Chesire). -- 1905. The Rock Tombs of El AlIlarna III, The Tombs
of Hu)'a a,lId Ahmes (Archaeological Survey of Egypt, 15th
CARTER, H., and P. ::\fEWBERRY. 1904. The Tomb of TllOutmo- Memoir, London).
sis lV nVestminster). -- 1920. "The \;York of the Tytu5 Memorial Fund." BMMA
CASSON, STANLEY. 1937. Auciellt Cyprus, Its Art and A'rchac- 15, 7, ,ect. II (Dec.).
ology (London). -- 1922. The Tomb of PWJle1llre at Thebes (MMA Egyptian
Catalogue 'IIlethodiqltc et raisoJl,Jle de la collection De Clercq I Expedition Publications, Robb de Peyster Tytus Memorial
(Fa,·;s, 1890). Series, vol. II, Kew York).
CATLING, HECTOR '1\'. 1963. "Minoan and Mycenaean Pottery: -- 1925. The Tomb of Two Scnlplot's at Thebes (MMA
Composition and Provenance." Archaeometry 6. Publications, Robb de Peyster Tytus Memorial Series, vol.
-- 1964. Cypriot BrOllzework in the MycenaeaH T¥orhl IV, New York).
(Oxford). 1926. "The Graphic \¥ork of the Expedition." BMMA
CATLlNG, HECTOR ,"V., and V. KARAGEORGI-llS. 1960. "Milloika 21, 3, pa,·l. II (March).
in Cyprus." BSA 55. -- ]932. "The \iVork o[ the Graphic Branch of the Expedi-
CATLING, HECTOR VV., et al. 1963. "Correlations between Com- tion." EMMA 27, 3, sect. II (March).
position and Provenance of Mycenaean and 1trinoan Pottery." -- 1933. The Tomb of Nefer Hotep at Thebes I (MMA
BSA 58. Egyptian Expedition, New York).
CHAI'OUTHIER, F. 1937. "L'Orient et la Crete a
propos d'un 1935. "The Graphic 'Work of the Expedition." BMMA
cylinclre cretoise." ArchEph 1937. 30, 11, sect. II (Nov).
-- 1951. "De l'origine dl! prisme triangulaire dans la glyp- -- 1935. Pa.intings jrom the Tomb of Rekh-1JIi-ri!' at Thebes
tique minoenne." BSA 46. (MMA Egyptian Expedition Publications, vo1. X, New
CHEVRIER, M. HENRI. 1928. "Rapport sur les travaux de Kar- York).
nak (1927-1928). ASAE 28. -- 1943. The Tomb of Rekh-11Ii-re' at Thebes (MMA
CHfLDE, V. GORDON. 1963. The Broll:::e Age (New York). Egyptian Expedition Publications, 11, New York).
First published in 1930, reprinted by permission o[ the Cam- DAVIES, NORMAN DE:G., and R. O. FAULKNER. 1947. <lA Syrian
bridge University Press. Trading Venture to Egypt." J EA 33.
CHRISTOPHE, LOUIS A. 1951. UNotes geographiqucs it propos DAVIES, O. 1929. "Two NOI"th Greek Mining Towns." lHS 49.
des campagnes de Thoutmosis IlL" RevEgyptol. 6. -- 1928-1930. "The Copper Mines or Cyprus." BSA 30.
CLARK, j. G. D. 1952. Prehistoric Ew·ope, the Ecollomic Basis DAVIES, ROBE1(T"H. 1955. Deep Diving alld S1/bmarilIC Opera-
(London) . tiolls (London).
COGHLAN, H. H. 1939-1940. "Prehistoric Copper and Some DAVIS, T. lvf., and H. CARTEH. 1908. The TOlJlo oj J-TalshopsItIi
Experiments in Smelting." T1·ansactions of the Newcomer (London) .
Society for the Study of the History of Engineering altd DAWKINS, R. M. 1903-1904. "Excavations at Palaikastro Ill."
T edtl/ology 20. BSA 10.
 '·OL. 5i, 1''1'. 8, 196i] BIBLlOGRAPHY 9
DELAPORTE,L. J. 1910. Catalogl/e des cvtindres orientanx de 10 -- 1932. "Zur Altersfrag'e del' rhomboiden Eisenluppen."
Bibliotlieque Nationa!e I (Paris). Cerniania 16.
-- 1932. Catalogue des cvlindres orientanx, L011vre II FORSDYKE,E. J. 1925. Catalog lie of the Creek and Etruscow
(Paris). Vases in the British Mvseuni I, pt. 1 (London).
DEMARGNE,P. 1947. La Crete dedalique (Paris). -- 1926-1927. "The Mavro Spelio Cemetery at Knossos."
D~;ONNA, \,y. 1938. Le Mobiiier defien, exploration archeolo- BSA 28.
gique de Delos par l'Ecole Prcncoise d'Athenes 18 (Paris). Fox, PENELOPE..1951. Tutonlstunnen's Treasure (London).
DESBOROl:GII,V. R. D'A. 1964. Tire Last Mvcenaoans and their FRANKEN, H. J. 1965. "A Bronze Age Shrine and an Unknown
Successors (Oxford). Script; Iron Age Jordan Village." ILN 17.4.196j.
DJo;SHAYES,JEAN. 1960. Los Outils de bronoe, de I'Lndws Olt FRANKFoln, H. 1936-1937. "Notes on the Cretan Griffin."
Danube (lVo an n° MillblQire) (Institut Francais d'Archeo- BSA 37.
logic de Beyrouth, Tome LXXI, Paris). -- 1939. Cylinder Seals, A Docnmentory Essay on the Art
DESSENNE,A. 1957. Le SpJIil1x, elude 1collographique (Biblio- and ReligioH of the AllciCHt Near East (London).
theque des Ecoles Francaises d'Athenes et de Rome, fasc. 186, FRODIN, 0., and A \V. PERSSON. 1938. Asine. Results of lite
Paris). Swedish Excavations 1922-1930 (Stockholm).
DrEHL,ERIKA. 1965-1966. "Fragmente aus 'Samos IL" AA FRONDEVILLE, GUY DE. 1956. Les Visiteurs de la mer (Paris).
1965. FROST, HONOR. 1963. Under the Mcditerraneaw (London).
D1KAIOS, PORPHYRIOS. 1953. A Giade to the Cyprus Museu'lIL FUG.MANN, E. 1958. Hcnm, [onitles et recherches 1931-1938
• (2nd ed., Nicosia) . II, I (Copenhagen).
-- 1962. "The Excavations at Engomi, Cyprus by the De- FURT\VI\NGLER,ADOLF. 1896. Beschreibunq der geschllittenel~
partment of Antiquities." Bericlu iiber den V l nternsuionclen Steine £m Antiquorillm (Berlin).
Konqress fiir Vor-1II1d Friihgeschichte, Hamburg, 1958 (Ber- 1900. Die Antikcn Gemmen (3 v., Leipzig and Berlin).
lin). -- 1906. Aigilla l (Munich).
-- 1962. "The Bronze Statue of a Horned God from En- -- 1913. Kleine Schriiten II (Munich).
komi." AA 1962.
FURTWiiNGLER,AOOLF, and G. LOESCBCKE. ]886, M)Ikenische
-- 1963. "The Context of the Enk0111i Tablets." Ktuimos Vasew (Berlin).
2, I. FURUIlfARK,ARNE. 1941. The Chronology of Myce1laean Pot-
DIRINGER,DAVID. 1948. The Alphabet, A Key to the Histor-y /ery (Stockholm).
of Mankind (New York). -- 1941. The Mj'CellOCaHPot/cry: analysis and classification
DORPFELD,YV. 1902. Troia ?Iud Ilion (Athens). (Stockholm) .
DOR, LEOPOLD,et aZ. 1960. Kirrha (Paris). -- 1950. "The Settlement at Ialvsos and Aegean History
DHIVEI~,G. R. 1954. Sell/it£c Writing from Pic/ogra-Ph to c. 1550-1400 B.c." OpusArch 6, .
A IpJwbet (revised, London).
D Un-I:MLER,F. 1886. "l1!,ittlleihmgen von den grieclt£sclte Inse!Il:'
Atll.Ann II. GALLING,KURT. 1934-1937. Biblisches Realfexikon (Tiibingen).
DUGAN, JAMES. 1956. Man Under the Sea (New York). GARDINER,ALAN. 1950. Egyptian CramHUlr (2nd ed., London).
-- 1961. Egypt of the Pharaohs (Oxford).
DUNAND, M. 1937-1954. Fonilles de Byblos I and II (Paris).
GARSTANG,J. 1953. Prehistoric Mersin (Oxford).
DUNCAN, J. G. 1930. Corpus of Dated Palest-inion Pottery
(BritSchArch 49, London). GILBERT,PERRY, ed. 1964. Shar/~s alld Survival (New York).
GILL, M. A. V. 1961. "Note on the Hagia Pelagia Cylinder."
DUNHAM, Dows. 1955. The Royal Cemeteries of J(mh II.
NlIri (Boston). Bull. Instit. Class. Stud. London 8.
GJERSTAD,E. 1926. Studies on p,.ehislor·ic Cyprus (Uppsala).
DUSSAUD, H.. 1914. Les Civilisations prehellmiqlles (2nd ed.,
Paris). GLOTZ,GUSTAVE. 1925. The Aegean Civili::ation (New York)
and La Civilisation egeeune (2nd ed., Paris, 1937).
-- 1939. "Rapports entre la Crete ancicnne et 1a Babylonie."
Iraq 6. GLUECK,NELSON. 1935. "ExplOl'ations in Eastern Palestine 11."
AASOR 15.
GOEDICKE,H. 1961. "Die Siegelzylinder von Pepi L" MDIK
EMPLUOt,-, B., ef al. 1962. The Ne'w Scimce of Skin and SCflba 17.
Diving (New York).
GOLDMAN,H. 1956. Excavatiolls at Cadii Kule, Tarsus II
EncyclopMic P1'islllfl du lIIolide .fOlls-warilie (Paris. 1957). (Princeton).
EVANS, ARTHUR]. 1900-1901. "The Palace of Knossos." GORDON,Cvnus H. 1962. Before fhe Bible, Tire Com1lt01l
BSA 7.
Backgroul1d of Greek and I-febrew C1vili:::atiolls (New York).
-- 1906. "Minoan 'Weights and Mediums of Currcllcy, from GOWLAND,VV'. 1899. "The Early Metallurgy of Copper, Tin,
Crete, Mycenae, and Cyprus." Corolla Nrtmismatica (Fest~ and Iron in Europe, as Illustrated by Ancient Remains, and
schrift Head, Oxford).
the Primitive Processes Surviving in Japan," Archaeologia
-- 1909-1952. Scripta A/'illoa (2 v., Oxford). 56 (2nd series, 6).
-- 1910. "Mycenaean Tree and Pillar Cult and its Medi- GRACE,VIRGINIA. 1956. "The Canaanite ]ar." The Aegealt and
terranean Relations." JJ-fS 2l. the Near East (studies presented to Hetty Goldman, S.
-- 1921-1935. The Palace of M1'/WS (4 v" London). Weinberg, ed., New York).
EVANS, JOEIN. 1881. The Ancient Bronze Imp/eUlel/ts, l¥eap- GRAi'\T, ELlIIU. 1931-1932. Ailt Shcllls Excavations I, II
ous, and Omomellts of Creat· Bn:tain alld Ireland (New (Biblical and Kindred Studies, nos. 3,4, Haverford College,
Yo'k). Haverford, Pennsylvania).
-- 1934. RU/lleilelr, being Ai}~ Shems E.xcavatio1lS III (Hav-
FnUJEN, D. 1924. Die Kretisch-mykel/ische J(tl1tur (2nd cd., er ford).
Berlin) .
GRANT, EUHu, and G. E. \\fRIGDT. 1938. Ailt Shems Excava-
FrrzGERALD, G. M. 1930. The Fom' Canaanite Temples of tions IV (Pottery) (Haverford).
Beth Shan II, Part II. The Pottery (Philadelphia). GREENBERG, J. 1963. Underwaler Photography Simplified (2nd
-- 1934. "Excavations at Beth-Shan in 1933." PEFQ 1934. ed., Coral Gables, Florida).
FaMEs, R. J. 1950. Metalll/rgy ilL Antiquity (Leiden). GRETSS,E. A. M. 1957. Anatomical Identi{tca.tioll of Some
FORRER,ROBERT. 1920. Die Provillzeinteiltlllg des Ass}'rischen Anciellt Eg)'plialt Plant Materials (Memoires de l'Institute
Reiches (Leipzig). d'Egypte, Cairo).
 10 CAPE GELIDONYA: A BRONZE AGE SH1PWRECK [TRANS. AMER, PHIL. SOC.

GRUMACH, ERNST. 1965. "Theban und das A!ter von Linear KANTOR, HELENE J. 1947. "The Aegean and the Orient in the
B." Kadanos 4. Second Millennium B.c." AJ A 5l.
GUlDO, MARGARET. 1963. Sardinia CAP and P 35, London). KARAGEORGHIS, V. 1962. Treasures in I.he C")IPruS ,M1tseUm
GUY, P. L. 0., and R. M. ENGBERG. 1938. lIt!egiddo Tombs (Republic of Cyprus, Dept. of Antiquities, Picture Book 1).
(alP XXXIII, Chicago). KARO, GEORGE. 1916. "Archaologische Funde im Jahre 1915."
AA 1916.
HAAS, HANS. 1954. 111en and Sharks (New York). 1930. "Schatz von Tiryns." AthM'itt 55.
HALL, H. R. 1915. Aegean Archaeology (London), -- 1948. "Art Salvaged from the Sea." Archaeology 1.
-- 1928. The Civili:::ation of Greece 1"n the Bronze Age -- 1959. Greiien. am Thron. Erineervnoew an Knossos
(London) . (Baden-Baden).
HAMILTON, R. \V. 1935. "Excavations at Tell Abu Hawam." K,\SCHNITZ-WEINBERG, G. 1949-1950. "Zur Herkunft der
QDAP 4. Spirale in der .A,gais." P Z 34/35 I.
HANFMANN, G. M. A. 1951. "The Bronze Age in the Near KENNA, V. E.G. 1960. Cretan Secds: with a Cctaloque of the
East: A Review Article." AlA 55. Minoan Gems in the Ashmolean Mnseuni (Oxford).
-- 1960. "Roebuck, Ionian Trade and Colonization." Gno- KIRK, G. S. 1949. "Ships on Geometric Vases." BSA 44.
mon 32. KOSAY, Hx xnr Z. 1951. Atocc Hiiyiik Kozisi, 1937-1939
HARDEN, DONALD. 1962. The Phoenicians CAP and P 26, (Ankara) .
New York). KUNZE, E. 1931. Kretische Broneereliefs. (Stuttgart).
HAWKES, C. F. C. 1936-1937. "The Double Axe in Pre-historic KUPPER, ].-R. 1959. "Sceaux-cylindres du temps de Zimri-
Europe." BSA 37.
HAYES, WILLIAM C. 1959. The Scepter of Emlpt II (Cam-
Lim." RAssyr 53. •
bridge, Mass.).
LAL, B. B. 1953. "Prctohistoric Investigation." Ancient India
HAYES, WILLlAi\{ C, et al. 1962. "Chronology: Egypt, West- 9.
ern Asia and the Aegean Bronze Age." CAH fasc. 4. LAMON, R. S., and G. M. SHIPTON. 1939. Megiddo I (OIP
HAZZIDAKIS, J. 1921. Tvlissos a
l'ipoque niinoennc (Paris). 42, Chicago).
-- 1934. "Les Villas minoennes de Tylissos." Et. cret. 3. LEVI, DORa. 1953-1954. "Una gemma cretese-micenea.' Arch-
HELCK, 'IN .: 1962. "Die Beziehungen agyptens zu Vorderasien Epi< pt. 3.
im 3. und 2. .lahrtausend v. Chr." Agyptologische Abhand- LILLlU, GIOVANNI. 1958. "Ciottdo incise prenuragico della
hmgen 5 (Wiesbaden ): Grotta Sarda di San Michele d'Ozieri-Sassari." Arch.Cl 10.
HERZFELD, E. 1941. Iran in the Ancient East (London).
LORIMER, H. L. 1950. Homer and the Monwments (London).
HrcKMAN, HANS. 1951. "Le Tambour-in rectaugulaire du LOUD, GORDDN. 1948. Megiddo II (OIP LXII, Chicago).
nouvel empire." ASAE 51. LUCAS, A. 1948. Ancient Egyptian Materials and Industries
HILl., GEORGE. 1940. A History of Cyprus (Cambridge). (London).
HOLSCHER, U. 1939. The Temples of the Eighteenth Dynasty LULLIES, R. 1955. Eine Sammlung qriecliisclier Kleininmst
(OIP XU, The Excavation of Medinet Habu, vol. II, (M iinchen ) .
Chicago).
-- 1941. The Mortuary Temple of Ramus Ill, part I
MACALISTER, R. A. STEWART. 1912. The Exccuatiow of Geser
(OIP LIV, The Excavation of Medinet Habu, vol. III,
Chicago) . II (London).
HOGARTH, D. G. 1920. Hittite Seals with Particular Reference MACDONALD, E., ct al. 1932. Beth Pelet II (BritSchArch 52,
London) .
to the Ashmolean Collection (Oxford).
MAcKA Y, E. J. H. 1935. The Indus Civiti=ation (London).
HOOD, M. S. F., and P. DE JONG. 1952. "Late Minoan Warrior-
-- 1938. Further Excavations at Mohenjo-Daro I (New
Graves from Ayios Ioannis and the New Hospital Site at
Knossos." BSA 47. Delhi).
HORN, S. H. 1962. HScarabs from Shechem." INES 21. MACKAY, E. J. H., F. PETRIE, and M. MURRAY. 1952. Ancie'llt
HORNUNG, E. 1964. "Untersuchungen zur Chronologie und
Casa V and City of Shepherd Kings (BritSchArch 64,
London) .
Geschichte des Nellcn Reiches." A'gyptofogische Ablzalldhmg
11 (Wiesbaden). :MACLEISH, K. 1961. "Sea Search into History at Caesarea."
Life 50 (May 5).
HUTCHINSON, R. W. 1939-1940. "Unpublished Objects from
Palaikastro and Praisos." BSA 40. MAIURT, A. 1923-1924. "Jalisos-Scavi della Missionc Archeo-
1948. "Notes on Minoan Chronology." Antiqnity 22. logica Ttaliana a Rodi." ASAtelle 6/7.
1950. "Battle-axes in the Aegean." PPS (new series) N[ARINATOS, SPYRJDON. 1933. "La Marine creto-mycenienne."
16. BCH 57.
1962. Prehistoric Crete (Harmondsworth, Baltimore). -- 1962. "Zur Frage der Grotte von Arkalochori." Kad-
HUXLEY, G. 1961. Crete and the Lllwians (Oxford). mos 1.
MASSON, OLIVIER. 1957. "Cylindres et cachets chypriotes
IMMERWAHR, S.'1RA A. 1960. "Mycenaean Trade and Coloniza- portant des caracteres chypro-minoens." BCH 81.
tion." Archaeology 13. MATTHIAE, P. 1962. Ars Syra (Centro di stlldi semitici, Serie
archeologica 4, Rome).
J ACOPf, G. 1930-193], "Nuovi scavi nella necropoli micenea di MATZ, FRIEDRICH. 1928. Die friihkretischen Siegel (Berlin).
Jalisso." ASAtene 13/14. -- 1950. "Zur agaischen Chronologie der fruhen BrOllze-
JAROBJDES, S. 1956. "'AlIauKa<jJr, ~IJK1I"ai·Kwll rarpwlI II€pa:r~ ...." zeit." H1:storia l.
Praktika 1953. -- 1954. Handb1lclt der ArcJriiologl:e II (Munich).
]lRKU, ANTON. 1957. Dl:e Welt der Bibel; iiini laltrtausende MAXWELL-HySLOP, R. 1953. "Bronze lugged axe- or adze-
in Pali:istil1a-Syrien (Stuttgart). blades from Asia." Iraq 15.
-- 1963. Geschichte Paliistina-Syriens im orientalischeu Al- MAY, HERBERT GORDON. 1935. Material Renwins of tlte
tertu1Jt (Aalen). Megiddo C"lt (alP XXVI, Chicago).
JUCKER, HANS. 1965. "IPromenade archcologique' dllrch die MCCOWN, CHESTER CARLTON. 1947. Tell en-Nasbelt I (Ber-
Ausstel1ung der Sammlung Kofler im Kunsthaus Zurich." keley and New Haven). .
Alllike KUlIst 8. 1I1ed-inet Hablt V. 1957 (OIP 83, Chicago).

l
 -
VOL. 57, PT. 8, 1967J BlBLlOGRAPHY 11
MELLAART, JAMES. 1959. "The Royal Treasure of Dorak." PERSSON, AXEL \V. 1931. Tire Royal Tombs at Dendra, near
ILN 235. kHdea (Lund).
MELLlNK, M. J. 1955. "Review of Biesantz, Kretisch-M'y- -- 1942. NC'lC/ Tombs at Dendra Near Midea (Lund).
kenisctie Sieqelbikier." Al A 59. PETRIE, W. M. FLINDERS. 1886. Nmskrctis I (Egypt Explora-
__ 1959, 1963. "Archaeology in Asia Minor." AlA 63 tion Fund, 3rd Memoir, London).
and 67. -- 1888. Nebeslieti alld Dejennetv (Tanis II) (Egypt Ex-
MERUN, A, and L. POLNSSOT. 1930. Crtueres at candilabres ploration Fund, 4th Memoir, London).
de morbre trowues en 'Iller pres de Jlt{alzdia (Tunis). 1891. Ilalnni, KaJ11I1t and Gvrob (London).
MILANI, L. A. 1912. Il R. Micseo Archeolooico di Firenee -- 1897. Si« Temples at Thebes 1896 (London).
(2 v., Florence). -- 1902. AbJdos I (Egypt Exploration Fund, 22nd Memoir,
MILCHHOFER, A. 1876. "Die Ausgrabungen in Mykene." London).
AthM.£tt 1. ~- 1903. Abvd os 1I (Egypt Exploration Fund, 24th Memoir,
1v10D£, H. 1950. "Dber einige syrische Siegelzylinder aus London) .
Schweizer Samrnlungen." Archiv Orientobw 18, 4. -- 1917. Tools and Weapows (Brit Sch.Arch in Egypt, 3D,
MONTELIUS, O. 1924. La Greco prectassiqne I (Stockholm). London) .
MONTET, P. 1952. Las Eniomes de Tallis (Paris), -- 1925. Buttons and: Design Scarabs (BritSchArch in
MONTGOMERY, J. A. 1934. "Ras Sharura Notes-IL" lAOS 54. Egypt, 38, London).
MOORTGAT, A. 1932. Die bikiende KHIlSt .. (Berlin). -- 1926. Ancient Weights alld Measures (BritSchArch in
... -- 1940. Vordcrasiatisctie Rollsiegei
MOOR'rGAT-CORRENS, U. 1955. "Neue Anhaltspunkte
(Berlin).
zur zeit-
Egypt, 39, London) .
-- 1927. Objects of Daily Use (BritSchArch in Egypt, 42,
lichen Ordnung syrischer Glyptik." ZAssyr 51. London).
MUHLESTEINE, H. 1964. "Review of Ktunnos L" Kratylos 9. 1928. Germ- (Brit Sch.Arch in Egypt, 43, London).
MULLER, V. 1918. "Die Ziernadel aus dem III. mykenischen -- 1930. Beth-Pctet I (BritSchArch in Egypt, 48, London).
Schachtgrab." AthM'itt 43. -- 1931-1934. Ancient Gcea I-IV (BritSchArch in Egypt,
1927. "Studieu ZUf kretisch-mykenischen Kunst 11." 53-56, London).
Jd142. -- 1954 ed. "Measures and Weights, Ancient." Ellcyclope&ia
MURRAY, A. S., et al. 1900. Excavations in: Cyprus (London). Britannica 15.
MYLONAS, GEORGE. 1962. "Three Late Mycenaean Knives." PETRIE, \V. M. FLINDlms, and ]. G. DUNCAN. 1906. Hvksos
AJA 66. and Israelite Cities (Brit Scb.Arch in Egypt, 12, London).
MYRES, ]. L. 1946. "The Minoan Signary." IRS 66. PLATON, N. 1955. Guide to the Archaeolog-ical Mwscmn of
Herocliow (Herakleion).
KAGEL. \V., and E. STROMMENGER. 1958. "Alalalj und Siegel- -- 1963. "A New Minoan Palace." Archaeology 16, 4.
kunst." ICS 12. PLATON, N., ct at, 1964. "Oriental Seals from the Palace of
NAVILLE, EDOUARD. 1907, 1908. The Kith Dynasty Temple at Cadmus; Unique Discoveries in Boeotian Thebes; Ivories
Deir El-Bahari, Parts I and VI (Egyptian Exploration and Linear-B from Thebes." ILN Nov. 28 and Dec. 5.
Fund, 28th and 29th Memoirs, London). POIDEBARD, A. 1939. Un grand port dispar1t: Tyr (2 v. Paris).
l\l~:WBERRY, P. ]893. Beni Hasan I (Egypt Exploration Fund POIDEBARD, A, and ]. LAUFFREY. 1951. S1'don: Amenag~ments
publications of The Archaeological Survey 1, London). a-ntiqlles du Port de Saida (Beirut).
-- 1907. The Timins Collection of Ancient Egyptian Scar- PORADA, E. 1957. "Mycenae 1939-1956, 1957, pt. II.2. The
abs alld Cylinder Seals (London). Faience Cylil1der." BSA 52.
NILSSON, MARTIN P. 1933. Homer alUi Mycenae (London). -- 1965. "Cylinder Seals from Thebes; A Preliminary Re-
-- 1949. "Oriental Import in Minoan and Mycenaean port" (Summary of AlA Paper). AlA 69.
Greece." Archiv Or-ientalni 17, 2. POIlADA, E., and B. BUCHANAN. 1948. Corpus of AnC1'ent Near
-- 1950. The Minoan-Mycenaean Religion and its Survival Eastem Seals -in North American Collect1'ollS I (Bollingen
-in Creek Religion (2nd ed., Lund). Series 14, New York).
NORTH, R. 1955. "Metallurgy in the Ancient Near East." PORTER, B., R. Moss, and E. BURNEY. 1960. Topographical
Orlentalia 24. Bibliography of Ancient Eg)lptian Hieroglyphic Te;rts, Re-
NOUGAYROL, JEAN. 1960. "I\:ouveaux textes accadiens de Ras licfs, aHd Paintings I, The Theban Necropolis, part I, Private
Shamra." eRAI 1960. Tombs (2nd ed., Oxford).
POST. G. 1933. Flora of S}wia, Palestine alld Sinai II (Beirut).
OHLY, D., and G. KLEIN~X 1951. "GCl11l11en der Sammlung PRITCHARD, JAMES B. 1954. The Ancient Near Eaj't 'il~Pictures
Arndt." Miinchner Jahrbuch der bildenden KUllst, N.F. 2. Relating to the Old Testament (Princeton).
OPPENHEIM, M. VON. 1962. Tell Halaf IV (Berlin). -- 1955. Ancient Near Ea,stern Texts Relating to tire Old
OSTEN, H. H. VON DER. 1934. Ancient Or·iental Sea,ls 1:n the T csfament (Princeton).
Collection of Mr. Edward T. Newell (OIP 22, Chicago). -- 1963. The Bronze Age Cemetery at Gibeon (University
-- 1957. Altorielltahsche S/:ege/steine del' Sammlllllg Han Museum Monograph, Philadelphia).
Silvius VOn Aulock (Uppsala). PRZEWORSKJ, S. 1939. Die Metaltindllstrie Anatoliens iJ~ der
-- 1961. "Altorientalische Siegelsteine." Bulletin of the Zeit von 1500-700 vor Clir. (Intern. Al"chiv. iiir Ethnogra-
i\!IedethMJS'IlI-useet, Stockholm, 1. phie XXXVI, Suppl., Leiden).
OrrEN, H. 1963. "Neue QueJien zum Ausklang des Hethit-
ischen Reiches." MDOG 94. QUIBELL,]. E. 1898. The R01l1eSseHJn (London).

PALMER, LEONARD R. 1961. .I.l1ycenaeans alld Minoans (Lon- RAO, S. R. 1963. "A 'Persian Gulf' Seal from Lothal." An-
don). tiqwity 37.
PARKER, BARBARA. 1949. Cylil1der Seals from Palestine." RAU, C. 1884. "Prehistoric Fishing in Europe and Korth
Iraq 11. Amer.ica." Smithsonian Contributiolls to Knowledge 25.
PARROT, A. 1953. Archeologie mesopotQ'IIlienne II (Paris). Rb:131KOFF, DIMlTRT, and PAUL CHERNEY. 1957. A Guide to
PENDLEBURY, J. D. S. 1930. Aegyptiaca: a catalogue of Egyp- Underwater Photography (New York).
tian objeetsin the Aegean area (Cambridge). REINECKE, PAUL. 1938. "Nelle friihbrot1zezeitliche Hortfunde
-- 1939. The Archaeology of Crete (London). aus SLidbayern." Ce1"Jllania 22.
 -----------------------.
12 CAPE CELIDONYA: A BRONZE ACE SHIPWRECK [TI{ANS. A;>.tER. PHIL. SOC.

REISNER, G, A. 1917. "The Barkal Temples in 1916." lEA 4. SHEAR, T. L£sLJE. 1940. 'The Campaign of 1939." Hesperia
RICHTER, G. M. 1\. 1915. Greek, Etruscan alld Roman Bronzes 9,3.
(The Metropolitan Museum of Art, New York}. SJOQVIST, ERIC. 1940. Problems of the Late Cypriot Browse
RIIS, P. J. 1939. "Rod-Tripods." ActaA 10. Age (Stockholm).
ROES, ANNE. 1937. "Une pierre gravee syro-hittite trouve a SKINNER, F. G. 1954. "Measures and Weights." A History of
Argos." EeR 61. Technology I (Oxford).
ROLFE, JOliN C. 1890. "Discoveries at Anthedon." AlA (old SMITH, S. 1945. "Middle Minoan I-II and Babylonian Chro-
series) 6. nology." AlA 49.
DE LA ROQUE, F. Brssox. 1950. Tresor de Tod (Catalogue -- 1950. The Slat lie of Ldriuii (Occasional publications of
General des antiquites egyptiennes du Musee du Caire, 102, the British Institute of Archaeology at Ankara, no. I,
Cairo) . London).
ROWE, A. 1936. A Cata10gne of Egyptian Scarabs, Scorcboids, STARR, R. F. 1937. Nusi, Report on the Excavations at Yorgan
Seals and Anndets in the Palestine Archaeological Museu.m Tepa near Kirlenk, Iraq . I (Cambridge, Mass.).
(Cairo). STEtNDORFF, GEORGE, and KEJTII C. St::ELE. 1957. When Egypt
-- 1946. "Discovery of the Famous Temple and Enclosure Ruled the East (2nd ed., Chicago).
o i Serapis at Alexandria." ASAE suppl. Cahier no. 2. STROM!lIENGER, EVA. 1961. "Rollsiegelfalschungen." Berliner
ROWTON, M. B. 1951. C<Tuppn and the Date of Hammurabi." Lahrbucl, fiir Vor- Wid Frilhgeschichte 1.
JNES 10. STUDDINGS, FRANK H. 1951. M'jlcellOea}t Pottery from the
RYAN, ERIC J., and G. F. BASS. 1962. "Underwater Surveying Levant (Cambridge). •
and Draughting-A Technique," Antiqu-ity 36. -- 1954. "Mycenae 1939-1953. Part VII. A Bronze Found-
er's Hoard." BSA 49.
S:\NDARS, N. K. 1955. "The Antiquity of the One-edged Bronze
Knife ill the Aegean." PPS 2l. TACK HOLM, V. 1956. Student's Flora of Egypt (Cairo).
SANTEI<IlE, H. GALLET DE, and J. TRElIlWX. 1947-1948. "Rap- TATLLIEZ, P., F. DUMAS, J.-Y. COUSTEAU, et at. 1957. The
port sur le depot egeeu et geometrique de l' Artemision it Complete Mcnnal of Free Divillg (New York).
DOlos." BeH 71/72. TAILUEZ, P. 1961. Nouvelles plongees saus ciibl e (Paris).
SAVE-SODERBERGH, TORGNY. 1957. Four Eighteenth Dynasty TALLQVTST, K. L. 1914. Assvruui Personal Names (Acta socie-
Tombs (Private Tombs at Thebes, vol. I, Oxford). tatis scientiarum f ennicac, Helsingfor-s, 43, no. 1, Helsing-
SCHACHERMEYR, FRiTZ. 1962. "For-schungsbericht iiber die fors ) .
Ausgrabungen und Neu lunde zur agaischen Fruhzeit 1957- TAYLOR, JOAN DU PLAT. 1940. "Mines Where the Mycenaeans
1960:' AA 1962, pt. 2. Got Their CopJ,Jer, Discovered in Cyprus." 1LN 24 (Feb.).
SCH.:;'.n:R, J. 1957. Stl/dien zic den griechiscllen Reliefspithoi -- 1952. "A Late Bronze Ag-e Settlement at Apliki, Cyprus."
des 8.-6. Lahrhwiderts v. Chr. ails Kretc, Rliodos, Tenos und Ant! 32.
Boiotiew (Kallmunz}. TAYLOR, JOAN DU PLAT, et at. 1957. ~~I'J'rto'll-P£gadhes. A Late
SCHAEFFER, C. F. A. 1936. Missions en Chvbre, 1932-1935 Bronec Age Sonctttarv 1'n Cyprus (Oxford).
(Paris.) TAYLOUR, \,VILLIAM. 1958. Mycellaean Pottery 1'n Italjl (Cam-
-- 1937. "Les Fouilles de Ras Shamra-Ugarit, huitieme bridge) .
campagne." Syria 18. THOMPSON, DOROTHY B. 1960. "The House of Simon the
-- 1948. Stmligrapliie comparee et chroltologie de l'Asie Shoemaker." Archaeology 13.
occidental (llr et II" 1II'iflenaires) (London). THROCKMORTON, PETER. 1960. "Thirty-three Centuries Under
-- 1952. Enkomi-Alasia (Publications de la Mission Ar- the Sea." NatGeo 117 (May).
cheologique Fram;aise et de la Mission du Gouvernement de -- 1962. "Oldest Known Shipwreck Yields Bronze Age
Chypre a Enkomi, Paris). Cargo." NatGeo 121, 5 (May).
-- 1939-1962. Ugaritica I, II, III, IV (Institut Fral1l:;'ais -- 1964. Lost Ships (Boston, Toronto).
d'Archeologie de Beyrouth, Tomes 31, 47, 64, 74, Paris). THROCKtllORTON, PETER, and JOHN M. BULLITT. 1963. "Under-
-- 1965. "'An Ingot God from Cyprus" (in Notes and water Surveys in Greece: 1962." E.'1;ped-ition5, 2.
News). Antiqnity 39. TOULOUPA, EvI. 1964. "Bericht tiber die l1el1en Ausgrabl1ngen
SCHLJEMANN, HEINRICH. 1880. llios, the City and Cmtntry of in ThebCll." Kad1ll0s 3.
the T1'ojaJls (London). TRENDALJ., A. D. 1948. I-Iandbook to the Nicholson Ail/sel/"Ill
-- 1884. Troja (Leipzig).
(2nd cd., Sydney).
SCHMIDT, HUBERT. 1902. IIe'inriclt Schlielllann's Sammlung
TSOUNTAS, CHR. 1888. 'AI'aIJ'Ka.¢al raq,wl' leI' 1\rLJK~I"l.L~, ArchEph
troja,nischer Altertiilller (Berlin).
pt. 3.
SCHUMACHER, G. 1908. Tell el-M1tteseUim I (Leipzig).
-- 1889. "Ept:lwaL il' TV AaKWI'tKV Kal 0 ra¢os TOU Baq;.o:wv, Arch-
Sco'IT, H. B. Y. 1959. "Weights and Measures of the Bible."
Epl<.
BiblArcl~ 22,2 (May).
TUCHELT, K. F. 1962. "Tiergefasse in Kopf- lind Protol11en~
SEAGER, R. B. 1912. Explorations in the Island of MachIas
gestalt." 1st. Forsch 22.
(Boston, New York).
TUFNELL, OLGA, et 01. 1940, 1953, 1958. Lachish (Tell ed-
SEGRE, ANGELO. 1944. "Babylonian, Assyrian, and Persian
Measures." JAOS 64. Dnweir) II, III, IV (London).
SEIRADAKI, MERCY. 1960. "Pottery from Karphi." BSA 55. TYLECOTE, R. F. 1962. }It!etalhwgy in Archaeology (London).
SELLERS, O. R. 1962. "Weights and Measures." IDE IV
(New York, Nashville). U.S, Nav.11 Diving Manual. 1959 (NAVSHIPS 250-538, Navy
SELLIN, E. 1927. "Die Ausgrabung von Sichem." ZDPV 50. Department, Washington).
SELTMAN, CHARLES. 1924. Athens, Its H1'st01'y and Coinage
Before the Persia,n Invasion (Camhridge). VALMIN, NATAN. 1937. Poids priltistm-iques grecs de Malthi
-- 1955. Greek Co'ills (2nd ed., London). en Messenie (Lund).
SEYMOUR, TnOMAS D. 1907. Life in the Homeric Age (New VANDERPOOL, E. V. 1957, 1963. "News Letter From Greece."
Ym·k). AlA 61 and 67.
SEYRIG, H. 1956. "Cylindre representant une tauromachie." VENTRIS, MICHAEL, and JOHN CHADWICK. 1959. Documents
S}-'ria 33. 1'n M'},'cenaealb G1'eek (Cambridge).
u-------------------------
VOL. 57, PT. 8, 1967J BIBLIOGRAPHY 13
VERCOUTTEIt, J, 1956. L'E'.gypte et te nunuie egeen pdhcl!enique 'VA},lPUR, ]. C. 1947. Tell en-Ncsbeti II. The Pottery
(Institut francais d'Archeologie Orientale Tome XXII, (Berkeley).
Paris) . \VARD, \VILLIA1It H. 1909. Cyfinders and other Ancient Orien-
VJ<:RMEULE, E. T. 1961. "Review of Kenna, Cretan Seals." tal Seals iw the Librar-y of J. Pierpont Morgan (New York).
TIIC Art Bulletin 43. -- 1910. The Seal Cylinders of vVestem Asia (Washing-
__ 1964. Greece in the Bronze Age (Chicago, London). ton) .
VIRE\', PH1LlPPE. 1891. Sept Tombecn» ihebcincs de la X'Vl l l" \iVATERER, JOlI7\T V.,T. 1956. "Leather." A Histor.\' of Tech-
d'\'Jlastie (Mem. Miss. 2, Paris). nolog'y II (Oxford).
\'oN BECKERATH, J. 1962. "Queen Twosre as Guardian of VVATEHMAK, STANTON A. 1960. "Three Thousand Years Under
Siptab." J EA 48. the Sea." Explorers Lournai 38, 3 (Oct.).
\'ON BrSSING, FR. VII. FREI HERR. 1928-1929. "Problerne der VVATZINGER, C. 1929. Tell el-Mvteseltini II (Leipzig).
agyptischen Vcrgeschichte." AOF S. VVEBER, O. 1920. Altorientalische Siegelbilder I 1/nd II; Der
Alte Orient 17/18 (Leipzig).
\VACE, A. J. B. 1932. Chamber Tombs at Mycenae. (=Ar- VIENIG, S. 1962-1963. "Ein Siegelzylinder mit dern Namen
ctmeotoaia 82). Pepi's 1." Zeitsclwift fiir iigypt. Sprache 1md Attertunis-
-- 1939. "Mycenae, 1939." IRS 59. lewnde 88.
-- 1949. M:ycenae: An Archaeological History and Guide 'VISEMAN, D.]. 1953. The Alalakh Tablets (London).
(Princeton) . -- 1958. Giitter lind Menscben im Rotlsiegel Westosicns
-- 1953. "Mycenae: Preliminary Report on the Excavations (Prague).
of 1952." BSA 48. WOOLLEY, L. 1935. Alalakh. An Account of the Excavations
-- 1955. "Mycenae: Preliminary Report on the Excavations
at Tell Atchana in Hctcy, 1937-1949 (London).
of 1954." BSA 50. \VOOLNER, DIANA. 1957. "Graffiti of Ships at Tarxien, Malta."
Antiquity 30.
'VACE, A. ]. B., and C. \V. BLEGEN. 1939-1940. "Pottery as
Evidence lor Trade and Colonisation." Klio 32.
\VRESZINSKI, V·l. 1923. Atlas 2m Altaegyptischen Kultur-
geschichte I, II (Leipzig).
\VACE, A. J, B., and F. H. STUBBlNGS. 1962. A Companion to
'WRIGHT, G. ERNEST. 1943. "I Samuel 13 :19-21." Bibl Arcb 6.
Homer (London). -- 1957. Biblical A1'chaeology (Philadelphia, London).
\VAINWRJGHT, G. A. 1913 "Alashia-Alasia and Asy." Ktio 14.
-- 1934. "The Occurrence of Tin and Copper Near Byblos." XANTHOUDIDES, S. 1906. "'EK Kpr(T17"" AI'chEph.
lEA 20. -- 1924. The Vaulted Tombs of Messara (London).
1943. "Egyptian Bronze-Making." Antiqnity 17. XENAKI-SAKKELLARIOU, AGNES. 1958. "Les Cachets rninoens
1944. "Egyptian Bronze-Making Again." Alltiqu.ity 18. de la collection Giamalakis." Etudes critoises 10, Paris.
1944. "Early Tin in the Aegean." Ant'iqltity 18.
1952. "Asiatic Keftiu." AlA 56. YADIN, Y. 1963. The Art of Warfare ,in Biblical Lands (New
\VALDSTEIN, C. 1905. Tile Argl:ve Heraeuni II (Boston, New York, Toronto, London).
York) . YAD1N, Y., et al. 1958, 1960. Hazar I, II (Jerusalem).
VilALTEI<S, H. B. 1926. Ctualoone of tlie Engraved Gems and
Cameos, Greeh, Etru.sCQHand Homan ia the British Museum ZEUNER, F. E. 1954. "Cultivation of Plants." A History of
(London) . Technology I (Oxford).
 II. THE EXCAVATION

GEORGE F. BASS AND PETER THROCJOiORTON

The excavation began on June 14, 1960, after we bad

pitched our camp on a narrow beach opposite Su Ada.'
Although only a strip of sand and pebbles, surrounded
on all sides by unscalable cliffs, the camp site offered a
source of fresh water and a suitable anchorage. It lay
nearly twenty miles from the nearest town, Finike, and
was a thirty to forty minute sail from the wreck in
either of our two Bodrum boats: the Mandalinri, cap-
tained by Kemal Aras, and the Luifi Gelil (fig. 4),
captained by Nazif Ccyme».
Most of tile staff slept on mattresses spread out on
• the sand under a large tent, but shelters were scattered
along the beach for the women members. Living condi-
tions were not pleasant. Rocks fell from the cliff during
the night, and on one occasion draftsman I-:Ionor Frost
found a boulder lying on her pillow just before she FIG. 4. The Lutfi Gdit under sail.
retired for the night. Flies forced 1\1r. Ball to draw
most of the finds while crouched on the ground under During the working season, we sailed each mornmg
a tiny mosquito net. The heat was intense, often in one or both of our boats to the wreck site and moored
reaching ]] 0° F. in the shade by ten o'clock in the to a buoy anchored just over the wreck (Jig. 5).~
morning. We could spare one of our boats to bring
food from Finike only once a week; with no means of I. PROBLEMS OF UNDERWATER EXCAVATJON
refrigeration, meat lasted only a few days even after
Underwater archaeology, in general, demands not so
being cooked and heavily salted. The heat also caused
much new methods of excavation as adaptation of
serious problems in the development of film.
present land techniques to the underwater environment.
In spite of these difficulties, we were able to work We began work at Gelidonya convinced that there wer-e
with relative efficiency. Two small generators, which no inherent technical limitations to prevent us from
furnished power for the darkroom enlarger, also sup- applying the same standards that are considered desir-
plied the entire camp with electricity for lights. The able on land. Almost always in the past, underwater
darkroom itself was a small cave closed with canvas, excavations had been no more than salvage operations
and could be used only at night. Other "work r00111S," performed by professional divers. We felt that we must
for cleaning finds, filling air tanks, and cooking, were
treat this wreck as an archaeological site, and that its
simply small areas of the beach roofed with old
parachutes,
The cliffs protected us from the north winds, but by
September we had received the first warnings of the
seasonal lodos blowing from the south. The first time
that waves washed across the beach, we were unpre-
pared, and material was buried under sand before it
could be moved. From that time, everything in the
camp which was not being used was stored on a small
hill, which had been formed on the beach by a rock
slide some years before. Finally, three months after
we had first arrived, the sky was darkened by threaten-
ing clouds. Fearing that the fir-st rain of the season
might cause another landslide, we broke camp during
the night and sailed away in the early hours of the
morning. Vve reached Finike just as the heavens
opened and it poured. FIG. 5. The Lutfi Gclil anchored over the wreck.
---
1Su Ada (Water Island), long noted for its supply of fresh Fuller accounts of living conditions and camp life appear
2
water, is identified by Beaufort (Kammuinia, 39) as "the island in: P. Throckmorton, NatGeo 121, 5 (May, 1962) 697 fl.;
of Grambousa (the Cnnnbnse of Strabo, and Diowysia of Lost Ships (Boston and Toronto, 1964); Atkintic Monthly
Scylax and Pliny)." 213,6 (June, 1964). G. Bass, Elopedilion 3. 2 (1961) 2 ff.
21
22 CAPE CELlDONYA: A BRONZE ACE SH1PWRECK [TRANS. AMER. PHJL. SOC.

location fifteen fathoms below the surface of the Aegean vague pains in the bones or itching to paralysis, blind-
should not be an excuse for improper excavation. ness, or unconsciousness.
Many unique problems, however, confront the under- The only preventative for the bends is a controlled
water excavator. The time which he may spend daily ascent following each dive. Schedules have been pre-
on each site is often severely limited; the distance from pared which allow the diver to rise to the surface in
which he can photograph and even see is dependent on stages. These schedules, or decompression tables, are
the clarity of the water; and the water itself causes based on calculations of the amounts of nitrogen ab-
optical distortion, which makes everything seem one- sorbed or lost by different types of tissues at various
third oversize and which affects the diver's judgment depths and during various lengths of time." Three
of distance. Although diving with the Cousteau-Gagnan decompression tables, similar but differing in detail, are
aqualung system used at Gelidouya is simple, safe, and in general use: the British, American, and French.
easy, the diver is an intruder in an alien and potentially The French Navy repetitive diving tables were used
hostile world, where his life may be endangered by a at Cape Celidonya."
Humber of factors. Divers were lost at Antikythera, The treatment for the bends is recompression fol-
Artemision, and Grand Cor.gloue. The main problems lowed by slow decompression, and this must be carried
are: S out immediately to be effective. The diver is put under-
pressure in a chamber until the bubbles which have
1. PRESSURE formed in the circulatory system are reduced in size
and the nitrogen can be eliminated normally. During
The Greek sponge divers who died at Artemision and treatment the pressure is reduced very gradually, fol-
Antikythera, and an American diver who was later lowing predetermined schedules which also appear as
paralyzed at Yassi Ada/ were victims of "the bends." tables, to prevent a reoccurrence of the bends.
An understanding of the bends depends upon the knowl- The wreck lay at depths of 26 to 28 meters. Because
edge that air is compressible while water is not, and of our isolated position, far from the large chamber in
that the pressure of water increases with its depth.
Istanbul, and because our portable chamber was held by
Regardless of the apparatus he uses, a diver must re- customs, we added a safety factor and calculated our
ceive air at a pressure equivalent to the pressure sur- dives as being at the next deeper depths on the tables.
rounding him in order to breathe. Without air at vVe decided that 11 or 16 minutes of decompression at
ambient pressure, a diver's lungs, sinuses, and middle- 10 feet would be too long and tiring for the divers, and,
ear cavities would be squeezed and crushed as the air
therefore, limited each person's first dive of the day
in them was reduced in size. to 40 minutes, with a 6-minute decompression stop at
Tailliez and his collaborators in the Undersea Re-
3 meters following the working time on the bottom.
search and Development Group of the French Navy
Six hours after the first dive, we normally made a
define the phenomenon of the bends as follows:
second dive of 28 minutes on the bottom, followed by
During his descent and stay below, the diver breathes another 6-minute decompression stop. When a 6-hour
air at a pressure greater than the atmospheric, and a cer- interval between dives was not possible, decompression
tain amount of nitrogen is absorbed by the blood each time
times were necessarily lengthened.
it passes through his lungs.
The tissues bathed by the blood in turn take up more A timekeeper on the surface kept a careful log of
and more nitrogen, depending on the length of the dive. diving times and signalled the divers to ascend by
This phenomenon is reversed during the ascent. The hammering on a length of pipe hung from the side of
excess nitrogen dissolved in the various tissues is carried
back to the lungs by the blood, where it is eliminated during the boat; one diver in each group usually wore an
respiration. If the ascent is too rapid, the difference be- underwater watch as an added safety measure. After
tween the pressure of the nitrogen dissolved in the various weeks and months of diving, divers tended to become
tissues and the hydrostatic pressure may become so great careless, although discipline was usually well kept. The
that bubbles begin to form in the blood, as in a bottle of
champagne that has just been opened. These bubbles pro- overzealous diver who overstayed his time on the bot-
duce compressed air iJlness, also known as caisson disease, tom would then find himself being kept on the decom-
"the bends," or decompression sickness .... 5 pression stop for up to an hour or more, sometimes into
The symptoms of this disease are various and un- the cold twilight; a glance at the tables shows that the
pleasant, although not always fatal. They range from G Ibid., 132, "Complete safety can never be had except by
a Formore complete coverage of diving and its problems
a prolonging the stops inordinately." Even divers carefully fol-
see: U.S. Navy Diving ManHal, NAVSHIPS 250~538 (Navy lowing the tables risk suffering from the bends; B. Empleton,
Department, Washington 25, D.C., 1959) ; P. Taillcz, F. Dumas, et al., op, cit. (supra, n. 3) 70.
J.-Y. Cousteau, et ol., The Complete Manual of Free Diving r We now use the new U.S. Navy repetitive tables, which
(Kew York, 1957); B. Empleton, E. Lanphier, J. Young, L. give a higher safety factor than the new French Navy tables,
Goff, et a/., Tile l.lew Science of SkilL and Scuba Diving (New which have, in turn, replaced the French tables used at Cape
York,1962). Gelidonya. At the suggestion of the U.S. Navy doctors, for
'G. Bass, NatCeo 124, 1 (July, 1963) 156. added safety, we now use the tables for the next deeper depth
5 P. Taillez, et ot., op, cit. (supra, n. 3) 97-98. and the next longer time than we are actually diving.
VOL. 57, PT. 8, [9671 BASS AND THROCKMORTON: THE EXCAVATION 23
required decompression time rises sharply after 35 been almost invariably against people swimming on the
minutes at our working depth. surface, where the splashing seems to attract sharks.
Divers complained of minor symptoms of the bends \Ale discussed the problem with Frederic Dumas, who
after weeks of daily repetitive dives, but these, whether has had a great deal of first-hand experience with
real or imagined, proved to be of no consequence except sharks under water, and it was agreed that if we were
where peace of mind was concerned. After a day's bothered by aggressive or overly curious Sharks, we
rest, divers invariably felt better, and fatigue after dives would suspend operations until they had gone away.
was reduced by slightly extending decompression times. It was never necessary to do this, although at least one
Nitrogen narcosis, also caused by breathing air under diver reported seeing a shark.
pressure, was not a serious problem at the depths in Bites from moray eels were avoided by taking care
which we worked. The mental faculties of some of the before putting hands into the cracks or holes which
divers seemed to have been dulled more than those of might house the creatures.
others, but advance planning of di ves left few decisions Our most common complaints stemmed fr0111 our
to be made by a diver once he was working on the being so often in the salt water. Infected ears kept
sea bed. some of the divers out of the water for several days
Of the other illnesses caused by pressure or pressure at a time, and even the tiniest cuts and bruises became
changes, air embolisms are certainly the most important. more severe until the sufferer could let them dry thor-
An embolism is most commonly caused by panic, which oughly over a period of days. We found that a pro-
might lead a diver to ascend rapidly without exhaling; tective coating of collodion allowed wounds to heal
trapped air can rupture the lungs as it expands, forcing more quickly.
bubbles into the blood. Gooel training anel physical fit-
ness are the best preventatives. 3. VISIBILITY

2. OTHER HAZARDS The sea is filled with particles of dirt and organic
A diver, dependent on an artificial supply of air, con- matter which can severely limit visibility. The Aegean,
stantly faces the dangers inherent in being in an un- however, has very little plankton, and thus provides an
familiar environment. Drowning is probably the major unusually high range of vision. At Gelidonya, on the
cause of death among skin divers. Engine exhaust, clearest clays, divers working 011 the wreck could be
entering an air hose or a tank being filled, can prove seen from the surface by swimmers equipped with
fatal. Strong currents can sweep divers away, or kink masks. Effective range for photography, however, is
their air hoses. Any dizziness or illness which on land seldom more than thirty feet; matter suspended in the
would be of no consequence can become most serious water, but not noticeable to divers, affects the resolving
under water; a diver who is forced to vomit is in power of the camera lenses. For good, clear photo-
trouble. graphs, the maximum range at Gelidonya was nearer
At Gelidonya we avoided serious injuries by follow- fifteen feet, even on clear days."
ing the general rules of diving found in most manuals The index of refraction of water causes objects to
on the subject. All of the divers had been trained In appear larger and nearer than they really are, both to
advance in the correct procedures for clearing water the diver and to the lens. The focal length of a lens
from their air hoses, in replacing masks under water under water is increased by about one-third; normal
where they can be knocked off, and in sharing their lenses become telephoto, and wide-angle lenses become
mouthpieces with other divers. \Ale generally followed normal. These problems are partially solved by the
the rule that aqualung divers must never work alone; use of a wide-angle lens, which covers a large area
011 several occasions possibly fatal accidents were from a short distance away. Such lenses, however,
avoided by the instant reaction of a diver to his partner's cause distortion when taking photographs from the
kinkeel air hose or empty tank reserve. Air intakes distance at which they are most useful under water.
were kept upwind of running compressors and other Water acts as a blue filter, which progressively ab-
engines to prevent carbon-monoxide poisoning. Be- sorbs the different colors of the spectrum as the depth
cause of the almost constant current, divers descended increases. Red goes first, being seriously affected for
and ascended on a line running between the boat and purposes of photography after one meter and disappear-
the wreck site; a dinghy with an outboard motor was ing completely after five meters. Orange is affected
on hand to rescue any diver who might be swept away, after two to three meters and disappears after ten.
Sharks do not normally present a danger to divers Yellow is affected after five meters, but is still dis-
in the Aegean." Attacks reported in recent years had
9 For an introduction to the problems of underwater photog-
8 Hans Haas, in Men and Sharks (New York, 1954), de- raphy see: Encyclopedie Prisinc dn M onde Sous-Mcrine (Paris,
scribes the difficulties his expedition experienced in trying to 1957), 392 ff.; Dimitri Rcbikoff and Paul Cherney, A Guide to
attract sharks in the Aegean in 1941. For further studies of Underwater Photography (New York, 1957); J. Greenberg,
shark behavior see: Perry Gilbert, ed., Sharks a.nd Survival Underuater Photogmphy Simplified (2nd ed., Coral Gables,
(New York, 1964). Florida, 1963).
 -
24 CAPE GELlDONYA: A BRONZE AGE SH1PWRECK [TRA"<S. IUtER. l'Hll •. soc.

tinguishable down to 'thirty or forty meters. On the ballooning, a broken faceplate, an entangled hose, or
wreck everything appeared greenish blue. Blood from compressor failure combined with a faulty non-return
minor cuts, for example, was green. Artificial lighting, valve on the diver's hose are all possible causes of death
by flash, allowed the camera to record colors which the which do not face the aqualung diver.
diver did not normally see. A flash, attached to its The advantages of this type of dress, however, are
underwater camera case by a short arm, as was ours, many for divers involved in other types of work. The
can harm a picture if mud has been stirred up by diver is heavy and can perform difficult manual tasks.
excavation in progress; light is reflected almost di- He can remain warm and dry over long periods of time.
rectly back into the lens from scattered particles be- I-Ie is able to work in mud, and can intentionally bal-
tween the photographer and his subject. At Gelidonya loon himself out if he becomes stuck. At Gelidonya,
there was always sufficient light during the daytime for helmeted sponge divers were used only for such heavy
black and white photography. work as swinging a sledge hammer, for they could not
work on the actual wreck site without risking damage to
II. METHODS OF EXCAVATION delicate material with their heavy shoes, and they could
110t work at all when the current was running.
1. DIVING The invention of the aqualung revolutionized diving,
The choices of diving equipment were relatively 51111- but its principle is quite simple. Air is supplied to the
p1e, for the same limitations of time and depth hold diver from metal hottles worn on the back and filled
true for all compressed-air equipment. Use of gas with compressed air. The air passes through a regulat-
mixtures, such as helium and oxygen, which can par- ing device which delivers it to the diver at the same
tially overcome some of these limitations, would have pressure as that of the sea surrounding his lungs.
proved too complex and costly for our operation. Exhalations are exhausted immediately, so that it is
Oxygen rebreathing equipment is generally unsatisfac- impossible for carbon dioxide to build up in the air
tory in the Aegean because it is limited to a working supply. The diver's body is surrounded by water, but
depth of 30 feet; well-preserved wrecks generally lie as the body has approximately the same specific gravity
below this depth, where they have been protected against as sea water, the pressure has no affect on it; the only
wa ve action. air required is the amount needed for breathing.
Although various self-contained underwater breathing When using the aqualung, the diver wears rubber
devices were developed before Cousteau and Gagnan fins on his feet for ease of movement, a plate-glass mask
invented the aqualung in 1943, the helmeted diving dress for clear visibility, and weights designed to give him
was standard among divers 1111til that time.!" 'The neutral buoyancy. A tight-fitting foam-rubber suit al-
helmet diver wears a rubberized can vas suit with a lows water to enter and lie next to the skin, but does
metal helmet and breastplate. His hands, the only parts not allow it to circulate; this water is soon warmed by
of his body exposed to the water, protrude through body heat and protects the diver against cold. Even in
rubber cuffs. Compressed air, supplied through a heavy the relatively warm Aegean, the colder water beneath
hose from a boat above, fills the helmet and expands the surface would quickly draw off much of the body
the suit to just above the diver's waistline, so that his heat of a naked diver, especially if he were drawing or
lungs are surrounded by compressed air. His feet are performing some other quiet task for forty minutes.P
weighted by heavy lead shoes, and the buoyancy of the A variation on the aqualung, known as the "hookah"
air which fills the upper part of his suit is compensated or "narghile," differs from the aqualung only in that the
for by lead weights strapped around his waist or slung air is supplied from the surface through a hose. The
at breast and upper small of his back. He must main- hose joins the same regulator that is otherwise mounted
tain a vertical position at all times, for if he leans over on air tanks (fig. 23).
air is likely to rush to his feet. As the exhaust valve At Gelidonya, all three systems of diving were used,
is in the helmet, this can result in his being pulled to although the helmeted sponge divers were employed, as
the surface out of control by the overly buoyant suit. has been mentioned, only for the very heaviest work.
It requires considerable skill to maintain the balance The aqualung offered great mobility and freedom to its
between the air in the suit and the lead weights, and wearers, but it had one major disadvantage. The bot-
the diver must continually manipulate the exhaust valve tles must be filled with high-pressure air at between
in his helmet. 1,800 and 3,000 pounds per square inch (i.e., up to 200
There are other disadvantages to the helmet suit. atmospheres). Compressors capable of producing air
Movement is difficult and visibility is poor. Accidental at such high pressures are delicate devices. Ours ran
up to ten hours a day; compressors normally built for
10 Robert H. Davies, Deep Diving alld Submarine 0 beraticns
(London, 1955) is the classic handbook on underwater work, such strenuous service are so bulky and heavy that they
especially in regard to the era before the aqualung. For an
11 The high heat capacity and heat conductivity of water
excellent history of diving see: James Dugan, Man Under the
Sea (New York, 1956), published in England as Man Evplores make suits desirable in temperatures below about 70e F., B.
the Sea (Harmondsworth, 1960). Ernpleton, et ai., op, cit. (s1Ipra, n. 3) 25-26.
VOL. 57, ('T. 8, 1967J BASS AND THROCKMORTON: THE EXCAVATlON 2S
could not have been used at Geliclonya. OUf portable With proper hose, there are other advantages to using
four-stage Bauer compressor served us well, but no "narghiles" on an excavation. The diver has constant
compressor of its size, overloaded as it was, could have contact with a tender on the surface who can feel pre-
been expected to operate continually under the condi- determined signals pulled on the hose. There is not
tions at Gelidonya. The small compressors which we the danger of running out of air, especially during
had brought as reserves proved to be utterly unreliable. heavy exertion, which can shorten the working time of
As an emergency measure, while repairs were being a diver breathing from bottles. The major disadvantage
made all the high-pressure compressor, we rigged a is that the diver must be particularly careful not to
"narghile" system with air-lift hose running from entangle his hose with another hose or with measuring
N!a'Jldatinfi)s diving compressor. This system worked strings and other obstacles. In a strong current the
quite well except for the soft hose which, not having pull on the diver's hose also makes it difficult for him
been designed for diving, tended to kink in the strong to move about.
current, cutting off the diver's air supply. Normally
the kinks could be shaken out of the hose. or removed 2. PLOTTING THE \-VRECK
by the diver's swimming upward for a few feet to
relreve the tension on the hose. When these measures After a preliminary inspection of the site, we began
failed, the diver ascended sharing his partner's mouth- mapping it by making a photographic montage. Over-
piece and air supply. At a last resort. the diver could lapping pictures were taken by a diver swimming in an
release his weight belt and, exhaling to prevent an approximately straight line above the wreck. A bubble
embolism, make a free ascent to the surface. level and a plumb line attached to his camera assured
Man,dalil1{i's compressor, a low-pressure standard that the pictures were taken vertically and from a fixed
height above the sea bottom; strings stretched across
Siebe Gorman diving compressor driven by pulleys and
belts from the sponge-boat's one-cylinder diesel engine, the wreck guided the photographer's path. A two-
meter surveying pole, marked in twenty-centimeter seg-
proved very reliable and gave trouble only once, when
ments, was included in each frame to allow the negatives
a valve blew out. A new valve was made from a piece
to be enlarged at a fixed scale.
of an old shoe and the compressor was running again Distortion caused by the index of refraction of water
the next day. This kind of repair work would have was a serious problem only when the overlap between
been impossible with a precision-made high-pressure two photographs was small. Otherwise, the edges of
compressor. the picture were cropped, leaving the less distorted

FIG. 6. Photographic montage of entire wreck as when first seen; north at bottom.
26 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

FIG. 7. Photographic montage showing same area as in fig. 6, after excavation.

central portions to be used in the montage. A greater The importance of photography in an underwater
problem was the difference in scale of objects at dif- excavation cannot be overemphasized, although this
ferent levels on the sea bed, which usually prevented necessitates a source of electricity for enlarging even in
truly accurate joining of pictures. This problem could the most remote areas. The archaeologist on land can
have been diminished by taking the photographs from inspect his site whenever he wishes, but the time limita-
a greater height, but this would have brought about a tions imposed by the decompression problem make this
loss of clarity. impossible in an underwater excavation. Further, the
Two montages of the entire site were made: one at land archaeologist can watch the excavation in progress
the beginning of the season, when the wreck had not and can discuss the work on the spot with members of
yet been cleaned or seriously disturbed (fig. 6), and a his staff. Because we seldom dived in groups of more
second near the end of the season after much of the than two or three divers, it was possible for several days
cargo had been raised (fig. 7; in this, the sponge seen at to pass before anyone diver could be on the site with
the right side [west] of figure 6 is just below the diver's the archaeological director, the draftsman, or the photog-
right leg, and the pierced stone "anchor" at the left end rapher to show directly a point of interest. For this
[east] of figure 6 has been replaced by a stone with reason, it was necessary each morning to di scuss and
a circle marking the position of the original hole). plan that day's operations with the aid of photographs
Neither was intended as an accurate plan from which taken on the previous day. '\'e took an average of fifty
measurements could be taken, but both served as reliable underwater photographs each day, and nearly all have
checks on drawings. proved of value when consulted during later study of
Close-up photographs of objects in situ, and a photo- the wreck.
graphic record of our methods of excavation were taken Confused reports, caused by haste or by slight nitro-
with both color and black and white film. Kodak gen narcosis, can often be clarified by photographs.
Ektachrome (High Speed) and Kodachrome were used Tcleally, these photographs would be shown after each
for color, with the best results from Kodachrome used dive to the next team of excavators. A Polaroid
with clear No.5 flash bulbs. For black and white pic- camera, mounted in a special l-ousing." showed promise
tures, Plus X or Veri chrome Pan film was normally before it was lost in the strong current during the first
exposed for 1/125 second at f/S, but the exposures days of the excavation.
varied according to the time of day; these were de- While the photographic plan was being made, we
veloped daily using D 76 stock solution, mixed with an snrveyed the wreck with methods similar to those used
equal amount of water, for six minutes. Contact prints
12 The underwater case for the Polaroid camera was de-
were made as soon as possible, and pictures were signed and constructed by the French Navy's Undersea Research
selected for enlargement immediately. Group (O.F.R.S.).
VOL. 57, PT. 8, 1967] BASS AND THROCKMORTON: THE EXCAVATfON 27
all land sites. Iron spikes were driven into the rock
around the wreck to serve as permanent bench marks,
and objects in the wreck were triangulated from them
(fig. 8). This was not completely satisfactory because
the horizontal measurements would have been accurate
only if the strings and meter tapes used in the triangula-
tions had been horizontal. As there is no horizon under
water, and as the weightlessness of the diver's body
allows him to stand on a steep slope without realizing
that he is not standing upright, visual estimates of the
horizontal were meaningless. The only solution to this
problem was to keep the pole as parallel as possible to
some natural line in the surrounding rocks which ap-
peared to be vertical, and then to move the string at-
tached to the pole up and down until it seemed to be at
a right angle. No more than thirty-five measurements
could be made in one dive, but by repeated checks, we
finally arrived at measurements which seemed to be
accurate to within a very Iew centimeters.
Detailed scale drawings of small areas or groups of
objects were recorded on sheets of frosted plastic at-
tached to clipboards. Ordinary pencils were used, but
FIG. 9. Duthuit drawing on frosted plastic. Photograph taken
from Platform towards Boulder.

the wooden halves tended to separate under water unless

taped in advance (fig. 9).
Using preliminary measurements and the first photo-
graphic montage, we were able to make a working plan
of the site while most of the cargo of the ship was still
hidden beneath sea growth. The natural contours of
the sea bed broke the wreck into easily defined areas,
which were labeled according to the divers' 1110stfamiliar
terms for them (fig. 43, key). Thus, Area G was the
"gulley" between a large fallen boulder (B) and the
base of the cliff which ran up to form the rocky edge
of the island. Just west of G and B was a flat "sandy"
area (S), which separated the gulley from a rocky
"platform" (P), about a meter higher than S. When
several objects were discovered still farther west, this
new area became E, for "extension," and objects scat-
tered by the current over a large area to the north of all
previously named areas were simply from TvI, or "mis-
cellaneous." There was no need for a grid system,
which we had considered using.
When first discovered, the only parts of the wreck
that were visible were the copper ingots of the platform
and a few scattered pieces which lay on the sand (fig. 6).
Everything else was covered with a heavy layer of
coralline sea growth. Captain Kemal's original estimate
of "two t011S of bronze" was based on his assumption
that the platform was a solid mass of metal. Tt was
only after some days that we found this was 110t the
case; the northern edge of the platform was merely a
stone ledge. At the same time, however, we discovered
FrG. 8. Triangulating on the sea bed. that the gulley was full of metal.
28 CAPE GELJDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOc.

After removing the material which was not solidly

cemented together by the sea growth, we were faced
with the large masses of metal which had been com-
pletely encompassed by the living rock. VVe found,
when cutting with hammer and chisel around pieces of
metal, that corrosion products had prevented the con-
cretion touching the metal from becoming rock hard.
A chisel placed at the juncture of the rock and the metal
so that it acted as a wedge would, when given a sharp
blow with a hammer, shear even quite fragile metal
objects from the bedrock to which they were attached.
A seemingly safer method of removing these objects, by
cutting into the concretion around them without touch-
ing the actual artifacts, proved to be less practical; we
found that fragile metal was sometimes shattered by the
continual vibration of heavy blows nearby. -
Cutting ingots free of their cases of limestone pre-
sented a different problem. If we had tried to chisel
tree the individual ingots, we would have faced several
summer-s' work, in the course of which much of the
material would probably have been damaged. Dumas
suggested that an automobile jack placed under a mass
FJG. 10. Dumas places stone "anchor" into lifting basket. of ingots would lift it from the sea bed, shearing not
across the metal, but on the line where the metal joined
the rock.
3. EXCAVATING THE WRECK
We placed a hydraulic jack with a lifting capacity of
Loose and scattered pieces, after being tagged with three tons into a slot chiseled beneath the southern edge
identification labels, and then drawn and photographed of the ingots piled on the platform. Planks and a steel
in situ, were placed in small wicker baskets. These bar over the jack distributed the thrust. The jack was
pumped carefully and, when strained almost to capacity,
baskets, along with individual ingots, were raised to the
broke off the entire mass exactly at the juncture of the
surface in a steel wire lifting basket by a winch on the
metal and rock.
Luffi. Gelil (fig. 10).

,•
~.~

FIG. 11. Duthuit guides sheet-wrapped lump being lifted FIG. 12. Dumas uses crowbar to pry lump loose from
by winch from the Lntfi Gelil. area P; jack handle protrudes in foreground.
VOL. 57, PT. 8, 1967J BASS AND THROCKMORTON THE EXCAVATION 29

FIG. 15. Cleaning concretion away from cargo

Ftc. 13. Bass, Duthuit, Iliing, and Throckmorton fit lumps in lumps from area G.
of concretion together at camp.
thick, was knocked away (figs. 15-17). Sandbags were
The freed lump was winched to the deck of the L1ttfi used to prop up individual artifacts as they became free,
Gelil after a diver had wrapped a bed sheet around it to so that eventually the objects lay in the positions they
protect the matting and rope he had seen stuck to its had held when the ship finally settled and disintegrated
bottom (fig. 11). on the bottom of the sea.
After this, raising of "lumps" of concreted cargo, The draftsmen next drew the objects, which still held
weighing several hundred pounds each, became standard their relative positions to one another, and added these
practice. Masses of encrusted metal were studied for details to their plans, which had previously shown only
what seemed to be the natural lines along which they the masses of limestone with occasional corners of pro-
would break, and cuts were made along these lines truding metal.
before jacking (fig. 12). If a Jump failed to break In the gulley were found the only substantial remains
loose, we tried another angle of thrust. No material of wood, but these protruded from beneath a mass of
was damaged by this seemingly violent method. ingots. ,\1e feared to raise this mass by means of the
After cleaning the first mass of cargo almost im- cable and winch on the Llttft. Gelil lest she roll slightly
mediately after it arrived on board the LlItfi Celil, we during the operation; if this had occurred while the
realized that evidence of lading was being lost. From ingots were only a few centimeters above the sea bed,
that time, uncleaned lumps were carried back to the the wood might have been violently crushed. The mass
camp and fitted together (fig. 13). After an entire area was, therefore, tied to a large plastic balloon which ....
vas
was reassembled on land as it had rested on the sea filled on the bottom with air from a diver's mouthpiece
bed (fig. 14), the heavy limestone, tip to 15 centimeters

FIG. 14. Lumps and ingots from area G after having FIG. 16. Lumps from area G when completely
been fitted together on land. cleaned of concretion.
30 CAPE GELlDOI\'YA: A BRONZE AGE SHIPWRECK [TRANS. A~rER. i-au.. soc.

F1G. 17. Lumps from area P when almost completely

cleaned; white material on right is tin oxide.

(fig. IS). After filling the balloon with air until its
upward thrust reduced the weight of the conglomerate FIG. 19. I11ing attempts to chisel beneath lump
to ten or fifteen pounds, two divers were able to carry containing wood fragments.
the mass carefully away from the area of danger. There
a second balloon was attached and filled, and the cargo overlaid with the usual coralline sea growth (fig. 19).
was borne easily to the surface where it was picked up As with the metal cargo of the ancient vessel, we decided
and winched onto the boat. Each balloon was capable that it would be sensible to raise the wood in one unit
of lifting between three and four hundred pounds, so and to excavate and draw it on land, where both
that 110 lump was too heavy to be raised in this way archaeologists and draftsmen would be able to work
during the course of the excavation. at their leisure.
The wood itself, however, was difficult to raise in one This proved to be the most difficult task of the excava-
piece. The fragments, mostly twigs and branches, lay tion. For three weeks, the entire staff worked in shifts,
partially cemented to the rock bottom, which was marble cutting through the solid bedrock on which the wood
lay. An air drill or air hammer would have saved much
time, and would have caused less damage to the fragile
material; during the weeks of work, with one diver
holding the chisel while another swung the sledge
hammer, it was impossible to avoid disturbing the wood.
This was the only time that we were aided by Mondo-
lillfi's helmeted sponge divers. Their heavy dress made
them well suited to the job of s\vinging the sledge, but
it was often necessary for an aqualung diver to guide
their leaden feet through the wreck with his hands.
They were unable to work in even moderate currents
(fig. 20).
When this lump was finally free, it was installed in
a box under water, lashed securely, and raised with
Lutfi Gelil's winch. It was carefully taken apart and
drawn weeks later in Bodrum (see chap. IV).
Finally there was the job of removing the sand which
covered the sea bed, and searching through it for small,
loose objects. At first this was clone manually by sim-
ply lifting handfuls of sand into a bucket, which could
be emptied away from the wreck. This proved to be
extremely time-consuming, however, and our air lift
FIG. 18. Dumas and Duthuit fill lifting balloon in area G. was lowered onto the site.
•

\'OL. 57, PT. 8, 1967] BASS AND THROCK~IORTON: THE EXCAVATIOI\ 31

The principle of the air lift is simple. The pressure
of the water at lower depths is greater than that of water
above. If a pipe runs from the sea bed upward, and air
is released into this pipe near its bottom, the air will
rise through the pipe toward the surface, expanding as
it goes. This expansion, contained and directed only
upward by the pipe, causes a suction at the, ba~e of the
pipe, and this pulls water, sand, and m~ld with It.
Two air lifts were used at Cape Gelidonya (fig. 21).
The larger was ten centimeters in diameter, made of
three lengths of light sheet-iron pipe, each five meters
long; this was suspended vertically fr0111 an oil-drum
float. At the lower end was a six-meter length of
flexible, but non-collapsible, wire-reinforced rubber hose
of the same diameter. Air was pumped from Manda-
liJ/(i's compressor through a small hose to an air intake
near the lower end of the air lift. A loosely meshed bag
tied over the upper end of the air lift caught coarse
sand and any tiny objects which were inadvertantly
sucked up, but allowed most of tile finer sand to be
carried away by the current (fig. 22).
Using an air lift for the first time, we ran into several
problems. If it was not vertical in the water, the pipe
tended to clog with sand. This forced air from the
flexible lower end, giving it enough buoyancy to lift its
operator off the bottom. The strong current also made
positioning of the lift difficult at times.
The second air lift, onlv five centimeters in diameter FIG. 21. Both air lifts positioned on wreck
but over thirty-five rueter's long, was able to discharge with L1ttji Gel-if above.
onto the deck of the boat above.
The air lifts proved to be most valuable tools when to explore. Normally a diver fanned sand gently to-
used properly, but they were used only for cleaning ward its mouth, searching for and extracting artifacts
sand away. We did not want to raise objects with them before they were caught in the strong flow.
for two reasons: we feared that fragile pieces would
suffer from the trip through the pipe, and objects
found only in the air-lift bag or on the cleek above
could never be placed with accuracy on the plan of the
wreck. TIle lifts were never used directly in the sand.
therefore, except in sucking sand fr0111cracks and from
beneath the boulder, where it was otherwise impossible

Pre. 20. Captain Kemal wearing standard diving dress FIG. 22. Martens sweeps sand into mouth of
works with Duthuit. larger air lift held by Bass.
 -
32 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TR.Al\S. AMER. PHIL. SOC.

scale differences were devised; the most promising

method of mapping is that of photogrammetry, which
involves the use of carefully controlled stereophoto-
graphs. At larger sites, such as those at Methene, sur-
veying techniques similar to those used on land have
been successful, and the lack of a horizon has proved to
be no hindrance. More advanced methods of preserv-
ing wood and metal finds are now being used, but it is
too early to comment on their success. The first results
of these excavations are now being published."

4. rrrs PROGRESS OF THE EXCAVATION

In order to present better the evidence of the excava-

tion, and also to give an idea of the rate at which the
work progressed-which may be of help to others
scheduling similar operations in the future-a serie~
of plans are appended. These present the work in the
three main areas of the site (P, S, and G), broken into
several arbitrary phases. These phases are not pre-
tended as true stratigraphic phases as are found on land
excavations, but represent the course of excavation and
do, therefore, show the relationship of objects to one
another.
A word of explanation about the plans is necessary.
They are based largely on a series of plans drawn by
Miss Frost at the time of excavation, but they have
been contracted into fewer phases than originally drawn.
Further study of triangulation measurements and photo-
graphs has corrected the errors which occurred at the
beginning of the excavation, before we had devised a
reliable system of plotting. For that reason, the posi-
tions of objects in areas G and S can be taken as quite
FJG. 23. Duthuit holds metal detector while diving
accurate, but the relative positions of some of the re-
with hookah.
mains in area P were judged only after careful study
of dozens of photographs.
A last searching device was an underwater metal It should also be noted that the plans cannot truly
detector (fig. 23) brought by Luis MareJen of the represent what was seen on the ocean floor. Most of
National Geographic Society on a visit during the last the material was invisible beneath seaweed and concre-
days of the season. This located several deposits tion, and the positions of many objects were determined
which Illight otherwise have been missed, and assured only after the lumps of concretion had been raised,
us that no major pieces of metal had been overlooked cleaned, and redrawn; objects which were not visible
beneath the limestone growing over the wreck and sur- on the site are shown with broken lines. Further, areas
rounding area. Although delicate and liable to mal- of sand, rock, and concretion were not defined as clearly
function, these detectors can prove valuable in opera- on the sea bed as they are in the plans because of over-
tions such as ours. growth; these areas became much clearer as the site
H should be emphasized that the methods of diving was cleaned (compare figs. 6 and 7).
and excavating clescribed above were only those used at General elevations may be found on figure 37.
Cape Gelidonya, a pioneer effort, and many are obsolete
at the time this is written. At Yassi Ada, for example, Area F, Fllase 1 (fig. 24)
it was found that divers could decompress for twenty-
On June 14, when first seen. only the vague shapes
one minutes, twice a day, and that even forty-minute
of Tn 7 and several bun ingots, and the handles of In ]3
decompression periods were not overly tiring for most
divers when they needed to make longer dives. At the 13G. Bass. AA (1962) 537-564; American Scholar 32, Z
(1963) 241-254; NotCco 124, 1 (July, 1963) ; Archoeotcq» 18 1
1
same site, new techniques allowed extremely clear (1965) 7-14; Archaeotocv Under Wafer (London, New York,
photographs to be taken from thirty or more feet above 1966) chaps. 6 and 8; E. Ryan and G. Bass, Antiquity 36 (1962)
the wreck, and almost all of the mapping was clone by 252-261. P. Throckmorton and J. Bullitt, Expedition 5, 2
photography after methods of correcting distortion and (1963) 16-23.
\'01.. 57, p'r, 8, 1967]
BASS AND THROClOiIORTON: THE EXCAVATION 33

AREA P PHAIE 1

••••
"".", .... "W.mo<

FIG. 24.

and J n 36 were visible in the lump of concretion (P II) and In 13, and the ingots beneath them, was raised to
which encased or covered most of the cargo in the area. the surface with a cable.
On June 17, scattered pieces of broken ingots, tools,
and basketry were collected from the sand around P II. Area P, Phase 2 (fig. 25)
Marty, if not most, of these pieces had been dislodged
during the previous summer when several entire ingots The remaining half of P II, containing In 7 and other
had been broken loose and raised; the impressions of ingots, was moved slightly out of position during the
these ingots in the concretion were still clearly visible raising of the first half of the lump; thus the exact
~nd are shown on the plan by hatched lines. Two bun relationship of In 15 and In 39 to the other ingots was
mgots (BI 14 and BI J9) may also have been dropped never firmly established. Careful study of photographs,
111 1959, and BI 8 was almost surely not in situ. however, has allowed us to accept the relationship shown
The area was drawn and photographed during the here with some confidence. The well-preserved bottom
following days, and more metal scraps, mostly un- of a wicker basket (BM 3, see Phase 1), which had
recognizable, were raised. On June 22, Duthuit and held scraps of broken tools and ingots, although resting
Dumas began to cut around P II so that the entire mass all In 10, stuck to concretion which had covered both
could be raised in one piece. It was not until June 29 In 7 and In 10, and was not revealed until June 30,
that an automobile jack could be placed under P II, to when the rest of P II was raised to the surface.
separate it from the sea bed, and the lump of concretion The large sponge, whose center was often used as a
then broke into two pieces; the piece containing In 10 fixed point for measurements, was found to contain half
34 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. AllIER. PHIL. SOC.

/
/
/
AREA , 'HAlE 2 I
I
I
/
~/

----

-=-=-=-
10 ....

.,."". ' .... W.m ••

FlG. 25.

of a bronze double axe which Illay have been dropped In early September, a general search of the area with
in 1959. air lift and metal cletector located a number of bronzes,
which must have come originally from area P, in the
Area P, Phase 3 (fig. 26) crevices under the base of the cliff: this area, an
When P II had been raised, two clusters of objects "extension" of area P, was labeled area E.
were found by In 15 and In 39; these were labeled as
sub-areas P V and P VI. It is assumed that they Areas G and s. Phase 1 (fig· 27)
represented the contents of two perishable containers
The "gulley" (area G), between the cliff and the
(scraps of matting suggest basketry), and included C0111-
large boulder, and the "sandy" area (area S) as first
plete ancI broken bronze implements, casting waste, and
found (ef. fig. 37). In 21, in area S, has been shown
two stone weights. The glass beads were imbedded in
in the various phases of area P (figs. 24-26), so that
encrustation, but had evidently been stored, along with
the relationship of the areas in the various plans may
a bronze bracelet, in a pottery container which was be notecl. As in area P, most of the ingots were in-
found crushed around them (fig. 141). Also in the visible beneath thick layers of concretion and sea
general area under P II were several lumps of white
growth, and their positions (shown with broken lines)
material, later found to be tin oxide, which had un-
were not actually determined until the lumps containing
doubtedly caused the serious damage to In 39 through
electrolysis. Sub-areas P V and P VI were excavated them were plotted, raised, and broken apart. Roman
mostly between June 31 and July 11, although the lump numerals in this case denote masses which were to be
containing the beads was not raised until August 11. raised intact.
 VOL. st, PT. 8, 196i] BASS AND TI-1ROCK~[ORTON: TI-IE EXCAVATION 35

/
/
;
AREA P PHASE I ;
/
I Ell
/
/'
//

, I

11'0.:
'21=
.. ~'C= ,
"° 5
0 ,"
"'05c::J
"'1Q ....
..
.. ' ·,
..
a'
.. .. '"
. ."3~0
"
Il f, •• c> ..
I

.:. ,

..' . . ".

P IV

'o.....
••••
• , •• Ioy ••••• Womo.

FrG.26.

Areas G and S, Phase 2 (fig. 28) oxide). but leaving In 29 1:nsitu. The raising of the
On June 18, the concretion between In 19 and In 21, stone "anvil" (St 7) revealed small bronze and stone
bearing the impression of an ingot raised in 1959, was objects beneath.
broken in half and raised. Stuck to its bottom were
pieces of pottery (P 14 and a fragment of P 2) ; BI 15 Areas G and S, Phase 4 (fig. 30)
and the fragment of another bun ingot also came to In breaking pieces of concretion away from the south
light. Lumps G I and II were broken away on June 20, edge of area G, we revealed metal fragments, ballast
revealing bronzes, a slab ingot, and traces of wood stones, and the ends of twigs. The air lift was used
beneath G IV. G II was not actually raised until J nly 6, for the first time on ] uly 15, and by sweeping sand
and on the same day G III (containing In 16--19) was down to it from the area formerly occupied by G III
attached to a rope to be raised. and St 7, we uncovered a number of bronze tools and
wood fragments, including twigs.
Areas G and S, Phase 3 (fig. 29)
Areas G and S, Phase 5 (fig. 31)
G V (which proved to he only a very large rock)
From July 16 to August 1, work in area G 'was con-
and In 24, resting on G V, were raised on July 9,
fined mainly to cleaning the sand from the floor of the
along with G IV. G VI (containing In 26 and In 27) "gulley," ancl it was during this time that the scarabs
and G VII were raised two days later, revealing wood and the first of the maceheads (St 1) came to light. At
fragments and white substance (which prove to be tin the same time, many more twigs were uncovered in the
36 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHJL. SOC.

AREAS G8<S PHASII 1

''''''

,"

.
, '. ,

,/
. - ...
la,m •

•• _ .. "

FIG. 27.

AREAS G 8<S PHASE 2

.........

·10 om.

!!i=!! wo!!' •.
FIG. 28.
 VOL. 57, PT. 8, 1967] BASS AND THROCK~'iORTON: THE EXCAVATION 37
AREAl Ga.l PHAIE J

"c.." •

CD• N

(:-..:-,~--_/ ...... ()
I,,»

I
•

,
,
.ii'_._
I

,I
..""".
FIG. 29.

AREAl Ga.l PHAIE 4

,
I
-. '.'

FIG. 30.
38 CAPE GELIDONYA, A BRONZE AGE SHIPWRECK [TRA:"S. A!\lER. L'HIL. soc,

AREAl G8<I PHAIE S

0 ........

,"
/"/,
'--..
~.
~-''<''\
InU

•. ". ".•. lO,m. .

..
:~-:!!I-.
F,G. 31.

AREAl G&I PHAIE 6

•......
F,G. 32.
 VOL. 57, PT. 8, 1967] BASS AND THROCIC\IORTON: THE EXCAVATION 39

AREAl G&I PHAI E 7

'''-" ,

)
)

I
•

FIG. 33.
_.-
10'""

breaking away of concretion just below In 22-23, and weights, bronzes, and ballast stones) which had been
In 30-32. Some of the wood near In 21 was removed. crushed beneath the ingots and concretion. This mate-
rial rested on the twigs, which were covered (except
Areas G ond S, Phase 6 (ji.g. 32) for thei r protruding ends) by a mixture of concretion
vVe determined to raise the remaining large mass of and sand. Under G VIn was found a large stone
concreted material (G VIII) from above the remains which, when removed, revealed that the tin oxide
of wood, and this was chipped loose and tied with ropes seemed to be coming from a box-like cavity in the
by August 2. The following day it was raised to the concretion. In 33 was removed.
surface in our first use of lifting balloons during the The remainder of the season, until the first week of
excavation. In 32, and an adjoining ingot fragment, September, was spent in cutting around and raising the
were also raised; this broke the concretion into smaller lump which held the twigs. During this time the lamp
lumps ( G X. XI, XII) which were raised separately.
(P 29) and the second ruacehead (St 2) were found by
New wood was appearing near In 21.
In 29. \i\1hen work was completed, all of the remains
Areas G and S, Phase 7 (fig. 33) had been raised to the surface and the area had been
The removal of G VIII-XII revealed a jumbled mass searched with a metal detector for any invisible
of material (including metal foil, slab ingots, whetstones, deposits of bronze.
 IV. THE SHIP AND ITS LADING'

GEORGE F. BASS

Little was preserved of the ship itself, for the rocky lines on the plan, were determined by the impressions
sea bottom onto which it sank was covered by only a which they had left in the sea bed.) As the ship broke
few centimeters of sancl. With no protective coating of into pieces, various parts of it finally did tip over, to a
mud, most of the hull had disappeared. The arrange- greater or lesser degree depending on the slope.
ment of the heavy cargo (fig. 37) on the sea bed, how- Most of the ingots were found still stacked in three
ever, indicates that the ship had not tipped over before major piles, as they must have been in the ship; those
sinking, but had settled fairly evenly, probably after that settled on a flat bottom, as in area P, had barely
taking on water through one or more holes torn by shifted from their original positions (fig. 17). The
jagged rocks. The keel may have come to rest on the stacking arrangement in areas G and S were still ob-
high rock just south of In 21, and broken gradually vious, although the piles had slid downhill when the
under the weight of the cargo; this would have allowed side of the ship gave way (fig. 16). Figure 91, in
the ingots at the two ends of the ship to come to rest chapter V, presents the most probable order in which
as they did in areas G and P, and would explain why the ingots were originally stacked, based on their posi-
the central group of ingots seems to have separated, tions on the sea bed (figs. 38, 39, and 40). Traces of
with 1110Stingots settling into area S, but with two fall- matting found on many of the ingots, especially those
ing to the southwest. (These latter two ingots were in area P, suggest that the ingots were wrapped to-
raised in 1959, but their positions, indicated by dotted gether, or that there were at least layers of matting be-
tween ingots.
1 A detailed study of all of the underwater photographs and _ Most of the smaller bun and slab ingots also were
drawings led to few definite conclusions about the wooden found in stacks, but some of the bun ingots were found
remains. When a better preserved ship of the same period is
separately. The remainder of the metal cargo, which
excavated in the future, a restudy of those records would
probably result in the proper identification of some of the consisted of complete and fragmentary bronze imple-
wooden remains of this ship. ments, was scattered throughout the site (fig. 41).

DISTRIBUTION OF INGOTS 5
I
L 0 ~ INC DOWN
,(-

FIG. 37.
44
VOL. .57, 1'T. 8, 1967] BASS: THE SHIP AND ITS LADING 45

In29

INGOT FRAGMENTS

FIG. 38. Ingots from area G as seen from north.

Often the bronzes were found in small groups, which lamp, maceheacls, whetstones, an astragal, a cylinder
suggested that they had been carried in bags or baskets, seal, weights, and traces of food (figs. 43 and 44)-
and this suggestion was borne out by the discovery of gives fairly certain evidence that the living quarters of
a well-preserved portion of a wicker basket which was the ship came to rest in area G. (Some of the smaller
tightly packed with bits and pieces of metal. pieces were found in crevices under the Boulder, but
The distribution of this heavy cargo suggests that the they were surely carried there from area G by the cur-
ship was about ten meters long; this is only slightly rent.) Although there was nothing to suggest that
shorter than the figure of forty feet, based on intervals there was an actual cabin here, we may assume that
between oarsmen, suggested by Seymour for a Homeric some sort of shelter was provided and that this was, as
merchant ship." is still normal, near the stern. A more tenuous reason
Little could be learnecl from a study of the distribu- for believing that the stern was in area G was the dis-
tion of pottery in the wreck (fig. 42). covery of the triangular, pierced stone which lay farther
The distribution of what I consider the personal pos- east than any of the cargo (figs. 6, 10, and 37).
sessions of the captain and his crew-scarabs, the ship's Whether or not this was a crude stone anchor was de-
bated throughout the course of the excavation; the con-
sensus, after close examination, was that it was only a
natural stone, but it does deserve mention.
It was virtually impossible to determine with any cer-
tainty which members of the ship were represented in
the jumble of wood. The wider planks generally ran ill
an east-west direction, which suggested that they were
strakes running the length of the ship. The only wood

FIG. 39. Ingots from area S as seen [rom north.

2 Life -in the Homeric Aoe, 308. :rIG. 40. Ingots from area [I.
46 CAPE CELIDONYA: A BRONZE AGE SH1PWRECK [TRANS.• "'ER. PHIL. SOc.

_AlSEn IN lU9, POS,T'O'l5 UNKNOWN'

{}...., "",,, "".",."."
,,,.,,,.,,,.,,,.,,,,,.,.,,,,,,,.,,,,,,,,,,,.,,,
,:]..
.

'H' O,N,", ... ., '0' .. ON'" w~o.. 'O""ON' wn' NQ'""'''lY O",.,.,N'. '.... IN." •• ,
... ~o".'"THI,"'AlCO, (H,,,,, •••

FIG. 41.

DISTRIBUTION OF POTTERY

j
( \~~-~.~--

el'"
\
5L!',NG D.~WN
\.
o o

IOUlPU

•.
,", ."",.. ,,",0"'--- •• ,
,. !, ••

FIG. 42.
 VOL. 57, PT. 8, 1967] BASS: THE SHIP AND ITS LADING 47

,
•

..
""

. "
.--..--
~"'''_:~
~
.-~- CAPE GELIDONYAWRECK
FIG. 43.

DISTRIBUTION OF WEIGHTS

, . '\""".. ~25
{'"
/:y JL~~~~~r~-~/
.. ~/\r: it"
luI (; vf::/
j fI
a
~
..
.
..
~J'."
I
;i;;~'
....
.
~
~
'L
> r~~\~,... '~.
" t;~ \

--~_~~ Kn

~ ~:;E GELlD:~YA ~=K

FIG. 44.
48 CAPE GELIDONYA, A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

that was preserved in its original context. however, was

that raised in a lump of concretion and disassembled on
land (figs. 45 and 46). This showed positively what
had been suggested earlier, that the narrower pieces of
wood ran perpendicularly under the broader pieces,
and were fastened to them with treenails (1.5 to 2 em.
in diameter). The spacing of the treenails, or wooden
pegs, would have allowed the wider pieces to lie closely
side by side, so that we may have remains of the ceiling
(the inner lining) of the ship.
Running also in an east-west direction, and higher
011 the slope 011 which the wooel lay, was a much larger
piece of wood (pres. L. .24, W .. 143, Th .. 057) which
may be seen in figures 45 and 46. That it was not part
of a gu nwale was proved by the discovery of many ,
weights and other objects, including some of the 116 •
kilograms of ballast stones from the wreck, even higher
on the slope; these objects would have settled to the
bottom of the ship, and there seems no conceivable way
in which they could have reached their positions unless
the large piece of wood was either the keel or, more
FIG. 45. Eg-demir. drawing remains of wood in situ~' probably, simply a heavier strake. The possibility
Wd 1 at extreme left. naturally exists, however, that any of the wood frag-

Semi

FIG. 46. Wood remains after having been cleaned of concretion 011 land.
 VOL. 57, PT. 8, 1967] BASS: THE SHIP AND ITS LADIl\G 49

I FIG. 48. Detail of some of brushwood stilllying across

• planks of hull after cleaning; cj. fig. 47.

the freighter on which he was traveling in 1959 "took

on bundles of cuttings either of rushes or of twigs" in
Beirut. "Inquiry of what was being done with them
received the reply that they were to be used as a type of
FIG. 47. Brushwood remains resting on south edge of area G. matting on which later cargo of grapes and citrus fruit
would be placed ... at Cyprus and at Gaza." 3
The brushwood aids considerably in an understanding
merits were simply pieces of chests or other furnishings of Odyssey 5. 256-257, which had previously pre-
in the ship. sented difficulties in translation:
Over the "ceiling strakes." and perpendicular to them,
lay a large pile of sticks (figs. 47 and 48). They varied ¢pa..~€ Of J.l.tv PI1r€(1(H OLajJ.'Trf.pf.5 olcrvIV'WH
in size, but none was probably ever more than a meter d.!}.LaT05 f.I'Aap (J.l.f.v· 7rO'A'At]1I O'E7rf.Xf.VaTO VA1jv.
long; their ends were often preserved, cut diagonally
The VAlJ has been considered part of the wattle fenc-
across as if by an axe, and bark and twigs remained
ing which Odysseus built as protection against the
011 most of them. They surely had nothing to do with
waves, or at least a backing for such a fence; it has
the construction of the ship, and G. Gruben suggests
even been translated as ballast or a bed. It now seems
that they might have served to protect the thin planks
that only literal translation is required, for brushwood
against the heavy metal cargo; it is certain that they
was probably the usual dunnage in such early ships.
lay beneath the ingots.
From the cargo and from the few fragments of wood
Ships following the same route (see chap. XIV) as found at Gelic1onya, we have gained a picture of a small
did the Gelidonya ship continue this practice of cushion-
ing their cargoes. G. L. Harrington informs me that 3 From a letter to the author, dated 2 April, 1963.

FIG. 49. Syrian merchant ships in port; from Tomb of Kenamon at Thebes (from Davies and Faulkner).
50 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRAKS. AMER. l'HIL. soc.

.- .-_.~-~._-. coastal vessel about the same SIze as those used by

K /!!~\:::/r sponge fisherman throughout the eastern Mediterranean.
(1."",/'/ \ \. ~ '~ It had 011 board at least a ton of cargo, with, in addition,

, \~ij~ti2~TI.~~:~~~~~lt:\
whatever perishable cargo was contained in its storage
Jars. Unfortunately, however, the hull itself gives no
clue as to the nationality of the ship (cf. figs. 49 and
50)!

:r''·····---------- -----::.-.~--
---- -----_.--- ---- -.~..::'.- ----.-~---"--: 4 The ships shown here are Syrian. FOI- the scanty evidence
0__~_~ ~ • _••. _ ~ ._: we have of the appearance of Bronze Age ships, mostly from
fanciful and inaccurate drawings and carvings, see Spyridon
FIG. 50. Syrian ship; from Tomb of Nebamun at Thebes Marinatos, "La Marine creto-mycenienne," BeH 57 (1933)
(from Save-Soderbergh, courtesy Griffith Institute). 170-235, and G. S. Kirk, "Ships on Geometric Vases," BSA 44

a
, ,

I
I
,

Wd4
,
Oems 10 20

FIG. 51. "Vood fragments.

VOL 57, 1'1'. 8, 1967J BASS: THE SHIP AND ITS LADING 51
were found loose in the sand and could not be related
to other parts of the "..'reck: 5
Wd 1. Largest fragment, well preserved on three sides.
Pres. L. .24, W . .143, Th .. 057. It had badly
warped from drying before being photographed or
drawn.
Wd 2. Fig. 51. Fragment of a plank with peg hole, and
squared edges. Pres, L. .154, W. 053, Th. 013,
D. of peg hole .018.
Wd 3. Fig. 51. Ragged fragment of plank, with many
knots. Pres. L. .18, W . .135, Th .. 018, D. of peg
hole .022.
Wd 4. Fig. 51. Fragment of plank with two peg holes.
One poorly preserved face with slight convex
curve and traces of bark. Knots at either end.
Pres. L. .195, \iV .. 05, Th. after some drying .01,
D. of peg holes .018, holes .123 apart.
Wd 5. Fig. 52. Short length of plank with ends cut to fit
other pieces of plank at either end; its position may
be noted in figure 46. Another adjoining piece,
FIG. 52. W d 5. extending downward, was seen under water, but it
quickly fell away. It would seem that these
Following is a list of representative wood fragments separate pieces were held in position by the pres-
from the wreck; twenty-five other pieces were cata- sure of strakes on either side of them, for they
logued, but these were too fragmentary to be meaning- were not fastened together in any way. L. .134,
W .. 077, Th .. 024.
ful. The original positions of a few of the pieces may Wd 6. Fig. 5]. Large fragment of planking attached to
be seen in figures 33 and 46, but many of the examples concretion holding two metal tools. Pres. L. ,21,
W . .19.
(1949) 116-117, which includes several of this period. To the Wd 7. Fig. 51. Fragment of split branch with copper or
bibliographies found in these articles may be added James bronze nail through knot; fish bone and fiber at-
Mellaart, "The Royal Treasure of Dorak," ILN 235 (1959) tached to concretion. Pres. L. .085, W .. 0l2.
754, with fig. 2, and the crude graffiti at Enkomi, Schaeffer, Wd 8. Fig. 53. An assortment of fragments from the
Enkomi-Atasia 102, with fig. 38, and on Malta, Diana Woolner, ___ --":cJrushwooddunnage, drawn to the same scale.
"Graffiti of Ships at Tarxien, Malta," Antiquity 30 (1957)
60-67. S For identification of types of 'wood used, see append. 2.

>.
-. "

FIG. 53. Selection of twigs from brushwood dunnage.

 V. THE INGOTS

GEORGE F. BASS

The bulk of the cargo consisted of copper and bronze whole ingots will be seen to consist of joined halves.
ingots and ingot fragments. Three major types were It is probable that all of these were broken in modern
present; so-called "oxhide" ingots, discoid "bun" ingots, times by the discoverers of the wreck, during attempts
and flat oval bar ingots, here called "slab" ingots. The to free them from the sea bed.) Lastly, there were al-
presence of rectangular bars of tin is also probable. most seventy-five kilograms of ingot fragments which
had been cut or broken away at random from complete
J. "OXHlDE" INGOTS ingots.
These are flat, oblong pieces of copper; (see append. 3) All examples are of Buchholz's Type 2 (fig. 55).'
roughly four centimeters thick and averaging sixty by We shall follow his typology for the sake of continuity,
forty-five centimeters in length and width; each has although, as we shall see, these types may not be
protrusions or handles at its four corners. One side of chronologically separated to the extent he believed.
each ingot is always rough and bubbly, while the other The Gelidonya ingots may be grouped further into •
is much smoother; the smoother side does, however, three subtypes. Type 2a has its sides pinched in to
usually contain low mounds and tiny air holes, and is form a waist, but its ends are almost straight across in
often outlined by a raised rim. some cases; the handles are thick extensions of the
At least thirty-four ingots of this shape were carried sides. Type 2b is the same, except that it has a more
in the ship, stacked in neat piles with matting serving pronounced rim on its smoother side. Type 2c has
as some sort of wrapping or packing (see chap. IV). incurving sides which extend to form outcurving taper-
They are more or less complete depending on their ing handles; a slight rim on the smoother side is usual.
situation on the site; those lying in contact with tin The rough, bubbly sides of twenty-four of the ingots
(figs. 40 and 59: 22), for example, have been badly cor- contain signs which seem to have been stamped into the
roded by the natural electrolysis which took place in the newly poured, still soft metal, usually at the center of
salt water, and some are missing almost half of their the ingot near one end. In four cases there are addi-
original metal. tional signs, which also seem to have been made with
Five half ingots seem to have been broken or cut in stamps, on the smoother sides of the ingots; all of these
antiquity, although the breaks are often so corroded signs are approximately the same, EB or ®' and are
that it is impossible to know whether in each case the found in the centers of the ingots. A few of the ingots
missing half was simply eaten away by corrosion; none have, 111addition, "secondary signs" which were incised
of these halves join or are from the same ingot. We into the metal after it had cooled and hardened. These
may presume that these ingots were fairly carefully cut last signs are, without exception, on the smoother sides.
or sawed in half, however, because twelve ingot corners The following catalogue lists [or each ingot the over-
were found, each of which was certainly cut out with all length, the maximum width (recorded across the
two deliberate strokes running at approximate right handles), the minimum width (recorded across the
angles to one another (fig. 54).' (A number of the waist), and the thickness (usually greatest on handles
and at edges). Weights of well-preserved ingots are
accepted as being very nearly original, for so little cor-
rosion has often taken place that surface details are still
sharp, while shiny copper appears under the scratch
of a pin. Original weights of ingots missing handles
and other bits have been estimated by placing extra
handles and other pieces of matching size and shape on
the scales during weighing. The estimates of original
weight for badly corroded ingots were almost impossible
to make, for in some cases it was possible to drill almost
completely through an ingot without striking solid cop-
per; these pieces were always quite light and have lost

FIG. 54. Ingot

....... _---
handles cut or broken off in antiquity.
a great cleal of weight.

~ Hans-Gunter Buchholz, "Keftiubarrcn und Erxhandel in

--- zweiten vorchristlichen j ahrtausend," PZ 37 (1959) 4-6, with
1 We have evidence that, at least at Mohenjo-Daro, copper fig. 2 on p. 7; "Der Kupferhandcl des zweiten vorchristlichen
ingots received preliminary saw cuts along which pieces were Fahrtausends im Spiegel del" Schriftforschung," Minoica, 93-
snapped off. E. J. H. Mackay, Moliewjo-Doro I, 452. 95, with fig. 2 on p. 96.

52
 VOL. 57, PT. 8, 1967] BASS: Tl-IE lNGOTS 53
In 1. Type 2b (fIg 56, a and b). L. .61, VV .. 35. min.

In 2.
W .22.Th
Impressed
045-05 Complete. Weight 19.95 kg.
sign (fig. 90: 39A or 33A).
scribed near one edge of smoother
2IBand22B).
Type 2b (fig. 56, a and b).
Signs in-
side (fig. 90:

L. .74, VV .. 35, min.

W . .25. Ingot made of two joining halves. Weight
21.75 kg. Estimated original weight 22.3 kg. Im-
CJ la

pressed sign (fig. 90: 23A).

In 3. Type 2c (fig. 56, a and b). L. .63, W . .45, min.
VV. 30, Th .. 03-.04. One handle missing.
18.5 kg. Est. orig. weight 19.5 kg.

W .. 28, Th .. 02-.04. Made of joining

Weight
Almost no
trace of rim. Impressed sign (fig. 90: 24A).
In 4. Type 2c (fig. 56, a and b). L. .66, W . .45, min.
halves.
Weight 18.15 kg. Est. orig. weight 18.8 kg. Slight
CJ lb

rim. Impressed sign (fig. 90: 25A). Traces of

fiber on one side.
In 5. Type 2c (fig. 56, a and b). L. .58, W .. 44, min.
• W . .26, Th .. 04. Made of joining halves.
22 kg. Slight traces of rim. Impressed
Weight
sign (fig.
2a
90: 26A). Traces of matting on smoother side.
In 6. Type 2b (fig. 56, a. and b). L. .75, W .. 37, min.
\iV. .22, Th. .025-.035. Detached handle joined.
Weight 20.7 kg. Impressed sign (fig. 90: 27A) .
Slightly off center.
In 7. Type 2c (fig. 57, a and b). L. .67, W . .45, min. 2b
W .. 26, Th .. 03-.04. Two handles broken. Weight
19 kg. Est. orig. weight 19.1 kg. Very slight
traces of rim. Impressed sign (fig. 90: 28A).
In 8. Type 2c (fig. 57, a and b). L. .67, W . .44, min.
VV . .26, Th .. 025-.042. Complete except for a few
grams missing from one edge. Weight 20.8 kg. 2c
Slight rim. Impressed sign (fig. 90: 25A).
In 9. Type 2c (fig. 57, a and b). Pres. L. .68, max. pres.
"'Ar . .44, min. W . .28, Th .. 025-.030. Two handles
missing. Weight 20.75 kg. Est. orig. weight 22.2
kg. Impressed sign (fig. 90: 25A). Matting ad-
heres to smoother side and one edge.
In 10. Type 2c (fig. 57, a and b). Pres. L. .53, W. ap-
prox . .43, min. W .. 26, Th .. 025. Three handles
missing. Weight 15.6 kg. Est. orig. weight 16.6
kg. No sign. Traces of rope from netting and 3
basket adhere to one side.
In 11. Type 2b (ng. 57, a and b). L. .78. pres. W .. 33.
min. W . .23, Th .. 04. Two handles missing. Weight
16.85 kg. Est. orig. weight 18 kg. Impressed sign
(fig. 90: 29A), near base of one handle. Traces Ingot Types Adapted From Buchholz
of matting on smoother side.
In 12. Type 2c (fig. 57, a. and b). L. .66, Vv. approx . .45,
min. V·/. .27, Th .. 04. Two handles chipped. Weight PZ 37 (1959) p, 7
23.4 kg. Est. orig. weight 23.6 kg. Slight rim. FIG. 55.
Impressed sign (fig. 90: 30A). Matting over
smoother side, one edge, and end. mig. weight 20.35 kg. Any sign would have been
In 13. Type 2a (fig. 58, a and b). L. .67, W .. 38, min. obliterated by surface damage through corrosion.
W . .25, Th .. 025-.04. Complete. Weight 25.9 kg. In 17. Type 2b (fig. 58, a and b). L. .58, W .. 32, min.
Oval concavity, near center of one edge of rough W . .205, Th .. 045. Small area of smoother side
side, may be a sign or only a blemish. missing, but weight taken with 1-1.25 kg. of con-
In 14. Type 2a (fig. 58, a and b). L. .70, W .. 36, min. cretion still adhering. Weight 23.0 kg. Est. orig.
W .. 23, Th .. 023-.04. Complete. Weight 20 kg. weight 21.75-22 kg. Impressed sign (fig. 90: 24A).
Impressed sign (fig. 90: 30A). In 18. Type 2a (fig. 58, a and b). L. approx .. 64, W. un-
In 15. Type 2c (fig. 58, a and b). Original L. unknown, known, min. W . .195, Th .. 032~.036. Three handles
W. apprcx . .45, min. VV. .26, Th .. 025. Two handles missing. Weight 17 kg. Est. orig. weight 19 kg.
missing; metal in very poor condition. Weight 17.5 No sign.
kg. Est. orig. weight 19.5 kg. Oval concavity near In 19. Type 2b (fig. 59, a and b). L. .54, W .. 315, min.
base of one handle, on rough side, may be sign or W . .195, Th .. 04. Complete. Weight 22.2 kg. Im-
only a blemish. pressed sign (fig. 90: 28A).
In 16. Type 2b (fig. 58, a and b). L .. 54, W. 31, min. In 20. Type 2b (fig. 59, a and b). L. .63, W .. 365, min.
VV . .19, Th .. 04-.045. Part of one handle and part W . .22, Th .. 038-.048. Complete. Weight 21.9 kg.
of rough surface missing. Weight 20.2 kg. Est. Impressed sign (fig. 90: 30A).
54 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. A~IER. PHIL. SOC.

•
In 1 In2

)Scm,

In4
In3

In'

In6

FIG. 56. Ingots In I-In 6.

 VOL . s t, I'T. 8, \SJ6i] BASS: THE INGOTS 55

• 1.1
I••

35cms

I.' In 10

In12
I n

FIG. 57. Ingots In 7-In 12.

56 CAPE CELIDONYA: A BRONZE ACE SHIPWRECK [TRANS. A,,]ER. PHIL. soc.

•
In13 In 14

35 em. •

In16

In 17 In 1.
FIG. 58. Ingots In 13-In 18.
 VOL. 57, PT. 8, 1967J BASS; THE INGOTS 57
In 21. Type 2c (fig. 59, a and b). 1. .64, W. .44, min. In 35. Type 2c (fig. 61, a and b) Half ingot. Max. pres.
W .. 26, Th .. 025-.04. Complete. Weight 19.7 kg. L. .34, W. .44, Tb. 04 at edge, .02 in center.
No sign. Weight 8.5 kg. Sign (fig. 90; 40B) incised on
In 22. Type 2c (fig. 59, c and b). L. .59. No other dimen- smoother side near handle; no impressed sign pre-
sions preserved. Half of ingot eaten away by con- served on this half.
tact with tin. Weight 10.7 kg. Est. orig. weight In 36. Type 2c (fig. 61, a and b). Half ingot. Max. pres.
21 kg. (I) Impressed sign (fig. 90; 28A). L. .36, VV. AS, Th .. 026 near center. Weight 8.75
In 23. Type 2c (fig. 59, a and b). L. .68, v«. approx .. 45,
• mill. \IV. approx . .25, Th .. 025-.035. One comer
kg. Incised signs (fig. 90: 42B) near base of handle
.on smoother side; no impressed sign preserved on
and part of midsection eaten away by contact with this half.
tin. Weight 15.5 kg. Est. orig. weight approx. In 37. Type 2c (fig. 61, a and b). Half ingot. Max. pres.
22 kg. (?) Impressed sign (fig. SlO:27A). Traces 1.. .31, W . .40, Th .. 025-.045. Weight 8.85 kg.
of matting on smoother side. Rim is more pronounced on rough side, but also oc-
In 24. Type 2c (fig. 59, a and b). 1. .615, W . .45, min. curs on smoother side. Impressed sign (fig. 90:
\hi. .26, Th .. 03-.04. One handle chipped very 41A).
slightly. , Weight 17 kg. Impressed sign (fig. 90: In 38. Type 2c (fig. 61, a and b). Half ingot. Max. pres.
2SA). Impressed sign (fig. YO: 34A), in center of L. .33, 'lvV. AS} Th .. 04. Vi eight 9.5 kg. Rims on
smoother side. both sides. No sign.
In 25. Type 2c (fig. 60, a and b). 1. .59, W . .40, min . In 39. Type 2c (fig. 61). Half ingot. Max. pres. 1. .37,
• W . .26, Th .. 028-.04. Tip of one handle missing. W . .40, Th .. 03. Weight 10 kg. No sign.
Weight 21.3 kg. Est. orig. weight 21.45 kg. Im-
pressed sign (fig. 90: 27A). Impressed sign (fig. Ingots} and representations of ingots, of this general
90: 35A) in center of smoother side. shape have been found scattered widely throughout the
In 26. Type 2c (fig. 60, a and b). L.. 60, W . .43, min. W . Mediterranean area (fig. 92). The following list of
.255, Th .. 025-.053. Large section of corner miss- actual examples, which must be considered in a study
ing. Weight 16 kg. Est. orig. weight 18 kg. of the Gelidonya ingots, is based on H. G. Buchholz's
No sign on rough side. Impressed sign (fig. 90:
3SA) in center of smoother side. catalogue," where fuller descriptions and complete bib-
In 27. Type 2b Uig. 60, a. and b). 1. .62, W .. 37, min. liographies for each may also be found; I have added
\!I/ . .23, Th .. 035-.05. Part of one handle missing. my subtype letters where possible to the type numbers
Weight 23.8 kg. Est. orig. weight 24.1 kg. Im- which he has given. Dr. Buchholz has kindly supplied
pressed sign (fig. 90: 25A) near center of one edge
me with additional references which he had noted since
of rough side. Possible impressed sign (fig. 90:
35A) in center of smoother side, but this may be his study appeared, and I have added the new finds
only a blemish. from Mycenae and Zakro, and those newly listed by H.
In 28. Type 2a (fig. 60, a. and b). 1. .55, W .. 37, min. Catling." I have excluded the ingot front Bogazkoy 5
W . .23, Th .. 035-.045. Two handles and part of as it does not seem to me to be truly related to the
midsection missing. Weight 18 kg. Est. orig.
weight 21.5 kg. Very slight ridges. Impressed sign others. For these reasons, the numbering here does not
(fig. 90; 36A) in center of rough side. coincide with that of Buchholz's original catalogue.
In 29. Type 2b (fig. 60, a. and b). Max. pres. 1. .455,
• W .. 303, min. W .. 22, Th .. 035-.055. Two handles I. Syria-Palestine
eaten away by contact with tin. Weight 16.85 kg.
Est. orig. weight 20 kg. Impressed sign (fig. 90: 1. Tell Beit Mirsim.? Miniature ingot of Type lb.
30A). Broken, but probably originally about .16 long.
In 30. Type 2b (fig. 60, a and b). 1.. .58, W .. 345, min. Dated stratigraphically to the first half of the six-
W. 235, Th .. 025-.04. Complete. Weight 21.8 kg.
Rim sharply delineated by grooves. No sign. teenth century B.C.
In 31. Type 2c (fig. 61, a and b). L.. 62, W . .43, min. 2. Ras Sha111ra.7
W . .265, Th .. 025-.035. One corner and part of
midsection eaten away by corrosion. Weight 14.95 II. CYPY1tS
kg. Est. orig. weight approx. 18 kg (?) Incised
sign (fig. 90: 39B). Possible impressed 0 in 3. Enkomi." Type 3. Primary mark (fig. 90; 18A).
center of same, smoother side.
Stratigraphy not observed at time of excavation,
In 32. Type 2c (fig. 61, a and b). Max. pres. 1.. .49, W.
.43, min. W . .25, Th .. 02-.03. One end missing. but later elated to twelfth century B.C. by Schaeffer .
Weight 12.85 kg. Est. orig. weight approx. 16 kg. 4. Enk0111i.9 Four fragments of unknown type.
(1) Impressed sign (fig. 90; 30A). Incised sign
)
(fig. 90; 38B). SpZ 37 (1959) 28-39.
In 33. Type 2b (fig. 61, a and b). L.. 61, W .. 35, min. 4 H. Catling, C3JP'l"iot Broncetoork, 267~270.
W .. 235, Th .. 04-.045. Hole corroded through near 5 Buchholz catalogue no. 9, citing K. Bittel, Bogazkoy III, pl.
one edge. Weight 19.35 kg. Est. orig. weight 19.7 21; 3.
kg. Ingot bent so that rough side arches upward. 6 'vV. F. Albright, "Tell Beit Mirsim II," AASOR 17, 1936-
Impressed sign (fig. 90: 32A). 37 (1938) 54, with pl. 41; 13.
In 34. Type 2a or 2b (fig. 61, a and b). 1.. .622, W .. 345, 7 Dr. Buchholz has informed me of an unpublished fragment
min. v\l. unknown, Th. at handle .05, at edge .04. froin the 1960 excavations.
Extremely corroded metal in many pieces. Weight 8 Buchholz cat. 110. 1, with pJ. 3: 1; Schaeffer, Enkomi-
10.9 kg. Original weight impossible to estimate. A lasia, 30: no. 13, with pI. LXIII: 3.
Impressed sign (fig. 90; 39A). 9 Catting, C3,pr. Bronce., 267, no. 2.
58 CAPE GELIDONYA: A BRONZE AGE Sf-lIPWRECK I'TI~ANs. AMER. PHIl .. SOc.

In 19 In 20

35c:mf

1n21 In22

1.,,24

F1G. 59. Ingots In 19-1n 24.

 VOL. 57, PT. 8, 1967] BASS: THE INGOTS 59

In 25 In26
•

In27 In 2'

'029 (·,,30

FTG. 60. Ingots Tn 25-Tn 30.

 ..
I
60 CAPE GELIDONYA: A BRONZE AGE Sf-IIPWRECK [TRANS. AJ\lER. PHIL. SOC.

In 32

•
o

( ,
I

35cm.

In34

In35

FIG. 61. Ingots In 31-In 39.

•

!
I

! VOL. 57, PT. 8, 19671

5. Enkomi.!"
Stratigraphy
Type 3. Primary mark (fig. 90: 19A).
BASS:

not observed at time of excavation,

TH E INGOTS

24. Mcchlos.:"
25. Knossos."
No other information available.
Fragment. Late Minoan I/II (ca.
61

but later dated to 1400-1200 B.C. by Dikaios. and 1600-1400 n.c.):

to Iron Age I (twelfth to eleventh centuries B.C.) 26. Tylisscs.:" Type Ia.. Late Minoan I/II (ca. 1600-
by Schaeffer. 1400 B.C.).
6. Probably Enkomi.:1t Type 2. Primary mark diffi- 27-45. Hagia Triada." All Type I, but at least one
cult to identify. No stratigraphic evidence for elate. seems to approach Type 2c. Middle Minoan III
7. Enkon-i." Type 3. Corner fragment. or Late Minoan 1/11. Inscribed marks (fig. 90:
8-9. Maihiati." Type 2 or 3. More than twenty frag- I-lOB) .
ments of more than two oxhicle ingots. No strati- 46--51. Zakros.:" Types la and Ib, but one approach-
graphic evidence for date. ing 2c. No visible signs, but further cleaning
10. Mathiati.>' Miniature ingot of Type 2; less than planned at the time of this writing. Dated strati-
half preserved. No stratigraphic evidence for date. graphically to Late Minoan Ia (ca. 1600-1500
-l l . Probably Enkomi." Miniature ingot of Type 2. B.C.) .
L. ca.. 088. No stratigraphic evidence for date.
Signs. V. Greece
12. Bay of Soli. An oxhide ingot, raised from the sea, 52. Mycenae.":' Type Zb, Impressed sign (fig. 90:
was described by a commercial diver to Eric Ryan 17A). Fourteenth century B.C.
of the Gelidonya staff. 53. Mycenae." Fragment of Type 2. Late Helladic
13-17. Enkomi.:" Two of Type 2, two of Type 3, and III B or C.
, one of uncertain type are reported by Catling. 54. Mycenae." Uncertain type. Sixteen fragments.
I III. Asia Mil101'
55. Mycenae." Handle of uncertain type. No evidence
for date.
18. Near Antalya." Type Is, Found m the sea; no 56--74. Cyme (Euboea}." All Type I (probably all
evidence for date. 1b). Two fragmentary. Found in the sea; no
19. Near Antalya." Type lb. Found in the sea; 110 evidence for date.
evidence for elate.
20. Probably from near Antalya." Type 1. Probably VI. West of Greece
from the sea. P. Throckmorton reports that he has 75. Makarska, Dalmatian Coast." Miniature of Type
talked to sponge divers on Kalymnos who remem- 3. L. .066. No evidence for date.
bered many more ingots from this hoard; pictures 76. Cannatello, SiCily.30 Fragment. No evidence for
drawn for him by these divers all represented dating.
ingots of Type I. 77. San d'Antioco eli Bisarcio, Oaieri, Sardinia;" Type
2e. Impressed sign (fig. 90: 16A). Dated only
IV. Crete by comparison with Aegean ingots.
21. Palaikastro." Type and stratigraphic information 78-82. Serra Ixili. Sardinia;" At least two are of
unavailable. Type 2e. Marks: see fig. 90: !I-IS. No strati-
22. Zakro." Type 1. Ingot fragment. No strati- graphic evidence for date.
graphic information. 21 Buchholz cat. no. 15; J. Haz zidakis, T'vtissos, 57, with
23. Sitias." No other information available. fig. 31.
22Buch1101z cat. nos. ]6-34, with pis. 3: 8-10, and 4: 1-6.
10 Buchholz cat. 110. 2, with pl. 3: 2 i P. Dikaios, Guide Cypr. 23 My thanks to Prof. Platen for providing me with this
u,«, 100, with pl. XXVI: 1. information shortly after the discovery of the ingots. N.
11 Buchholz cat. no. 3. with p1. 3: 3; Schaeffer, Enhonn- Platen, Archaeology 16 (1963) 273; translated as "bronze"
Atosia, 30: no. 11, with pi. LXIII: 1. 011 275.
12 Catling, Cvpr. Bronse., 268, no. 5, with pl. 49: e. 24 Buchholz cat. 110. 35, with pl. 5: 1-2; C. Seltruau, Athens,
IS Ibid., 268, nos. 6 and 7, 284, with pl. 49: f, g; Buchholz its Hist. and Coinooe, 4-5, with figs. 3 and 4.
cat. no. 5; SeE III. g'. 328. 25 G. F. Mylonas, AlA 66 (1962) 406, with pl. 121: 2;
14 Cat ling, Cvpr. Bronce., 269, no. 2, with pl. 52: b: 37; Archaeology 15 (1962) 54.
set: lIT, fig. 374. 26 Wace, BSA 48 (1953) 7, with pl. 2(a): F. Stubbiugs,
15 Buchholz, PZ 37 (1959) 19-20, with fig. 9; Mino-ica., 105- BSA 49 (1954) 292,295; Catling, C:ypr. Bronce., 269, 295.
106, with figs. 4a and 4b. Catling, Cvpr, Bronee., 268, miniature ~j Carling, C)lpr. Bronoe., 269; in Nauplion 1v[uSet1ll1.
no. 1. ~8 Buchholz cat. nos. 38-56, with pl. 5: 3-4; Przeworski,
16 Catling, Cypr. Bronze., 269. .Metall. Anat., 92, with n. 6.
17 Buchholz cat. 110.6, with pI. 3: 5; S. Przeworski, Metalf. 29 Buchholz cat. no. 57, \yith pI. 5: 5. See, however, Catling,
Anat., 92, 103, with pI. XIII: 2. C-ypr. B'l"on::;e.)269, n. 3.
18 Buchholz cat. no. 7. gO Buchholz cat. no. 58.
19 Buchholz cat. no. 8, with pI. 3: 6: G. M. A. Richter, 31 Giovanni Lilliu, ArchCI 10 (1958) 192, with pis. 63-64.
Grl~. Etrusc. and Rom. Bronzes, 456--457, no. 1810, with fig. 32 Buchholz cat. 110S. 59-63. Lord William Taylour, .i11yc.
on 457. Pottery in Ital)l, 1i6, corrects common spelling of Ilixi to
20 See Buchholz catalogue for additional scanty references. Ixili.
62 CAPE GEL! DONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

FIG. 62. Tomb 119 at Thebes (after Wresainski ):

VII. Egypt
83-86. Four model ingots fr0111 foundation deposits at
Thebes (see append. 8). Type 2. One with in-
cised cartouche, Late thirteenth or early twelfth
century B.C.

VII. Other
87-106. Twenty other ingots, from Aigina, Athens,
and Sardinia have been reported, but almost
nothing is known of these;"
All of the ingots which have been analyzed are copper,
with the exceptions of numbers 20 and 75, both pur-
chased from dealers and both of which are bronze. FIG. 63. Tomb of Puyemre at Thebes (after Davies,
Representational evidence is also important for the courtesy Metropolitan Museum of Art).
study of the ingots. 1\105t general works on Aegean
prehistory, when mentioning oxhide ingots, refer to the 1. EGYPTIAN PAINTINGS AND RELIEFS SHOWING OX-
scenes from the Tomb of Rekh-mi-re' at Thebes. This HIDE INGOTS 35
seems to account for the belief that the ingots were from
Keftiu, or are Mycenean in character. Such misleading Time of H atsheps"t and/or Tuthanosis III (1490-
conclusions result from the lack of a thorough study of 1436 B.C.):
Egyptian representations. The following list, which in-
1. Tomb 119 (name lost) at Thebes.
cludes representations outside of Egypt, does not pre-
One scene of Syrians and men of Keftiu bringing
tend to be complete, but it is the fullest compilation to
products including metal ingots, and another of
date.84
Syrian vases and metal ingots." A bearded man
33 Buchholz cat. nos. 36, 37, 64-81.
34 It was important, in looking for representations of ingots, have led me to question so-called ingots that lack some metal-
to consider carefully the context of each scene. Of the many lurgica l context, such as those OIl the seals below.
pictures of food (loaves?) which exactly resemble oxbide and 35 Dates of rulers are from "V. C. Hayes, The Scepter of
bun ingots, especially when the original colors are not repro- Egypt II, xv. The numbers of the Tbeban tombs are the
duced or mentioned in publication, I give three examples: Nina official number-s of the Antiquities Service; more complete
de G. Davies and A. H. Gardiner, Tomb of Anienenitiet, pl. bibliographies for the Theban tombs may be found under these
VIII; N. de G. Davies, The Rock Tombs of Deir el Gebrinei numbers in Porter, Moss and Burney, Topographical Biblioq. I
II, pl. XX; E. Naville, The Xlth Dyn. Temple at Deir el-Bahari (2nd ed.}, from which I have taken the rulers under whom the
I, pl. XX. Other objects, including tambourines, also resemble tombs were painted.
ingots: H. Hickmann, ASAE 51 (1951) 317 ff. Such objects so Porter, Moss, and Burney, op, cit. (supra, 11. 35) 234.
 VOL. 57, 1'1'. 8, 1967] BASS: THE INGOTS 63

FIG. 64. Tomb of Useramon (from Davies, courtesy

Metropolitan Museum of Art) •
•

'- X ~ FIG. 66. Relief of Tuthmosis III at Karnak (from Wresziuski}.

FIG.
DOl
65. Tomb of Useramon (after Vercoutter).

may be seen carrying an ingot of Type Ib (fig.

62)."
2. Tomb of Puyemre at Thebes (no. 39).
Tribute containing copper (red) and lead or tin
(blue) ingots of Type 1, from men of Retnu
(Syria) (fig. 63)."
3, Tomb of Useramon at Thebes (no. 131).
Three registers depict "Reception of the spoil FIG. 67. Tomb of Rekh-mi-re' at Thebes (from Davies,
which the might of His Majesty brought from the courtesy Metropolitan Museum of Art).
northern countries, the confines of Asia, and the
islands in the midst of the sea .... " 39 (Vercout- silver?), and above this are oval ingots of v...hite
tel' translates: cc ••• from the northern countries of metal (fig. 65)."
the confines of Asia.") 40 Those shown bearing 4. Relief of Tuthmosis III at Karnak.
raw metal, including copper (light red) ingots of An ingot of Type 1b, identified by Wreszinski as
Type 1b (fig. 64) are almost certainly those meant silver, is included in the booty which the king
by the people {rom "the confines of Asia," 41 but presents to Amon front his Syrian campaigns (fig.
whether or not the islanders in the topmost register 66) ."
also carried ingots is not known because of the
poor state of preservation. Time of Tutlnnosis III or Amenhotep II (1469-
Asiatic tribute is also shown after being laid 1411 B.C.) "
down by the bearers. A stack of ingots of Type la 5. Tomb of Rekh-mi-re' at Thebes (no. 100).
includes both copper and white metal (lead, tin, or Copper ingots of Type 1b are brought by "the
--
3; Wresainski, Atlas I, 340; Buchholz, PZ 37 (1959) pl. chiefs of Keftiuland (and) the islands which are
1: 3-4. within the Great Sea," in the second register of the
eelvorman de G. Davies, Tomb of P'lIyellrre I, 80-81, with
south side of the west wall of the tomb (fig. 67) ,"
pl. XXXI. ---
39 N. de G. Davies, BMM A 21, 2 (March, 1926) 44, with 42 Vercoutter, op cit. (supra, 11. 40) 366, document 501, with
fig. 5 on 48. pl. LXV.
HI J. Vercoutter, L'jJ;g.vpte et Ie monde egeen p1·ehellenique, 4.3 Wreszinski, At/as II, 3Ja, and 3Jb: object 134.
130, U Norman de G. Davies, The Tomb of Rekh-m,i-reC I, 20;
41N. de G. Davies, or. cit. (supra, n. 39) 46. II, pls. XIX, XX. Vercoutter, op, cit. (sup'ra, n. 40) 57, insists
64 CAPE GELlDONYA: A BRONZE AGE SJ-l]PWRECK [TKANS. A~lER. I'l-JlL. soc.

FIG. 72. Tomb of Penljet at Thebes (from Wreszinski}.

FIG. 68. Tomb of Rekh-mi-re" (after Davies, courtesy
Metropolitan Museum of Art).
•

~fJC1FlF:~.
;~"~7"~:
I ql~"
\ u'··('~v-
A1'~
-,'t~I!'I~~ ...
I

.-r-qf ~.:~~:l.T~
t! •
~~~~rJ
..... iiiii.l ..c-:. -:.:::1 =-=-"~=:::. ~
'iI.\'-.:
~
FIG. 69. Tomb of Rekh-mi-re" (after Davies, courtesy
Metropolitan Museum of Art).

FJG. 73. Relief of Amenhotep 11 at Karnak (after Chevrier).

FIG. 70. Casting doors; from Tomb of Rekh-mi-re" (after

Davies, courtesy Metropolitan Museum of Art). while in the fourth register similar ingots are
brought by "the chiefs of Retnu and all the lands
of Further Asia" (fig. 68)." Each line of tribute-
bearers approaches a scribe who notes the booty
----r
~ ~\l.~~
Til:" =-
c
r~~lt~~~.
--. ~ti\ ~!
... ~>:
<:;.} ift ~ ~.} ;.,g;;..' .\
1I1~:
which is piled before him, and iri each case ingots
have been laid down already. In the pile made
r"") • ) ~~~ .... _~_
....:__ win _1,l~A-':.-=- by the men of Keftiu are seen four whitish ingots
of Type la (fig. 69)!' These ingots are slightly
---7,]~/I more gray than the brick-shaped white ingots
which are identified as "silver" above; 47 Vercout-
ter has suggested that these are possibly of lead
or electrum," and I would add the possibility of tin.
On the east half of the south wall is a scene of
Egyptian bronze-workers casting a door. Ingots
of Type Ib are brought to the fire by native porters
who are "Bringing Asiatic copper which His

~~~if)l
that "and" must be maintained in order that the grammar be
correct; island may not modi 1- y Kef tiu.
4;, Davies, Tomb of Rekh-mi-rr' I, 27; II, pls. XXII, XXIII.
46 Davies, Pa.intings [roni T. of Rekh-'mi-re', pl. n.
471bt"d.; T. of Rekh-nti-re", I, 21, identifies the ingots of
FTG. 7]. Bringing copper to metal workers seen in fig. 70 Type la also as silver.
(f r0111Davies, courtesy Metropolitan Museum of Art). 48 Vercoutter, op, cit, (supra, 11. 40) 365, document 492.
 VOL. 57, PT. 8, 1967J BASS: THE INGOTS 65
Majesty carried off from his victory in the land
of Retnu ... " (figs. 70 and 71)."
6. Tomb of Penhet at Thebes (no. 239).
Ingots of Type 2 carried by Syrians as tribute
• (fig. 72)."

Time of Amenliotep II (1436-1411 B.C.):

7. Relief of Amenhotep II at Karnak.
The king, riding in a chariot, has just pierced a
copper ingot of Type 2c (or lb ?) with five arrows '-"'"', -
(fig. 73). The accompanying text leaves 110 doubt
as to the material of the target: "The king, great
in strength, who displayed his dexterity before his
\~J
army, the mighty bowman who shoots at a plate FIG. 74. Tomb of Nebaruun at Thebes {from Save-
Soderbergh, courtesy Griffith Institute).
of copper, cleaves it as one does a clump of papyrus .
• 1-Ie disdains all wood, as befits his strength." "This
is the great plate of mined copper which His
Majesty shot, three fingers in thickness. The hero
pierced it with many shafts, three handbreadths of
them standing out at the back of the plate." 51
8. Tomb of Nebamiin at Thebes (no. 17).
Nebamun is visited by a rich Syrian, who has just
arrived by ship (fig. 50). Behind the visitor, bring-
ing gifts or payment for services (Nebamun was
a doctor) are two porters; one carries a copper
ingot of Type Ib which is almost rectangular (fig.
74) ."

Time of Tuthsnosis IV (1411-1397 B. C.): FIG. 75. Tomb of Amen-em-opet at Thebes (from Davies,
9. Tomb of Arnen-em-opet at Thebes (no. 276). courtesy Metropolitan Museum of Art).
Pink copper ingots of Type lb are brought by
Syrians, who also bring lead or tin (?) ingots in
a different shape (fig. 75); 53 this is, however,
called "Aegean tribute" by Vercoutter.:"
.~~ E:l-
10. Tomb of Hepu at Thebes (no. 66).
Egyptian bronze-workers seen with copper (reel
and pink) ingots of Type Ib (fig. 76)."

Tillie of Arnenkotep III 01' Ainenhotep IV (Akhena- ~~~_. ~

ten) (1397-1353 B.C.): Q .~c... _~~--=
FIG. 76. Tomb of Hepu at Thebes (f rom Coghlan,
11. The Tomb of Nebanmn and Ipuky at Thebes (no.
by Davies).
181) .
Scene of Egyptian metal-workers with copper
ingot, which is damaged in the painting but is
49 Davies, Paintings from T. of Rchh-mi-rc', pI. XXIl1;
T. of Rekh-mi-rc( I, 53-54; II, pIs. LII, LIII.
eo Wreszinski, Atlas I, pl , 373.
51 N. de G. Davies, BMMA 30,2 (1935) 51, 'with 49: fig. 4;
H. Chevrier, ASAE 28 (1928) 126, with fig. 5.
52 T. Save-Soderbergh, Four 18th D::.m. Tombs, 25-26, with
pl. XXIII; Wreszinski, At/as I, 115; Buchholz, PZ 37 (1959)
pl. 2: I.
sa N. de G. Davies, BMAIA 27, section 2 (March, 1932) 62,
with fig. 13.
54 Vercoutter, oo. cit. (supra, n. 40) 364-365, documents 497-
500, with pl. LXV.
ss H. H. Coghlan, Notes OJI the Prehist. l1fetallurgy of Cop-
per and Bronze, 68-69, with fig. 10; Wresaiuski, Atlas I, 226 FIG. 77. Tomb of Nebamun and Ipuky at Thebes (after
and 229, with inaccurate drawing on 229. Davies, courtesy Metropolitan Museum of Art).
66 CAPE GELIDONYA: A BRONZE AGE SH I PWRECK [TRANS. AMEll.. PHIL. SOC.

FIG. 80. Tomb of Huya at el Amarna (after Davies).

FTC. 78. Tomb of Meryra I at el Arnarua ; SCCllC

of Royal Storehouse (from Davies).

,
•
•
FIG. 81. Tomb of Huya (after Davies).

are Type Ib, while the longer could be considered

ingots of Type 2 (fig. 78)."
FIG. 79. Tomb of Meryra II at el Amama (after Davies). 13. Tomb of Meryra II at el Ama-na.
Ingots of Type 2 are shown as tribute brought by
probably of Type 2a-b, with rectangular white or Syrians (fig. 79). An inscription from the tomb
blue ingot below (fIg. 77).5G The latter has been of H uya (infra., no. 14) tells that this scene repre-
identified as lead 57 or tin. 58 sents tribute being brought from places including
Syria and the islands of the sea, but here we surely
Time of Amenhotep IV (Akhellaten) (.1370-1353 see Syrians."
B.C,) : 14. Tomb of Huya at el Amarna.
On the west wall we see the usual lines of
12. Tomb of Meryra I at el Amarna. foreigners bearing tribute. From Syria are brought
A scene of the Royal Granary and Storehouse copper (red) ingots of Type 1 (fig. 80). These
shows stacked ingots of several sizes; the shortest are probably all Ib, but one may be 1n.01 The
--- ----'
56 Norman de G. Davies, The T()IIl!J of Two Scnlntors, 63, nnN. de G. Davies, The Tomb of Mcrvra, pl. XXXI;
with pl. XI; N. de G. Davies, BMJ.l1A 15, section 2 (Dec., Buchholz, PZ 37 (1959) 15, with fig. 7.
1920) 39, fig. 10; Coghlan, op, cit. (supra. n. 55) 68-69, with 00 N. de G. Davies, The Tombs of Pcnehcsy and Meryra II,
fig. 9. The rectangular: ingot is reproduced as bright blue in 38, with pis. xxxvn and XXXIX.
Virey, Sept Toinbeavx thebaines. 61 N. de G. Davies, The Tombs of Hwya alld Ahmes, 11
r,r Davies, T. of Twa Sculptors, 63. (where ingots are identified as sacks), and pl. XIV. The
58 G. A. Wainwright, Antiqltity 17 (1943) 96. inscription states that tribute is brought from Syria, Kush
..,
l VOL. 57, PT. 8, 1967J BASS: THE INGOTS 67

FIG. 83. Relief or Ramesses III at Medinet Habu (from

• Medinet Habw V, courtesy the Oriental Institute) .

.~
•

FIG. 82. Tomb of Huy at Thebes (after Davies, courtesy

Metropolitan Museum of Art).

north wall of the tomb shows store chambers con-

•
•
taining, among other things, red metal ingots (fig.
81) ,62 These are of extreme interest, for two are
excellent examples of Type 3. Below, however,
are what seem to be ingots of Type 2, but one of
these has a protrusion which could be the neck
FIG. 84. Relief of Ramesses III at Medinet Habu (after
of a beast; whether or not this represents a dried
Medniet Habn V, courtesy the Oriental Institute).
skin we will discuss below.
2a, with four ingots in each stack, to Amon-Re
Time of Abhenaren or Tnt-ankh-amen (1370-1343 (fig. 83). A second scene of the same subject
B.C). shows ingots of Type 2 in three stacks (fig. 84),
15. Tomb of Huy at Thebes (no. 40). marked with the hieroglyphs for silver, copper,
The chieftains of Retnu (Syria) bring tribute to and lead." At least one scene, and possibly both,
the king (fig. 82). The copper ingot (reported as was copied directly from the thirteenth-century
deep red or black) is closest to Buchholz's descrip- Ra-nesseum.?"
tion of Type 3 (almost straight sides), yet he calls
it an ingot of Type 2.63 II. POSSIBLE REPRESENTATIONS OF OX HIDE INGOTS
ON SEALS

TitHe of Romesses IT! (1192-1160 B.C.).- Crete

16. Relief of Ramesses III at Medinet Habu. 17. A flat, rectangular bead-seal of green steatite, Iror»
A relief shows Ramesses III offering a hoard of Central Crete, presents a possible ingot of Type 1a
treasures, including four stacks of ingots of Type (fig.85)." Actually, the object looks far less like
an ingot in the photograph than in the drawing,
(Nubia), the western and eastern countries, and the islands in
and I consider Evans' tentative identification as an
the middle of the sea. The ingots are carried by the Syrians,
according to Davies (pp. 9,11). G4 ifedinci Habn V, pl. 328.
62 iu«, pl. XVI. f'>5Jbid., pl. 333; Buchholz, PZ 37 (1959) 24, fig. 11: d,
as Nina Davies and A. H. Gardiner, The Tomb of J-Iuy, 29, where they appear as ingots of Type 1.
with pl. XIX; Buchholz, PZ 37 (1959) pl. 2: 2. Of'> Arthur Evans, Scripto Minoa I, 151: P. 13(b), with pl. 1.
68 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRAl\'S. AMER. PHIL. SOC.

r- o 0 ()

o()Of,(
..' 0
0
() 00 00

FIG. 85. Bead from Crete (after Evans, Scripta Milwa I).

Frc.86. Seals from Cyprus (after Ward}. •

FIG. 88. Bronze stand from Cyprus (after Schaeffer,

Enkomi-Alasia) ,

and 3, actual examples of which have not been

found on Crete.
Two of rhe tablets indicate weights of ingots in
an unknown standard, possibly a talent. Oa 730
FIG. 87. Tablets from Knossos (after Buchholz, PZ 37).
states that 60 ingots weigh 52 2/30 units, and Oa
733 states that 10 ingots weigh either 6 or 8 units.
ingot to be improbable. Of interest, however, is
On certain of the Sc tablets, corselets have been
the fact that an ox is also depicted on the seal,
erased and replaced by ingots, which led Evans to
making a combination found on the following
believe that an ingot contained enough copper (or
examples from Cyprus.
bronze?) to make a corselet.?" Ingots appear on
the following tablets: (Evans' numbers) 70 22Sb Nm
Cyprus
01,730 OJ 01 (LI), 733 OJ 11 (XXXIII), 734a
18. A number of cylinder seals from Cyprus also show Oh 01 (LI), 734b Oh 01, 247 Nj 81, 248 Nj 86
possible oxhide ingots of Type 1, with ox's heads (XXIV), 249 Nj 85 (XXIX), 246 Nj 82
above several (fig. 86).67 O. Masson, who dates (XXIX).
one example from Enkomi to 1550-1400 B.C., be-
lieves these to be ingots, but I find this identifica- IV. OTHER
tion questionable.?"
28. A bronze stand fr0111 Kourion on Cyprus, now in
the British Museum, contains four male figures in
III. REPRESENTATIONS OF Ox HIDE INGOTS ON LINEAR
its openwork sides. One of these carries an ingot
B TABLETS (fig. 87) of Type 2c (fig. 88). According to Harelen,
19-27. Ingots of Types Ia, Za-b, and 2c seem to be shown lithe figures are in Asiatic garb and wear shoes
on tablets from Knossos, but such crude pictures with turned-up toes that suggest Hittite prototypes
should probably 110t be used to identify types. Of and occur frequently on north Syrian sculptures
interest, however, is the fact that most of the ingots down to the ninth century B.C., if not later. This
represented have the "legs" or handles of Types 2 piece was probably made in Cyprus, but is linked
artistically with north Syria rather than Phoeni-
67 W. H. Ward, The Seal Cytind ers of Westen, Asia, 346-
349. 00 Arthur Evans, Pal. of linn. IV, 805-806.
68 O. Masson, BCH 81 (1957) 7-8, with fig. 1. 70 Arthur Evans, Scripta Minoa II.
 VOL. 57, PT. 8, 1967] BASS; THE INGOTS 69
cia." 71 S. Casson says that it is of "pure My- longer and became more pronounced, he believed, dur-
cenean style," but with Babylonian and Egyptian ing the evolution of the shape. He advanced the theory,
influence." while Watzinger sees a Phoenician therefore, that the legs were developed merely for ease
origin." It was probably made in the twelfth or of porterage, and the resemblance to a skin was purely
eleventh century B.C. iortu itous. 7 8
29. C. F. A. Schaeffer, Antiq·/tity 39 (1965): p. 56, While this is indeed the case, the chronological divi-
reports the discovery at Enkomi on Cyprus of a sion of types cannot be clearly defined. Ingots dated in
bronze god, .35 high, "standing barefoot upon an stratified excavations are, as we have seen, the rare
ingot of Cypriot type in the form of an ox-hide." exceptions, and evidence for dating from such actual
PL. xvib shows the statue with horned helmet, examples must be considered too scanty for drawing
shield, and spear. Schaeffer, AO" 21 (1966) 59 conclusions. We must, therefore, turn to representa-
ff., shows ingot to be of Type 2. The context of tions.
the find was Cypriot Iron Age I (LC III), or Ingots of Type 2, elated by Buchholz to around 1400
early twelfth century. B.C., are seen as early as the mid-fifteenth century in the
tomb of I'enhet;" and possibly as late as the twelfth
Using the evidence gathered above, we may begin to
• seek the purpose and origin of what have been called
century on the relief of Ramesses III (but see below,
p. 164). Ingots of Type 3, dated to around 1200 B.C.,
"four-tongued," "double-axe," "pillow-shaped," and
are twice depicted in the fourteenth century, in the
"oxhide" ingots. All but the last of these epithets may
tombs of Huya and Huy. Ingots of Type 1, elated to
be easily rejected. Terms based on shape, as Buchholz
around 1500 B.C. by Buchholz, are seen in two four-
has pointed out, do not describe all the variations of
teenth-century tombs, of Meryra I and Huya, at el
form found (fig. 55)."
Amarna. The most conclusive evidence that the various
The term "oxhide" is widely used by archaeologists
types were used contemporaneously is found in the latter
and, although it has not been universally accepted, its
tomb, 'where Types 1, 3, and probably 2 are seen
rejection presents some difficulty. The term arose be-
together.
cause of the simi1arity of the ingot shape to that of a
We must conclude that the three major types of
dried skin: each ingot, it was thought, was
ingots cannot always be separated chronologically, but
cast in the shape of an ox-hide, dried and stretched. The we may accept the appearance of ingots without legs as
obverse roughly represents a hairy hide, the reverse the slightly earlier than the appearance of those with legs.
raw inside of a skin with its edges curling inward. Neck Their continuation in use, after the development of
and tail are not shown because for practical purposes they
are not needed, but the roughened surface of the obverse handles, has no apparent explanation other than con-
which represents the hairy hide of the skin is even indicated servative local preference in some areas; those of us 'who
on Egyptian wall-paintings .... 75 often moved the oxhide ingots from place to place in
our camp near Gelidonya are satisfied that the shape
Furthermore, because the ox was a common unit of
with legs is much more easily handled than the block
value in antiquity, it was believed that each ingot
shape of modern ingots.
equaled the price of one ox in a pre-monetary form of
Attempts have also been made to show that other
currency."
resemblances to cow skins were merely the fortuitous
The oxhide theory has been attacked by many, in-
but quite natural results of casting. While again this
cluding the author," and the arguments against it often
seem overwhelming. 1-:1. G. Buchholz, in the most thor- 78 Buchholz, P Z 37 (1959) 2, 4; the additional suggestion
ough study of all ingots known before the Gelidonya (p. 2), that the shape was developed for long-distance commerce
excavation, has shown that the shape evolved over a because round ingots would be impractical for lading on shi ps,
is disproved by the discovery of many round bun ingots at
number of centuries. The earliest ingots, dated by him Celidonya. The opinion that the "legs" were only handles
from representations in tomb-paintings and from actual was previously held by H. R. Hall, Aeg. Archae .. 67; C.
examples in stratified excavations, seemed to have no Schaeffer, Enleoini-Atasia, 33; Arthur Evans, Pal. of 111';11.II,
"legs." If the casters had wished to imitate cow skins, 625; IV, 653.
,9 The date of the tomb of Penhet is important because it is
it seemed more natural for the earliest examples to have
the only fifteenth-century tomb listed with a certain ingot of
been the most realistic, while later shapes would have Type 2. Porter, Moss and Burney, Top. Biblio. I, 330, give
hecome more debased. Conversely, the legs grew the time as of Tuthmosis IV to Amenhotep II, which is
obviously impossible. This must be a misprint for either
,1 D. Harden, The Phoenicians, 195,with pl. 82. Tuthmosis III to Ameuhotep II, or Tuthrnosis IV to Amenho-
72 S. Casson, Aile. CjlprHs. 128,with pl. 8. tep III; I have selected the earlier possibility on the basis of
73 Buchholz, PZ 37 (1959) 20, with n. 43 for further bib- the date of ca. 1450 B.C. given by Wreszinski. If this date is
liography, and fig. 4 on p. 12. incorrect, we would have only the evidence of the Linear B
«ru« 1-2 tablets for the existence of ingots of Type 2 before 1400 B.C.,
75C. T~Seltman,
Athens, its Hist. and Coinage, 4-5. which would be an additional point in favor of L. R. Palmer's
1, 5.
761hid., dating of the tablets from Knossos after 1400 B.C. (Palmer,
H Al A 65 (1961) 272-273. J1'[ 'yccnaew/s and J1,1
moans'),
 .....
70 CAPE CELIDONYA: A BRONZE ACE SHIPWRECK [TRANS. AMI;:R. PHIL. soc.

side of the ingot must have been in the bottom of

the mold and must find another explanation. For this
a I asked a sculptor, Robert Barnes, to make a small mold
for an oxhide ingot from a photograph of one from our
~loard. First he outlinecl the shape of the ingot deeply
111 clay. Most of the clay within this outline was then
e.ither dug out or pressed down with his hand. The par-
ticular example III the photograph was of one of the
ingots 'without the thick rim on its bottom, so there was
b
no attempt to copy that feature. The outline remained
FIG. 89a. Usual draft of Cape Gelidonya ingots. deeper than the rest of the mold, however, and would
FIG. 89b. Section through hypothetical mold shows ridges that
have produced such a rim. Inadvertantly the cause of
must have caused grooves inside ingot rims.
the rim was dis~overed, and also the reason why it
appears on. most ingots but not on all; the rim simply
seems to be true, the explanations for these features results accidentally when all of the clay in the mold is
ha ve been incorrect. 80
The rough, "hairy" side has been thought to result
not removed to the level of the outline. There is some •
evidence that the rims were desirable, however, though
from the bottom of the mold in which the ingot was not mandatory. Some of the rims are sharply de-
cast, which would necessitate that the impressed signs lineated by grooves. These grooves would have oc-
were made by reliefs placed in the molds. Actually, the curred if the caster, fearing that the outline would be
rough side was upper in the mold, and resulted from mas~lecl or lost, had not pressed down the clay near the
the reaction of the solidifying metal with the atmosphere. outline (fig. 89b). Only two explanations for inten-
Its occurrence 011 bun ingots, where there was certainly tional rims come to mind: the rim was desired to give
110 intention of representing hair, is mentioned by a b~tter edge for grasping and lifting the ingot, or it was
Gowland: "This phenomenon is technically termed desired to represent the curling under of a cowhide.
'r ising' and is seen daily in all copper works." 81 Manu- The bottom of the mold made by Barnes further re-
facturers who want smooth surfaces on their copper tained a number of concavities where it had been pressed
ingots n~v./ obtain these by partially filling the empty or pounded down. These would have caused ' in casting ,
molds with powdered charcoal. The charcoal floats to Iow mounds such as we find on 111oSt,but not all, of
the top of the molten metal, when it is poured, and forms the Gelidonya ingots.
a protective covering against the air. Agricola, writing in the sixteenth century of the
A further indication that the rough side was the modern era, describes the making of forehearths and
upper side in the mold comes fr0111the Gelidonya ingots. dipping pots to catch newly smelted metal, and the pro-
Careful inspection of the ingots revealed that in almost cedure was then the same: a mixture of charcoal dust
every case the rough surface was larger than the and earth was packed tightly together and then hollowed
smoother (fig. 89n). Only if the draft was made in this Ol~t with a knife. The concavity was then packed again
manner could the finished and hardened ingot be drawn with a rammer." A mold made in this manner would
from its mold without the destruction of the undercut produce ingots with the various features we have de-
edges of the mold. Authorities have been misled in the scribed; it seems quite certain that Bronze Age ingot
past because the usual view of the cross section of an molds were made in much the same way.
ingo~ shows the edges beveled in the opposite direction,
VVe.have s~lown .that the ingots were not necessarily
making the rough surface smaller than the smoother cast with the intention of imitating oxhides. Even more
surface." This is decidedly atypical, although it occurs conclus.ive evidence is the appearance of ingots of the
on one of the Gelidonya ingots (21), while two of the same Size and shape, but made of different metals. In
ingots (14 and 18) show no draft at all. This suggests the tomb of Rekh-mi-re' are red copper ingots stacked
that ease of drawing was a convenience rather than a with si111il~rwhite ingots of lead, electrum. or tin (Ver-
necessity, and that molds may have been made for only coutter P0111ts out that these unidentified white ingots
one casting apiece. This is further suggested by the are grayer than the brick-shaped white ingots in the
fact that no two of the Gelidonya ingots seem to have s~l11e painting wl~ich are identified by hieroglyphs as
come from the same mold. silver). In the tribute shown in the tomb of Useramon
The "curling under of the skin" has been explained may be seen similar ingots of copper and lead silver
as the natural rim left by shrinkage on top of the metal or tin again stacked together. At Medinet H'abu we
as it cooled and solidified.s" Vile have shown that this h~ve representations of oxhicle ingots marked with the
eoBuchholz, MiJloica, 95; Seltman, Athens, 4, n. 2; Schaef- hieroglyphs for lead, silver, and copper. We have also
fer, Enkomi-Alasia, 31. seen that at least one actual oxhide ingot, from the Bay
81 VV. Gowland, Archoeolooia 56 (2nd series 6, 1899) 289. of Antalya, was of bronze.
82 Buchholz, PZ 37 (1959) 13, with fig. 5 (after Pigorini).
83 Seltmau, Athens, 4, n. 2; Buchholz, PZ 37 (1959) !3. 84 Agricola, De Re Mctattica (trans. Hoover), 377-378, 386.
 VOL. 57, PT. 8, 1967J BASS: THE INGOTS i1
1£ a copper ingot of a certain size and shape equaled ingots, where admitted at all;" have sometimes been
the price of an ox, it is certain that a silver or lead ingot explained as local variations of the standard in use : they
of the same size and shape would not also equal the
varied greatly in w.eig!1tfor the simple reason that copper
price of an ox.
was less easy to ship tram place to place than its portable
It would seem that the oxhide theory could be com- ~olden" equivalent. Consequently in the gre~t "copper-
pletely discarded, but one disturbing piece of evidence Island, Cyprus, where the metal was common It was need-
remains. Upon one wall of the tomb of Buya are seen ful to pay, as the price of an oX a far heavier mass of
I

cO'pper.than one would pay in Mycenae where, as compared

what must be considered red ingots of copper. Those
WIth Cyprus, the copper was scarce. For this reason the
above are of Type 3, while those below are of Type 2. Cypriot ingot in the British Museum wei shs as much as
One ingot of Type 2, if it is an ingot, has a protrusion 37 kilogrammes, while the average weight l:>ofthe "ox-hide
between its handles which resembles the neck of an talents" used by the Achaeans in Greece was only 25-:\-
kilogrammes.e? -
animal. A comparison with an actual representation of
a dried skin on an Egyptian tomb painting, in the tomb A careful study of the weights of the Gelidonya ingots,
of Panehesy;" shows that the forms are almost identical. a closed deposit forming a coherent unit, and by far the
Has either the original artist or the copier made an largest hoard yet found, has shown that they vary COl1-
• error, or is it possible that a reel skin is shown stored
with copper ingots? The latter is improbable, for metal
:ider~bly. Even after taking into consideration changes
111 weight caused by corrosion, we see that the difference
objects are usually found together in such scenes, but between the lightest and heaviest is extreme: 13 kg.
in at least one instance we do see metal tribute with There seems to be no ratio in the weight variations. A
panther skins just above." The copy of the scene in study of the ingot fragments further indicates that these
the tomb of Meryra II indicates that the painting was in did not represent fractional parts, by weight, of whole
poor condition, and parts of these "ingots" are missing. ingots-a fact noted by Wace at lVIycenae,9o and by
The shape with the "neck" is missing its other end, but Hazzidakis at Tylissos.w
another recl shape directly beneath also resembles an \1\Tehave, I believe, evidence from contemporary docu-
ingot of Type 2; in this latter example the end which ments that ingots were of no set weight and needed care-
might also have a neck is gone, but the preserved end ful weighing "when sold or inventoried. Two of the
shows no evidence of the tail we might expect to see. tablets from Knossos show oxhide ingots with scales
We must assume that the neck is an error, therefore, but and numbers: Oa 730 states that sixty ingots weighed
this scene remains the most important piece of evidence 522/30 "talents;" 00 733 states that ten ingots weighed
which might suggest that cow or ox skins were imitated six or eight "talents." 02 If we assume that the "talent"
in copper. was a fixed unit, it is quite clear that the average weights
I continue to use the term "oxhide" only because it of the ingots on the two tablets were not the same.
is so widely spread. Even though it seems to be incor- Arthur Evans suggested that the tablets equate copper
rect, it at least defines a known cody of material, as does and bronze talents with a gold unit." for he believed
the term "Minyan" ware. To suggest yet another name the ingots to have weighed one talent each. It seems
at this time would only lead to more confusion. more reasonable to conclude that the tablets tell only
That the ingots were not cast to imitate oxhicles does how many "talents" a number of copper ingots of 1/.0
not negate the possibility that they were a form of cur- standard 'Uwight weighed.
rency. Obviously they were used for trade and, pos- From the evidence concerning shape ancl weight, we
sibly, tribute. They were also stored in Egyptian treas- can be certain that the ingots "were not a form of cur-
rency. They were merely pieces of raw copper, cast
uries, as we have seen, but so was grain "which was not
in shapes convenient for carrying, which would be
currency. J. G. D. Clark poses the general problem: handed over to smiths to cast into objects of copper and
How far trade was conducted in prehistoric Europe on a bronze. The Gelidonya wreck shows that pieces were
basis of pure barter, under which commodities were ex- broken off indiscriminately as they were needed. T. L.
changed for direct consumption only, and how far and Comparette, in presenting his argument that the large
when the use of one particular object of barter became so
regularized as to become a medium of exchange are ques- bronze bars usually known as aes signatwn were not
tions to which satisfactory answers can hardly be given.et examples of the first Roman coinage, has remarked well
88 J. D. S. Pcndlebury, Archae. of Crete, 212, says that the
Currency, to me, infers some standard of value which, weights of the ingots are practically uniform. Ventris and
in the case of raw metal of the same quality, can be based Chadwick, Docmnents, 57, speak of average weights of ingots
only on weight. Variations in weight among different equal to a talent each: "... such a talent tends to remain
constant, due to its limiting value as the largest ingot which
can conveniently be shouldered by one man. 'I
85 N. de G. Davies, Tombs of Ponehes» and Meryra IJ, pl. 89 Seltman, Greek Coins, 7--8.
XVIII. "BSA 48 (1953) 7, with pI. 2(a).
86 Nina and N. de G. Davies, lEA 26 (1940) 135, and pl. 91 Tyl1:ssos, 57.
XXIII. 92 Ventris and Chadwick, Documents, 57.
87 Clark, Proliist. Europe, 278.
9~ A. Evans, Corolla Ninn., 361, with fig. 14.
72 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRANS. A.\tER. mu.. SOC.

all the same phenomenon. Although he speaks of a of the Gelidonya ingots are marked with signs either
period almost a millennium after the Gelidonya ship impressed into the metal while it was still hot and soft,
sank, his statements are no less applicable here. The or incised into the metal after it had cooled and
bars, he says, are simply commercial ingots: solidified. Although we have proved that the majority
of the impressed signs were not made, as previously be-
... the fact that 1110St of the bars 11ave been found 1n a
fragmentary state has been considered by many scholars
lieved, by relief-stamps lying in the bottom of the molds
as strong proof that they were issued for use as money; for (only the E8 signs, on the smoother surfaces, were made
it is held that, whenever it was desired to obtain a sum less in this way), there can still be little doubt that they were
than the value of an entire bar, the practice was to break made at the place of manufacture. The incised signs
off from it a piece of the necessary weight and value. The
could have been made at any time and place thereafter.
absurdity of this theory has been pointed out repeatedly and
its chief weakness has been stated above. Considering the It is probable that the primary, impressed marks were
bars as money, in any sense, the breaking of them into frag- related directly either to mining or smelting; none of the
ments would be most remarkable; but it is not difficult to bronze bun or slab ingots, which could have been pro~
understand why commercial ingots should be found in frag- duced anywhere from either copper and tin ingots or
ments; it is in fact just what an expert worker in bronze
would expect. For at the present time, with greatly im- scrap bronze, were marked. It is further probable that
proved facilities for melting bronze, including an abundance
of cheap gas, it is the very common practice to reduce the
neither the impressed nor incised marks had much, if
any significance after the ingots were sold, for these
•
ingots to fragments because the smaller pieces pack better marks would have been lost as soon as the ingots were
in the crucible and at the same time the surface directly
exposed to the heat is thereby greatly increased.?"
broken or cut into pieces.
That the incised marks were made at almost the same
A number of scenes [rom Egyptian tomb paintings time as the primary marks is suggested by their occur-
show oxhide ingots about to be melted down and poured rence 011 some of the complete and broken ingots of the
into molds (figs. 76 and 77). Such an end was cer- Gelidonya ship, a merchantman undoubtedly not far
tainly the sale function of every ingot. along its proposed westward voyage. This negates the
\IVe have remaining the question of the meaning of possibility that the secondary signs were made after re-
the ingot signs. \Ale have seen that most, but not all, ceipt in another land. Nor is there any plausible reason
n r .'-/JNII1I1 52 (1918). why the merchant transporting the ingots should have

Rasia Triada Sardinia Llyc. EIlkom1

2 3 5 8 9 10 11 12 13 14 15 16 17 18 19

Ingot Primary T tV T T 1- T A
Marks Secondary
't' f \IV .j:' Q( r .h q. ..- '8 ~ or 't' I B
Linear A 'i r ih \ji T Of T T t T

Linear B
¥ \~ ~ 't' T --'& T T t r
Cypro-,Minoan &.
Cypr. Pot-Marks 'f Y I- '!' 9' T P i> r T '¥
11th Century W 6<1 I ~
NWSemitic

lApp ears later on Mesha Stone.

Cape GelldoD.Ya
20 21 22 23 24 25 26 27 28 28 30 31 32 33 34 35 36 37 38 39 40 41 42

Ingot Primary

seconder
r-r
j:.
n + V -tt-
* X
* T
• r-r :1: i¥; EB (;->
f t
-i=
F
~
t\.
A
Wark. 111 ~ B
Linear A
I f + 1\ t X T ffi ( f t T
Linear B t + $ :j: T i:IJ ( 'f t F
Cypro-M1noan &. (f,
Cypr. Pot-Marks
ill $" t 1\ ~ X of f T + T
11th Century
N'W Slllllitic
'i' ? T ";( 0 'f e
FIG. 90. Superficial resemblances of ingot marks to contemporary and slightly later scripts in the Eastern Mediterranean.
 VOL. 57, PT. 8, ]967J BASS: THE lNGOTS 73
marked them during the voyage. Primary and second-
ary marks should, therefore, be considered in the same
"script."
This script cannot be identified. Comparisons 'with
Linear A, Linear B, or Cypro-Minoan prove fruitless,
for no group of signs is uniquely related to anyone of
them (fig. 90). The suggestion that primary (im-
pressed) marks are Cypriot in origin, and secondary
(incised) marks are Cretan 05 is shown improbable by
the appearance of secondary marks in the Gelidonya FIe. 91. Restored order of stacking of ingots in ship. Ingots
ship which was undoubtedly sailing toward the Aegean. from area P (bow) at right, from area G (stern) at left.
If the signs are Cypro-Minoan, they may be equated Exact orientation of In 15 and III 39 uncertain.
with abbreviations for Cypriot place-names given in
Near Eastern and Egyptian documents: 1 si for Sullu if the ore had been mined, but with an additional mark
(Soloi) on an Assyrian cylinder seal, or EB lea for if "made from tin-stone collected on the ground after
• Kation (Kition) in the Egyptian list of probable Cypriot washing." 101 Of silver ingots he says that the Royal
cities from twelfth-century Medinet Habu.?" This is Inspector, after testing their quality, marked them with
only a game, however, which can be played equally well "the seal of the King or the Prince or the owner and
with the eleventh-century North Semitic alphabet and near the same, the amount of the weight." 102 ' ,

other names from lVIedinet Habu : f s for Salon-aski Copper ingots made today by the Kramer Ingot
Company of Chicago and Philadelphia are each stamped
(Salamis) and Sali (Soloi), IV le for Kation (Kition)
with a K; the top ingot in each batch or heat is identified
and Kerena (Kerynia), 0 (for Airnar (Marion), and with a tag after the ingots have been stacked. Ingots
+ or X t for Tamassos. specially made [or a certain customer are marked with
In seeking the meaning of the ingot marks, therefore, stripes of paint along their sides.
we cannot assume that they were phonetic. 110st are In all these cases the mark is associated with value,
simple arrangements of lines of the sort used as potter's whether giving the weigbt, or guaranteeing quality (by
and mason's marks, and some may even be found on identifying source of metal, manufacturer, or inspector).
ingots made in medieval Europe.": Metal ingots have For the Gelidonya ingots we may immediately con-
horne marks from the Bronze Age until the present, and clude that the marks did not designate weight. Ingots
a look at more recent, better understood evidence may weighing the same have different signs, while identical
provide a clue as to their purpose, whether phonetic or signs appear on ingots of varying weight. It seems also
not. on oxhide ingots. unlikely that the signs identified smelters. One would
Greek lead ingots from Laurion often were stamped suppose that the shape of Type 2a was preferred by one
with non-alphabetic trade-marks." Roman ingots bore smelter or locality and Type 2c b): another, yet identical
more information; many were inscribed with abbreviated marks occur on ingots of different types, while ingots
personal names." A copper bun ingot, found recently which are very nearly "twins" bear different marks.
in the sea, bears the inscription; M (etallum)P (ublicut1t) A second method of trying to determine whether the
NOMI(ne)PRIMULT(et)SILONIS CCXCVII PRO signs indicated specific smelters or localities has also
(clIrator)COL(oniae)ONOBENSIS. This gives the resulted in negative evidence. If the merchant ship
names of the conductores, or private lessees, of the mine had made several calls in ports along the coast, it would
{Primulus and Silo), the 'weight of the ingot, the con- seem likely that the ingots in the bottom of the hold
trol of the procurator representing the public treasury, were placed in first, with additional ingots piled on top
and the name of the place where the mine existed during later stops. A look at the lading of the ingots
(Onuba, modern Huelva, in Spain}.':" (fig. 91) reveals little. We see again that there is
Agricola, writing in the sixteenth century of our era, almost no pattern based on type, metal content (see
tells us that tin ingots were stamped by the Magistrate append. 3), or signs. The only point of interest is that
the ingots marked with EEl011 their smoother sides were
95 Buchholz, lVnl101:W 103-104.
j

961b£d., 104-105; Buchholz also rejects possibilities of read- together near the stern of the ship, although these marks
ing place-name abbreviations. S. Casson, Anc. Cjlp'rns, 130 were 011 ingots of different types, of different weights,
and 144, discusses the Medinet Habu inscription. and bearing different marks on their tops. It is un-
9, Agricola, fig. on p. 499; R. Forrer, Gerinania 16 (1932) fortunate that ingots 22 and 23, lying in the same area,
104. Most of the ingot marks are included in J. L. Myres'
list of arbitrary signs in IHS 66 (1946) 4. were so badly damaged by having been in contact with
9sDaremberg and Saglio, Dict., 1865 (Metallmn). tin that it is not possible to know whether they also bore
99 Forbes, }\if et. 'ill Alit .. 369. this mark.
100 F. Benoit, CalHa 20 (1962) 154-156, with fig. 19. See
also R. G. Collingwood, "Roman Britain," f:COII. Survey of 101 Agricola (trans. Hoover) 414.
Anc. Rome III, 38. 10~ibid., 489, with fig. on 488.
 [TRANS. A..\ll';R. l'HIL. SOC.
CAPE CELIDONYA: A BRONZE ACE SHIPWRECK
74
vVe have seen how the ingots were made and for
Too many ingots bore primary marks for these to
what purpose, but a number of questions remain. \lVho
have designated batches or heats; lack of a mark, in-
dealt with them, who manufactured them, and were
deed, might be considered a sign in itself, especially if
these necessarily the same people? The first conclusion
the signs were not phonetic. Further, analyses of the
that may be reached, by a study 01 their distribution
trace elements in the copper offered nothing to indicate
(fig. 92), is that ingots were transported almost ex-
that ingots with the same marks were more closely
clusively by ship. At the time of this writing, almost
related in their sources of metal than those with different
a third of all known ingots had been found in the sea
marks (append. 3) ; the danger of basing too much on
and others, with one exception, were found either on
such analyses, however, has been well stated by A.
islands or at mainland sites not far from the sea (this
Steinberg.':" does not include the four models from Egyptian
So far we have no positive evidence for the meaning
of the signs, yet they certainly served some rational pur- Thebes) .
The almost universal misconception that the presence
pose. The theory tentatively proposed here is that, in
of these ingots indicates Minoan, ·Mycenaean, or at least
spite of metal analyses, whose value has been shown to
Aegean trade has been mentioned, and references are too
be questionable, the primary marks identified the source
of the copper. Whether a mine, mine-owner, state
numerous to be noted. Such statements as that there •
is "no doubt that the trade of ox-hide ingots was in late
official, or mining guild was represented I will not ven-
Minoan or :Mycenaean hands," 105 are reasonable if one
ture to guess, but it is more logical that "Type-2a-ingot-
accepts that "al1 paintings show us the Keltiu hearing
maker" and "Type-2c-ingot-maker" each had some of
hide-form slabs of copper as tribute (or probably a well
Hare-merchant =1= 's" material, which they marked ac-
paid article!) to Egypt. ... J> 106 Indeed, the ingots
cordingly, than that "smelter "$ "was making ingots of have been named "Keftiu-ingots" by their most thor-
Types 2a, 2b, and 2c, while "smelter T " was making ough student. 107
almost identical ingots of the three major types. Less Such claims have little basis. I have Iound but one
common secondary marks might have been incised on instance. in the tomb of Rekh-llli-re(, where oxhide
the top ingots of stacks which were inventoried and ingots may definitely he assoicated with people from
readied for export, perhaps signifying the inspector who Keitin. Even here Syrian chiefs present similar ingots,
checked them or the merchant who was to receive them, and when the ingots are seen in the hands of Egyptian
or they may have been scratched on for identification of smiths in the same tomb, they arc identified as Asiatic
ownership by the merchant hirnsell after he had bought copper from Retnu.V" In eleven other representations
them. Both marks, as has been shown, lost all value we may be certain that the Egyptians considered the
soon thereafter, as ingots were apparently being broken ingots to be Syrian.
lip either before or during the voyage and, therefore, ] n Cyprus there may be pictures of ingots on seals,
cannot have designated destination. but these are questionable. The one sure representa-
\Ve may conclude the discussion of marks with the tion. on the bronze stand from Kourion, is carried by a
observation that they could be related to either private figure who may be identified as Syrian or Phoenician.
or state control.''" The absence of primary marks on Lastly we find ingots in Crete on Linear B tablets
the earlier ingots of Type 1 will he discussed below. and, doubtfully, all a seal. These indicate only that in-
103 Speaking on "Optical Spectrometry and 'Oriental' gots had arrived in Crete, which is known from actual
Bronzes" at the Annual Meeting of the Archaeological Institute examples. This no more suggests a Minoan trade or
of America in 1963, Mr. Steinberg pointed out that: "There is origin than their appearance in Egyptian storehouse
a certain margin of cxpcl·imental error 011 the part of the
laboratory that must be taken into account. There are also scenes suggests that the ingots were Egyptian.
certain metallurgical phenomena to consider ... it has been During the fifteenth century B.C., the only time "Vhen
shown that tin, lead, antimony and bismuth, when cooling, we may be sure that men 01 Keftiu arrived in Egypt
tend to segregate and form separate crystals in a eutectic with ingots, Keftiu boats (whether built by Keftiuan
structure: thus a sample taken with a very small drill might
completely lack these elements or, inversely, contain a pre- shipwrights, patterned after ships built in Keftiu. or
ponderance of them. Furthermore, bismuth and zinc, with built in Egypt for the Keftiu trade) were mentioned
melting-points of Zit and 419 degrees Centigrade, respectively, several times in Egyptian docurnenta.'?" This is not
might volatize or blow-off. . [The melting point a f copper the place to discuss the complex problem of the position
is 1083 C]."
0 Steinberg spoke of bronze, but these excerpts
would hold true also for copper. ers, and offer no information concerning control of the sources
104 I have not here pursued the evidence for control of metals
in the Bronze Age, which should provide a fruitful field for themselves.
i os Clark, Prchist. Europe. 258.
further study. L. R. Palmer, M)'c. alld Mill .. 100-101, 103,
ice Forbes, !lief. in: Ant.. 341.
sees in the Linear B tablets from Pylas an official responsible
107 Buchholz, PZ 37 (1959) 1.
for metal-workers, but the private nature of smiths in the 108 I can see no basis for the suggestion that "Asiatic COlmer"
Hittite world is shown by Hittite laws (176,B and 200,B) meant bronze; vVainwright, Antiquity 17 (1943) 96, and 18
translated by Goetze in Pritchard, ANET, 195, 197. Both
pieces of evidence, however, deal with those who worked with (1944) 101.
10!) Vercoutter, ap. cit., (.wpro, n. 40) 53-55.
metal which had already been obtained from miners and srnelt-
 VOL. 57, PT. 8, 1967J BASS: THE INGOTS 7S

DISTRIBUTION OF
OXHIDE INGOTS

X Single Ingotl type not

.. Hoard known
t::. Singi. Ingot I type 1
• Haard
o Single Ingotl type 2
• Hoard
a Singi. Ingotl type 3
• Hoard
II

Adapt.d fram luchhoh:, PZ 37 (1959) flg.6

FIG. 92.

of the land of. Keftiu, but if these ships were 1\1i11oa11, they took to making long trading voyages. Loaded with
or Minoan in character. we may suppose that the Egyptian and Assyrian goods, they called at various
Minoans held some part of the trade prior to 1400 B.C. places along the coast, including Argos," some years
This might also be a partial explanation of why only before the Trojan War ? 113 Are we certain that Phoeni-
actual ingots of Type 1, the earliest, have appeared on cian sea trade did not expand until "the power of the
Crete. Minoan and Mycenaean merchant venturers who had
During most of the period covered by the occurrence dominated the eastern Mediterranean sea routes for
of ingots, Mycenaean sea traffic is believed to have been many centuries was smashed by northern invaders
intense.v'? The distribution of Mycenaean pottery, like about 1200 B.C."? 1H- Perhaps the spread of oxhide
that of the ingots, indicates sea-borne commerce, which ingots will help to answer this question. Maritime
is believed to have been almost entirely in the hands of dominance may be surmised by the spread of goods, but
Mycenaean traders."!' During the same period, however, who would guess the size and extent of the modern
Syrian merchant-men arrived peacefully in "Egypt,1l2 Greek merchant fleet by the number of Greek goods
and does not Herodotus say of the Phoenicians that "as which are left at ports of call today? Middlemen leave
soon as they had penetrated into the Mediterranean and few traces.
settled in that part of the country where they are today,
113 Herod. 1.1, trans. by Aubrey de Selincourt (Penguin ed.,
110 H. Kantor, AlA 51 (1947) 17 £1., esp. p. 103. 1954).
III Harden, The Phoenicians, 157-158; \V. F. Albright, in
ttl S. Immerwahr, ArcJwcolog)1 13 (1960) 6, with map (pp.
2-3) showing distribution of Mycenaean pottery. The Bible and the Ancient .Vrar East, 342, says that Phoeni-
112 Save-Sodcrbergh, loco cit. (supra, n. 52). Norman Davies cian commercial activity did not expand in the Mediterranean
and R. O. Faulkner, lEA 33 (1947) 40 ff. until after co. 1050 B.C.
76 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

The evidence so far indicates that the ingot trade being carried by a Syrian merchantman which had 110t
was in the hands of Syrians after 1400 B.C., and at least yet reached as far west as the Aegean (see chap. XIV).
partially in their hands before that time. This does not, Finally, the study of the purely secondary marks on the
however, suggest that the ingots were of Syrian manu- earliest ingots is of little value: nearly identical symbols
Iacture.t'" The manufacture belonged to those who may be found as easily in the contemporary pseudo-
mined the copper. Although shipments or unrefined hieroglyphic script of Byblos (c--4, vv, )I) as in the
copper and copper ore fr0111 Cyprus to Ras Shamra have Linear A of Crete (cr, 'f), which shows the futil-
been suggested, based on the similarity of the slag at ity of pursuing such comparisons further in our present
Enkomi to that found at Ugarit,116 it seems certain that state of knowledge. What cannot be denied is that ore-
the copper was usually smelted and cast into ingots rich Cyprus is a likely source of the copper.
quite near the source of the raw material; the con- Catling has made a sensible argument against the
venience of transporting ingots, with their great saving theory that Cyprus was the original home of the ingots.
of weight and volume, instead of uusmelted are, is I-Ie believes that the unique oxhide form indicates a
obvious.'!" strong centralized authority which regulated copper and
Where were these copper mines and smelting places? its trade; the organization for such administrative con-
Three of the areas which yielded ingot hoards are known trol was lacking on Cyprus, he feels, during the earliest •
to have produced copper in antiquity: in Sardinia there period of the ingots.!" He goes on to suggest that
were copper mines in the Full Nuragic period, but not prior to 1400 B.C. the ingots were made in Crete, where
necessarily during the Bronze Age; 118 copper was they occur most often on land, with either imported or
mined near the site of the Cyme hoard before Strabo's local copper; after the fall of Knossos, the "vested in-
time; 119 and certainly the island of Cyprus, from which terest in the management of Mediterranean copper sup-
our very word "copper" is derived, was an important plies may be assumed to have passed to the Greeks of
mining center during the Bronze Age (attempts to chal- the mainland." 127
lenge this 120 have long since been disproved Ul). Then' The suggestion of a central control is appealing, but
is also some slight evidence that copper was mined in ingots were not necessarily produced on Crete. It will
Crete in Minoan times.!" Deposits are known in North be shown (chap. XIV) that the Gelidonya ship was
Syria/23 but whether or not these were exploited in a Syrian merchantman which was carrying a cargo of
antiquity is unknown. ingots westward before it sank. The occurrence of
Because of the rich copper ores on Cyprus, because ingots of Type 1 in the Bay of Antalya, only a few miles
miniature "votive" ingots have been found there, and from Cape Gelidonya, indicates that even the earliest
because of the primary (impressed) signs, which he ingots were transported along the same route; only the
feels are Cypro-Minoan.v" Buchholz concluded that ox- location and excavation of the ship that carried these
hide ingots probably originated on that island.v" He ingots can prove whether they were being transported
felt that the secondary (incised) marks, even those of westward or, I believe less likely, from Crete eastward.
later date, were closely related to Linear A; this sug- Crete, as Catling points out, was extraordinarily wealthy
gested that ingots were marked with secondary signs in copper and bronze objects during this early period.v"
when imported into Crete, and that they were sold in but it was a time when Asiatic and Egyptian imports
the western Mediterranean (Sardinia) by Aegean into Crete outnumbered Minoan imports into Egypt and
traders. the Syro-Palestinian coast.'?" The discovery of ingots
With the evidence at hand, most of these conclusions of Type 1 together with elephant tusks at Kato Zakro 180
were reasonable. Now, however, Buchholz himself has is reminiscent of scenes in the tomb of Rekh-mi-rc'
pointed out the miniature ingot from sixteenth-century which show the same objects being borne by men of
Palestine. Secondly, the Gelidonya ingots have revealed both Keftiu and Syria; 131 certainly the tusks were not
that primary and secondary marks existed on ingots native to Crete, and I would suggest that these two items
formed the partial cargo of a typical vessel which sailed
115 Lapis lazuli arrived in Egypt from Ke ltiu, but we can be
that same route, from Syria to Cyprus to the West, as
sure that it was not native to any of the possible locations of
Ke itiu itself; Vercoutter. oo. cit. (supra, 11. 40) 67. did the later Gelidonya ship.
'oc Schaeffer, Missions, 98-101. Ingots of Types 2 and 3 do not appear on Crete pos-
117 Clark, Prebist. Eu·rope, 258. sibly only because of that island's decline in purchasing
118 M. Guido, Sardinia, 151.
power after 1400 B.G. These ingots differ from the
119 Strabo X. 1. 9. c. 447.
120 O. Davies, BSA 30 (1930) 74 ff.
earlier ones in shape, method of marking, and disu-ibu-
121]. du Plat Taylor, JLN (24 February, 1940). .
122 R. Vi. Hutchinson, Prelust. Crete, 40, n. 2; 247, mentions 126 Carling, C'jJpr.Bronee., 266, 271.
exploitation of copper ore at Chrysokamino during the Middle 127 iu«, 271.
Minoan period. 128 Ibid.
1~8 G. A. Wainwr-ight, lEA 20(1934) 29 ff. 129 \"1. F. Albright, op, cit. (supra" 11.114) 337.
130 N. Platen, Arctweoloov 16 (1963) 273,275.
124 Buchholz, Minoicn, 103.
131 N. Davies, Tomb of Rekh-mi-reC II, pls. XX and XXIII.
rc PZ 37 (1959) 25.
 VOL. 57, l'T. 8, 1967] BASS: THE INGOTS 77
ticn. That they were still being transported by Syrians cas ion ally. It seems, therefore, that the incised marks
is proved not only by representational evidence, but also served the same function during all periods, but that the
by the Gelidonya ship itself; there is nothing which ties introduction of impressed marks 011 most of the later
the ingots to the Mycenaeans in any way. ingots signifies some administrative change which made
Most of the later ingots which have been found on them necessary.
land have been on Cyprus and Sardinia. Their appear- I suggest that the Minoans and Syrians shared the
ance on Cyprus may be attributed to their having been copper trade until 1400 B.C., but that after the fall of
produced there, but their appearance on Sardinia as the Knossos the monopoly passed into the hands of the
only recognizable import from the East is more difficult, Syrians alone. This would explain the administrative
at first, to explain. The arduous transportation of cop- change indicated by the new marking system, the occur-
per from copper-rich Cyprus to copper-rich Sardinia rence of early ingots on Crete, and the appearance of
even seems illogical. vVe admittedly have no evidence men front Keftiu bearing oxhic1e ingots in the fifteenth-
of Bronze Age mining on the latter island, but it is rea- century tomb of Rekh-mi-rev. Such a theory would not
sonable to suppose that men with the ability to find and be negated by the evidence from the Bay of Antalya,
mine copper on Cyprus, and to sail relatively great nor by the early representations of Syrians with ingots
• distances with it, would have had the ability to locate in Egypt, and it would not contradict Catling's ideas
and work the Sardinian ores; there is no alternative concerning the early connection between oxhide ingots
solution to the problem of why hoards of these ingots and Crete.
have been found in such a remote and seemingly other- An explanation for the problems raised by Minoans
wise ignored land. '1'0 the Classical Greeks the same and Semites sharing the earlier copper trade may be at
verb was used for "to explore" and "to mine"; surely hand. Cyrus Gordon has deciphered Linear A as
a large part of the ancient miner's life was spent, as it Northwest Semitic; 133 I am not linguistically competent
is today, exploring and prospecting in the farthest to judge the validity of his decipherment, which has not
corners of the map. been universally accepted, but the theories which would
The two islands had still more in COJ11mon. Both follow that acceptance fit well with the picture we have
became prime targets of later Phoenician colonization. drawn: "Semitic Minoans," ingot trade and all, were
Blakeway has shown how Greek colonization followed subdued on Crete by Mycenaean Greeks about 1400
earlier Greek trade west of Greece,t32 and we must sup- B.C.; ancient statements about the Minoan thalassocracy
pose that Phoenician settlements also folJowed patterns and very early Phoenician traders are more comple-
set down by earlier, Levantine trading posts. mentary than contradictory; and the third-century "con-
Finally we are able to see the central control that fusion" of Keftiu as Phoenicia 134 no longer necessitates
Catling has suggested as being necessary for the unique our either ignoring the translation or attempting to
oxhide shape. The ingots on both Cyprus and Sardinia move Keftiu from Crete to the Syrian coast.v"
were locally made, perhaps by native labor to fit Syrian Before ending the discussion of the oxhide ingots,
specifications, more probably under direct Syrian super- some mention must be made of Alasia, the great ex-
vision. Indeed, oxhide ingots could have been produced porter of copper. My belief that Syria controlled the
wherever copper was discovered and mined, and we copper trade offers one more objection to the identifica-
need no longer look for their "home." Because ingots tion of that land as Cyprus. The argument for Cyprus
were made to be melted down and cast soon after reach- seems most circular. Cyprus has been identified as Asy
ing their destinations, it is logical that they would be and Alasia because of the large amounts of copper ex-
found mostly in the areas where they were made or, in ported from that country (or countries), but one of the
the case of the underwater sites, where they were lost in major reasons for believing that this copper hailed di-
transport. This is, indeed, the case. They are found rectly from Cyprus is the evidence of documents telling
not at all in Egypt (except as models for foundation of the export! Wainwright's frequently discredited view
deposits) which country only imported them, and they that Alasia and Asy were in North Syria, even without
are found only rarely in Greece.
This brings us to the serious problem of Crete, for a '''lNES 21 (1962) 207-214; Grientolia 32 (1963) 292-297.
134 Vercoutter, oo. cit. (supra, n. 40) 100-101; G. A. Wain-
large number of ingots, all of Type 1, have been found
wright, AlA 56 (1952) 209.
on that island. V/ ere it not for the ingots in the Bay of 13~ Vercoutter, op, cit. (supra, 11. 40) 14----15,
summarizes the
Antalya, Catling's thesis that the earlier ingots were views of those who do not accept Crete as Keftiu: Wain-
made in Crete would be quite strong. The absence of weight, lEA 17 (1931) 26-43, and IHS 51 (1931) pre ler s
primary marks on the ingots of Type 1 also needs some Cilicia; Schaeffer, Ugaril.ica I, 34-35, proposes the Aegean-
-Mycenaean colonies at Ugarit and elsewhere on the Syrian
explanation. After the appearance of impressed marks coast; Furumark, OimsArclt 6 (1950) 243-244, eastern Cilicia;
on the later ingots of Types 2 and 3, the majority rather L. Christophe, RevEgyptol 6 (1951) 113, sees a country of
than the minority of the ingots became marked, while coastal Cilicia and the Syrian coast north of Ras Shamra.
the rare secondary marks continued to be used only oc- C. Gordon's view that Semitic Minoans were included among
those known as "Phoenicians" is explained in his Before the
'so A. Blakeway, BSA 33 (1933) 202. Bib/e.
78 CAPE GELTDONYA: A BRONZE ACE Sl-TIPWRECK [TRANS. AMER. PHIL. SOC.

the evidence of the oxhide ingots and the additional evi- was indeed the case, for only this and the slab ingot in
deuce of the Syrian ship at Gelic1onya, has seemed to me the table contained large proportions of tin. A qualita-
far more sensible than the views of his opposition.'?" tive analysis 141 revealed that the bun ingot was true
Catling has recently summarized the issues involved, bronze;':" with 87.28 per cent copper and 7.077 per cent
and also concludes that Alasia was probably on the tin.>"
mainland, possibly in North Syria.':" No marks were found on better preserved pieces,
The discovery of tablet KBo XII 38 at Bogazkoy aas which suggests that these ingots were all unmarked.
has been called the final answer to the problem by those In the following catalogue, figures represent maximum
favoring the Cyprus-Alasia relationship.':" It is said dimensions:
that "ships from Alasia in the midst of the sea" are men-
BI 1. Fig. 93, D . .20, Th . .035. Weight 3.35 kg. Com-
tioned, proving that Alasia was an island. This is not
plete. Convex side almost flat.
necessarily so. The passage in question uses the phrase BI 2. D .. 19, Th .. 04. Weight 4.5 kg. Complete. One
"in the midst of the sea" twice. The second usage can side very convex. Slight projecting Jug on one side.
refer only to where fighting took place, and the first can BI 3. Fig. 93. D . .195, Th .. 026. \i\t'eight 1.35 kg. Com-
refer as easily to the battles as to Alasia. That ships plete. One side cut straight across.
from Alasia sailed into the midst of the sea three times
BI 4.
BI 5.
D .. 18, Th .. 033. Weight 2.4 kg. Complete.
Fig. 93. D . .225, Th .. 054. Weight 5.5 kg. Com-
•
to do battle proves only that Alasia had a navy, which plete. Excrescence on one edge.
was already known.':" The possession of a navy no BI 6. D . .19, Th .. 03. Weight 1.45 kg. Complete. Ir-
more implies that Alasia was insular than that it was regular ridge around flat side, probably caused by
metal settling on edge of mold.
merely coastal. BI 7. Fig. 93. D .. 26, Th .. 038. Weight 3.95 kg. Com-
plete. Small excrescence all one edge.
II. "BU N" INGOTS BI S. Fig. 93. D . .199, Th .. 041. Weight 3 kg. Com-
plete. Large number of bubbles on convex side.
Found among the oxhide ingots were twelve complete BI 9. D . .205, Th .. 036. Weight 3 kg. Complete. One
and eight almost complete bun ingots, nine broken half side cut straight across.
ingots, and numerous smaller fragments. The whole BI 10. Fig. 93. D . .24, Th .. 04. Weight 2.7 kg. Com-
plete. One side cut across.
ingots were sometimes, and possibly always, stacked, BI 11. D . .215, Th .. 063. Weight 3 kg. Complete.
but fragments were probably carried 011 board the ship BI 12. D . .165, Th .. 045. Weight 2.5 kg. Complete.
in baskets (see chap. IV). BI 13. D .. 22, Weight 4.5 kg. Badly decayed and en-
Each bun ingot is a round cake, flat or slightly con- crusted.
BI 14. D . .195, Th .. 031. Pres. weight 2 kg. Edge missing
cave on one side and convex on the other. The flat in two places. Convex side more bubbly; other side
or concave side is normally very bubbly, which indicates partially concave and partially convex, with traces
that it was the top when the metal was cast. The con- of matting adhering.
vex side is often covered with low mounds, similar to BI 15. D . .135, Th .. 03. Most of edge missing.
BI 16. D . .18, Th .. 045. Weight 2.65 kg. Edges badly
those found on the oxhide ingots, which represent de-
damaged. Both sides convex.
pressions left in the mold when it was packed down BI 17. D . .23, Th .. 031. Weight 2.9 kg. Edge missing in
into clay. Occasionally, one edge of an ingot is either two places.
cut straight across or has a slight protrusion; either BI 18. D . .185, Th .. 031. Weight 1.2 kg. Part of edge
feature would seem to indicate the point at which the missing.
BI 19. D . .22, Th .. 035. Weight 3 kg. Edge missing in
molten metal flowed from a channel into its dished-out places.
mold. BI 20. D . .21, Th .. 033. Weight 3.5 kg. Part of edge
The bun ingots average about twenty centimeters in missing.
diameter and three to four centimeters in thickness, but BI 21. Fig. 93. D .. 18, Th .. 035. Weight 1.4 kg. Ap-
proximately half preserved.
the weights vary widely. These variations, even among BI 22. D . .18, Th .. 03. Weight 1.05 kg. Approximately
ingots of approximately the same dimensions, are due half preserved.
to the poor state of the metal, which in some instances BI 23. D . .26, Th .. 05. Weight 3.8 kg. Approximately
crumbled under slight pressure. This advanced decay, 2/3 preserved.
found in only one of the oxhide ingots, suggested that BI 24. Fig. 93. D .. 223, Th .. 04. Weight 2.05 kg. Ap-
proximately half preserved.
the bun ingots were of a different composition than the ---
more solid oxhide ingots. The qualitative analysis 141 I wish to thank the Ajax Metal Division of H. Kramer
(append. 3) of the one sample taken showed that this and Co. for making this and other analyses for me.
14:: Coghlan, op, cit. (sicbra, n. 55) 23-24, defines true, as
Wainwright, Klio 14 (1913) 1-36.
13(; opposed to accidental, bronze as that which contains more than
Carling, Csl»: Bronze .. 299-300.
]3, 3 per cent tin.
r aa H. Otten, MDOG 94 (1963) 20-21. 143 At the time of publication it had been possible to sample
t a M. Mellink, AlA 67 (1963) 176. only one bun ingot and one slab ingot. Because both proved
140 J. Nougayrcl, CRAI (1960) 166, also places Alasia on to be bronze, and because all of each class seemed different
Cyprus on the basis of documents which could, I feel, equally from the cxhide ingots, by direct observation, I am assuming
point to North Syria. that all o l the ingots of these two types were bronze.
 VOL. 57, PI. 8, 1<J67] BASS: THE l"COTS 79

-_ •.. - .....
•
Bll
-
BI3
-"- BI5

_5...._ B/ 7
•••••
B/ B
_5<",

8/ 10

_5~'"
BI24
B/ 21 BI30

FIG. 93. Bun ingots 1, 3, 5, 7, 8, 10,21,24, 30.

80 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK jTRANS. A!lII~H. PHIL. soc .

. ', "
.!

.' " '.: .

. ":.".'
.... ',.

.. '
-,':."
'\

....... :....
.-:' •
<>•• <>. m.tal
f.ogmen .. and
>-~-...o .1 ch,,""al

2lcm

(I "d"p •• d f<<>mlyle.<> ••• Met"Uu<gy" Ilg.$)

CONCEPTION Of fURNACE WHICH COULD HAVE MADE COPPER OR

BRONZE BUN INGOTS

FIG. 94. BI 30. FIG. 95.

BI 25. D . .205, Th .. 04. Weight 2 kg. Slightly less than range.>" The simplest are merely the cakes which
half preserved. formed in the bottoms of shallow pits over which either
BI 26. D . .185 Th .. 023. Weight 1.1 kg.
1
Approximately are was smelted or metal melted, and are not in fact
half preserved.
BI 27. D . .19, Th .. 04. Weight 4 kg. Approximately half
ingots at all. Gowland 147 and Percy 118 have recorded
preserved. the furnaces and operations which produced such pieces
BI 28. D. .20, Th .. 03. Weight 1.8 kg. Approximately in nineteenth-century Japan, Korea, and Sikkim.
half preserved. The excrescences found on a number of the Gelidonya
BI 29. D . .18. Approximately half preserved. cakes, as well as on several examples at Mohenjo-
In addition, there is one piece which is so much more daro,':" indicate that the type of furnace suggested by
regularly formed that it is difficult to believe that it falls Tylecote 150 for the manufacture of stamped Roman bun
into the same category: ingots was known at a 11111ch earlier date (fig. 95). In
BI 30. Figs. 93 and 94. D . .2[, Th .. 018. Weight 2.5 kg. such a furnace. the molten metal collected at the bottom,
Notch in one side. The convex side is very smooth, but rather than being allowed to solidify there, it was
suggesting that this was fanned in a hard mold,
perhaps made from the base of a rounded jar; the 146 Only a partial list, to indicate the wide spread of hun
flat side is likewise extremely smooth and flat. ingots, may be given here. India: Mackay, Motiewjo-Duro I,
lacking the normal rough surface. The metal of 451; S. R. Rao, Antiquity 37 (1963) 99, with pl. Il(a).
the piece has not been analyzed, so it is not possible Susa : ibid. Syria: Buchholz, PZ 37 (1959) 15. Anatolia :
to state whether it is copper; if it is, the smooth Praeworski, Metaf!. Anat., 22 f., 108 l., with pl. 3: 1. Cyprus:
surface may be "the result of skilful smelting of O. Davies, BSA 30 (1930) 78. Crete: Marinates, Kadsnos 1
pure oxidized orcs," as in the case of certain Welsh (1962) 91, with pl. 1. The Balkans: Buchholz, P Z 37 (1959)
ingots.w- Smooth surfaces may be obtained by 16. Sicily: L. B. Brea, Sicily, fig. 48. France, imported from
protecting the molten metal from the atmosphere.P" Spain in Roman times: F. Benoit, Gollia 20 (1962) 154-157,
with figs. 18-19. Germany: P. Reinecke, Ger-nowo 22 (1938)
Similar plano-convex disk-shaped ingots, in a variety pl. 1: 1-7. Great Britain in Roman limes and earlier: W. Gow-
of metals, cover a wide geographical and chronological land, Archaeoloqui 56 (2nd series 6, 1899) 287-289; Tylecote,
Met. in Archae., 21, 24, 29-34, 107.
141 Gowland, op, cit. (supra, n. 146) 278--285.

R. F. Tvlecote, Met. in Archae., 33.

144 148 Tylecote, Met. -in Arcliae., 27-28; citing J, Percy, Metal-
H. Kramer and Co. today produces copper ingots with
14.l'.i tvro» ... , 389.
smooth surfaces by floating powdered charcoal on the molten 141) Mackay, Mohewjo-Doro I, 451, with pl. 132,38-39.
150 Met. in Archae., 31, 33, with fig. 5.
metal.
 VOL. 57, PT. 8, 1967] BASS: TI-JE INGOTS 81
run off through a channel into an exterior mold. Some
of the metal undoubtedly froze in the channel each time
the furnace was tapped, and the resulting excrescences
either remained as lugs or were cut off close to the
cakes, leaving straight sections on their otherwise curved
edges; the latter phenomenon appears not only at
Gelidonya but also at Arkolochori.v-
The commonly known copper bun ingots were the
result of smelting, but, as we have seen, the bun ingots
at Gelidonya 'were probably all of bronze.v" This poses
the problem of whether they were produced by smelting
copper and tin ores together, or were an alloy made by
melting metallic copper and tin. Both methods were
possible. Gowland has produced bronze by smelting ~~
copper ore and tin are together in a primitive furnace
with a central cavity about the size of an average bun
• ingot from Celidonya.v"

Palestinian in origin.'>'
Both ores occur in Syria/53
and in this regard it is interesting to note that Buchholz
interprets representational examples in Egypt as Syro-

The occurrence of both metallic copper and tin on the

Gelidonya wreck, however, suggests that the bun ingots
were the result of the fusion of these two metals. Al-
_". -
~~

FIG. 96.
-
Slab ingots.

storehouse at Tell el-Amarna.v" Actual examples show

though the Tomb of the Two Sculptors at Thebes has
no standards of size or weight, and many were broken
shown that smiths sometimes worked directly from cop-
into small fragments which would more conveniently
per and tin ingots in casting bronze objects/55 such a
fit into crucibles for later melting. Indeed, close ex-
procedure might have caused weaknesses in the finished
amination of copper cakes from Europe has revealed
product owing to segregation of the metals; 156 a sounder
that they were broken up before they had cooled and
method would have been the preliminary making of hardened.v" a practice still used in Korea, at least during
ingots, which would have been remelted for bronze the last cennu-y."'"
casting. The scene of smiths at work in the tomb of Rekh-
Not to be overlooked is also the possibility that the mi-re' shows that oxhicle ancl bun ingots were used in-
ingots were made of scrap bronze, which was so readily discriminately, for both are seen being carried toward
available to those who laded the ship with quantities the furnace. That cxhide ingots at Geliclonya were
of broken bits and pieces. Brea states that irregular copper and the bun ingots bronze seems to have 110 par-
ingots of the Sicilian Adrano hoard, which contained at ticular significance for the time; we have seen that some
least one bun ingot, still showed traces of fibulae, brace- oxhide ingots from other sites were bronze, and that
lets, and other objects which had not completely most bun ingots were copper. VVe may conclude, there-
melted.'>" In this case, the end in view was simply the fore, that bun ingots were simply convenient cakes
which appeared in copper, bronze, iron.''" anel possibly
production of easily handled and bartered units and was
gold and silver or tin''"
not a metallurgical necessity.
As in the case of the oxhide ingots, these cakes were
Ill. SLAB I~GOTS
not used as currency, although both forms are seen in
great quantities and, apparently, in uniform sizes in the Nineteen slab ingots (fig. 96) were stacked in neat
piles of three or four ingots each. All were found in
151 Marinatos, Kadnios 1 (1962) pl. 1. area G, and all were probably complete at the time the
152 Tylecote, Met. in Archue., 28, citing Gowland, JiM 7 ship sank. They 'were by far the softest of the ingots,
(1912) 23-49. but hardened somewhat after drying; several of the first
153 Forbes, Met. in Ani., 239-240, doubts that these ores were
mined in antiquity. 158 N. de G. Davies, The Tomb of Mervra, pl. XXXI, de-
154 Buchholz, PZ 37 (1959) 15. scribes them as "loaves" (p. 37). It is difficult to interpret
155 Norman Davies, Tomb of Two Scuiptors, pl. XI. The such objects, especially when the original colors are not men-
identification of the white ingot as lead (p. 63) has been tioned in publication (s'lipra, n. 34).
rightly corrected by Wainwright, Antiquity 17 (1943) 97. 159 Childe, The Bronze Aqe, 29.
156 Dr. Robert Maddin, Chairman of the Dept. of Metal- 160 Gowland, op, cit. (supra, n. 146) Zi9.
lurgy at the University of Pennsylvania supplied me with this 161 Forrer, Gerwuuno 16 (1932) 217-218.
information. 162 Nina de G. Davies and Norman de G. Davies, The Tombs
157 Erea, Sicily, 194, with fig. 48 on p. 193. of J1f enhhcperrosonb , 11, with pl. XI.
82 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRANS. A1>IER. PHIL. SOC.

These ingots were first thought to be blanks which

were to have been hammered into tools 01- weapons;
actual castings, however, prove to be quite regular and
smooth (p. ll3). That they are ingots is indicated by
their similarity to somewhat earlier silver ingots from
Egypt.':" Although the Egyptian ingots are smaller,
they are identical in shape, and gold ingots found with
FIG. 97. 513.
them also bear the small cracks which appear on the
Gelidonya ingots; I do not know the cause of these
to be removed from the concretion partially crumbled cracks.
away. Two important factors must be considered in a study
The slab ingots are somewhat irregular in shape, but of these ingots. The first is that all came from the area
all are long, roughly oval pieces of ftatish metal which of the "captain's cabin." While it is true that metallic
lack the bubbly surfaces seen on oxhide and bun ingots. cargo was found there, the slab ingots form the only
They are between twenty and thirty centimeters in type of metallic object found only in that area (area G),
length, six and eight centimeters in width, and one to which also yields the scarabs, cylinder seal, maceheads,
one and a half centimeters thick. The qualitative analy-
•
knucklebone, lamp, food, and most of the weights and
sis of the only ingot of this type which was sampled whetstones. This might indicate only that they were
indicated that it had a high proportion of tin (append. ready to be utilized in making bronze objects, if it were
3). The quantitative analysis showed that at least that not for the second factor: the unusual regularity of their
ingot was of true bronze, as was the case with the bun weights.
ingot, consisting of 94.01 per cent copper and 5.27 A lmost all of the complete slab ingots weigh exactly
per cent tin.163 1.0 kg., with variations in weight among the others ex-
In the following catalogue, figures represent maximum plained by the addition of concretion or the subtraction
dimensions. None of the ingots bore marks, but the of chips of metal. Slab ingots 18 and 19 may be dis-
surfaces of most carried tiny hairline cracks which counted as fragments, but 14 and 17 seem to be nearly
radiated in clusters from various points. complete and these weigh .5 kg. each, or half the weight
SI I. L. .27, W .. 07, Th .. 015. Pres. weight 1.0 kg. End of most of the ingots; slab ingot 13 is probably another,
chipped. broken, "half unit." This leaves only ingot 6, ap-
S12. L. .265, W .. 075, Th .. 011. Weight 1.0 kg. proximately one anel a half times the weight of the
SI 3. Fig. 97. L. .277, W .. 074, Th .. 012. Weight 1.0 kg.
SI 4. L. .28, W .. 07, Th .. 012. Weight appr ox. 1.0 kg.
a verage ingot.
SIS. L. .302, W .. 075, Th .. 013. Weight l.l kg., with The poor state of preservation of these ingots, with
some concretion. the resultant unknowns and variables in weight, pro-
SI 6. L. .29, W .. 056, Th .. 017. Pres. weight 1.55 kg. hibits definite conclusions, but the strong probability
Edge chipped, but some concretion adhering when exists that they, of all the ingots, were the most likely
weighed.
SI 7. L. .29, W .. 071, Th .. 012. Weight 1.0 kg.
to have served as currency; their position on the ship,
SI 8. L. .297, W .. 075, Th .. 01. Pres. weight .9 kg., with with the other "personal possessions of the captain,"
10 to 20 grams missing fr0111 chip on one side. can only strengthen this probability. They are not, it
SI 9. L. .247, W .. 08, Th .. 008. Pres. weight 1.0 kg. seems, represented in contemporary art,165 and the con-
SIlO. L. .273, W .. 082, Th .. 008. Weight 1.0 kg. clusions must be based on the evidence of the ingots
SI II. L. .218, W .. 065, Th .. 01. Pres. weight .7 kg. One
end bent upward. thernsel ves.
SI 12. Pres. L. .23, W .. 07, Th .. 014. Pres. weight .8 kg.
End missing. IV. TIN (PROBABLY IN INGOT FORM)
SI 13. L. .IS, W .. 055, Th .. 012. Pres. weight .35 kg.
One end possibly broken off diagonally. Under the copper ingots of areas P and G were found
SI 14. L .. 206, W .. 075, Th .. 011. Pres. weight .5 kg. three piles of white, powdery tin oxide.v" The single
One end chipped. pile in area G, found beneath oxhide ingots 22 and 23,
SI 15. Pres. L. .235, W .. 07, Th .. 015. Pres. weight 1.0 kg. was collected and raised in a plastic bag. When dry, it
One end broken off. weighed 8 kg., which was approximately half of what
5I 16. Pres. L. .209, VI/.. 08, Th .. 01. Pres. weight .9 kg.
End broken off. had been preserved on the wreck. The tin oxide, how-
SI 17. L. .215. W .. 078, Th .. 013. Pres. weight .5 kg.
164 F. B. de 1a Roque, Trisor de Tad, pl. VI, no. 70513
Very worn edges j not kno-wn if complete or broken.
SI 18. Pres. L. .085, W .. 068, Th .. 01. Pres. weight .3 kg. (silver) and pl. IV, no. 70505 (gold).
1(J5 N. de G. Davies, Deir cl Gebrtnui I. pl. XIV, shows a man
End of ingot. Four impressed holes on convex side;
other side flat. approaching smiths, who are blowing 011 a fire through tubes,
SI 19. Pres. L. .105, W .. 075, Th .. 005. Pres. weight .2 kg. with a basket of red objects (called "food and jars of beer (?)"
Ingot end. on p. 20) which resemble slab ingots. The scene is from the
----=
My thanks
163 to M r. M. Stoltz of the Ajax Metal Division
Sixth Dyn. tomb of Aba.
1(J6 My thanks to the Turyag Laboratories of Izmir for this
for this analysis. identification.
 VOL. 57, PT. 8, 1967J BASS: THE INGOTS 83
ever, was found only where it had been covered and
preserved by masses of copper and sea concretion, and
it would be impossible to estimate the amount that
might have been washed away from more exposed areas.
Miss Frost noted that the tin oxide in Area G seemed
to extend from and below a squarish shell of concretion
(fig. 98), which was thought at the time to have been
formed over a box. Similar shells of concretion, how- . ".
.' '.
ever, are found over powdery iron oxide on later " ..
wrecks, iur and these shells preserve exactly the original
shapes of the corroded metal; if one end has not been .... '.'
covered by concretion, the oxide has sometimes com-
pletely disappeared, leaving an empty space. At Geli-
donya, therefore, we may have the shape of the end of
a tin ingot which was six centimeters on a side; the
• length of the bar is unknown. Such a shape would
correspond to the larger ingot which we have identified ., .
as tin in the tomb of Nebamun and Ipuky (I'. 65). . '.
The extreme decay of the tin may be explained by the
electrolytic action which so badly damaged the copper ' ...
ingots with which it was in contact (p. 52). ( C ...
Sources of tin ore throughout the Bronze Age are TIN OXIDE -,
debated by students of metallurgy. Lucas rue and Wain- L
J
wright 169 point to Syria, but Forbes, although not deny-
ing its existence, questions its importance there.v"
Przeworski 171 records tin in Anatolia, but Wainwright
states that there is no proof that it was mined there at
an early date.172 Wainwright then suggests Crisa, on
the north shore of the Gulf of Corinth;"? but Forbes Looking South in Area G
doubts that are was ever mined there; 114 the more re-
cent excavations throw no new light on this subject.v"
The presence of copper oxhide ingots on Sardinia,
FIG. 98.
whether imported or manufactured 10caH)', implies that
ships such as that wrecked at Cape Gelidonya sailed at
was "most likely" from Anatolia, "possibly" from Syria,
least that far west; this would take them far on the
and "even conceivably" from Central Greece.1T7
way to the tin fields of Spain, but we have no evidence
Although the tin at Gelidonya is not the earliest
that such voyages took place. Sardinia, itself, has tin
known, it is the earliest purely industrial tin; the pre-
ore, but whether it was known at such all early period
viously noted lack of tin bars or slabs in founder's
is not certain.F" What does seem to be certain is that
hoards 178 may now be explained as due only to the for-
tin was not mined on Cyprus; J. R. Stewart believes
tunes of excavation. The identification of the rectangu-
that the tin used there, at least in the Early Bronze Age,
lar ingot from the tomb of Nebamun and Ipuky as lead
107 Bass, NG 124 (July 1963) fig. on p. 142. J-I. Frost, was based largely on the belief that tin ingots did not
Under the Med., 20-21, 59-60. appear this early. Other re-interpretations in art and
HIS Lucas, Anc. Eq'jlpt. Mat. and Industr., 288--289.
literature may now be needed, and Albright has already
res JEA 20 (1934) 29-32; Antiquity 18 (1944) 101.
EO M ct. in Ant., 239-240, 252-253.
suggested that the Gelidonya tin provides an additional
171 Praeworski, Metall. 1'n Anat., 91, 101-102. argument for the identification of mulklt in the old
1'2 Antiq1/it.v 18 (1944) 59. Assyrian documents from Cappadocia as "tin" rather
m Ibid.; O. Davies, J HS 49 (1929) 92-93. than "lead:' although he points out that these documents
1~4Met. 1'n Ans., 244.
175 Dar, j annoray, and H. and M. van Effenterre, Kirrha.
are considerably older than the shipwreck.v?
Sec more recently S. Benton, "No Tin £1'0111 Kirrha in Phokis," m SGE IV, lA, 242.
A"tiq"ity 38 (1964) 138. 1,8 Forbes, Met. in Ani., 247.
a-o Guido, Sardinia, 28. 179 Letter dated 7 Dec. 1960.
 VI. THE BRONZES

GEORGE F. BASS

The study of the bronze tools, weapons, household l. PICKS

utensils, and personal objects has profited greatly from The 1110st numerous tools at Cape Gelidonya were
two works of major importance which have appeared picks (Catling's agricultural tool D).' Each was cast
since the Geliclonya excavation: J. Deshayes, Les Owtils as a long, thin bar with a T-shaped end. The sides of
de bronee, de l'Indus au Donube (IVO-[]6 '/'Jtillenaire), the T were rolled over to form an open socket, and the
and H. Catling, C:ypriot Broneeiuork in the l11ycenaean opposite end of the bar was hammered into a point. A
Worui? References in this chapter will be abbreviated sharp point is preserved only on B 15, for the otl:ers
to "Deshaves" and "Carling." Catling has published a have been either blunted or broken off. No two picks
list of the ~111etalobjects from Gelidonva," but it will be are exactly alike, but several distinct features may be
seen that his identifications differ from mine in a few noted. The bars of some (B 3, B 17, B 18) are
instances. Some of the differences may be ascribed to rectangular in section; the point of none of these is
personal opinion, but in some cases they are the result preserved. Other bars (B 1, B 16, B 19) are round
of less than complete information which Catling had in section; that these have been hammered out of flat
on hand at the time of his work, through no fault of his bars is apparent [rom the central grooves, on the same
own. The unusually difficult conditions under which side of the pick as the socket opening, seen on B 20 and
the objects were first catalogued at our camp on the B 21. This would suggest that the rectangular bars
beach led to incomplete information in some cases, and were simply unfinished, and that their points are missing
it was only during the following summers that the pieces only because they had not yet been hammered out, were
could be laid out and grouped for careful restudy in the it not for Cypr. IvI11s.Met. 1708, which is rectangular
Bodrum Museum. Indeed, I was not even aware of and pointeel. Still other bars (B 15) are flat on the
several objects, which had been raised in 1959, until side of the socket opening, but are otherwise rounded.
they were given to me to place in the Bodrum Museum B 6 and B 14 are almost oval in section. B 2 and B 11,
ill 1962. while rectangular in section, are not nearly as wide as
In addition to Desbayes and Catling, a number of those first mentioned.
studies of individual types, such as those of lugged The manner in which the sockets and their bars form
adzes by R. Maxwell-Hyslop, Aegean knives by N. angles varies. Sockets of picks most rectangulal~ in sec-
Sandars, tripods by J. L. Benson, and axe-adzes by tion are simply bent in toward the socket opemngs, ~o
C. Schaeffer, have offered great help in supplementing that the bar may lie flat with the socket raised in the air.
exca vation reports. Others, notably 13 1 and B 6, which are round or oval
In the following catalogue, area locations on the site in section, are flattened on the outer sides of their bends,
are given in parentheses; (1959) indicates that the so that they rest naturally with both bar and socket
object was raised during the summer when the 'wreck pointing slightly upward.
was first investigated, and a precise location is not Forty-seven pieces, mostly fragments. are catalo.g~~d.
known. Those pieces which have no area designations None of the fragments join, but there is the possibility
were mostly discovered during the breaking open of that some of the sockets and bars were originally from
oddlumps of concretion after the excavation had ended. the same picks. Vle may estimate only that the frag-
Although the uses of some of the implements arc ments are £1'0111 a minimum of tv..-enty-four picks.
ambiguous, the bronzes have been grouped as foll~~vs:
agricultural tools (I-VII), wooel-workmg tools (VlII- B 1. Figs. 99 and 101. L. .252, socket D .. 033, bar W.
.03 tapering to .01, bar Th: .014 t.o .01, so~ket Th .. ~1.
XI), metal-working tools (XII-XIJI) , light tools, Tip broken. Bar round l~ sectl~n, commg t.o. P01t1t
mostly for domestic work (XIV-XVII), weapons at end. but is rectangular 111 section at transition to
(XVIII), personal and household objects (XIX- socket (1959). "
XXV), and miscellaneous categories, including :111- B 2. Figs. 99 and 101. L. .245, socket W ... 030. Both
sides of socket broken (P). Very similar to Cypr.
finished castings, unidentified implements, and casting Mus. 1958 VI-24 12, from the Stylianou Hoard.'
waste (XXVI-XXX). All of the implements are B 3. Figs. 99 and 101. L. .205, socket D, .032.. Socket
thought to be of bronze, but the reader should bear III complete; end nussmg or unfinished. ~ incised on
mind that only a very few fragments have actually socket (P). A number of very similar}ools (Met.
1708, with more pointed end, 2164, 216.), 2168, and
been analyzed. 2173) are in the Cyprus Museum."
1 I wish to express my great thanks to Dr. Catling for suppl~- B 4. Figs. 99 and 101. L. .23, socket D .. 03. Point miss-
ing me with many references and comments before the publi- ing (1959).
__ -"0
cation of his book. Although our conclusions do not agree, I
could not have formed mine without the benefit of his work 3 Ibid., 82-83.
and ideas, for which I am especially grateful. 4 Ibid., 82, no. 1, with fig. 7:6 and pl. 4 .i.
5 Ibid. 110S. 3-8.
:.1 Catling, 292-294.

84
VOL. 57, PT. 8, Jl)(i7] BASS: THE BRONZES 85

81

86

82

87

I
_
cQ
.:..<, !
""":""'::=,"'~"""" ,,,-:~--~,
.~"

88
_.~~. ,
I

'~~".='=====-J-'
83

';"~"'~

C':================) 811

=======-='
84
It:
c:::-

814
I
===.

~.
'=:".=
.. ~~.)

•
~=-='=====~
~=. ====~·_I__ ---=--=--_:::;>- [ _~
--._-._-~
815
I

85

816
o 5 10 15 20

FIG. 99. Bronze picks: Bl to B16.

86 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

B18

B35 B45

E)"~
B3I

B19
C . "\tlJllL..'<ulttiIti1
d:lb;JlIw~l\I\'J\~

B36
'~~

B46

B34 B43 B47

B 20

o 10 20

FIG. 100. Bronze pick fragments: BI8 to B 47.

B 5. Figs. 99 and 101. L. .14, socket D. .03. Socket B 23. L. .06, max. pres. D .. 012. Round in section; tapers
chipped (G). to rounded end .
B 6. Figs. 99 and 101. 1.. .23, socket D. .034. Point B 24. L. .033, Vl.. 029, Th .. 006. Rectangular in section;
missing (G). point missing (P).
B 7. Figs. 99 and 101. L . .19, socket W .. 035. Socket B 25. 1.. .043, W .. 019, Th .. 011. Rectangular in section;
cracked; blunt end (1959). point missing.
B 8. Figs. 99 and 101. 1.. .215, socket W .. 035. Half of B 26. L. .026, VV.. 021, Th .. 013. Almost rectangular in
socket missing (1959). section, but rounded on one side.
B 27. 1.. .08. (E).
B 28. L. .11, D .. 015. Round in section; central groove
Bar fragments: on one side; point missing (5).
B 9. 1.. .105, max. W .. 024. Blunt end (G). B 29. L. .05, D .. 013. Round bar, perhaps pick fragment
B 10. 1.. .092, W .. 015, Th .. 019. Blunt point (G). (G).
B 11. Figs. 99 and 101. 1.. .265, W .. 025, Th .. 01. Un- B 30. 1.. .021, pres. W .. 026, Th .. 008. Rectangular in
usually long (G). section (P).
B 12. 1.. .15, W .. 028. Fairly sharp point (5).
B 13. 1.. .18, W .. 018, Th .. 011. Blunt point (G). Sockets:
B 14. Figs. 99 and 101. 1.. .225, W .. 028. Fairly sharp
point (1959). B 31. Fig. 100. L. .085, '1\', .032. Broken off at transition
B 15. Figs. 99 and 101. 1.. .16, W .. 025. Very sharp to bar (1959).
point (G). B 32. 1.. .059, D.. 035. Very fragmentary (G).
B 33. L. .08, D .. 035. Broken off at transition to bar (P).
B 16. Fig. 99. 1.. .18, W .. 02. Fairly sharp point (1959). B 34. Fig. 100. 1.. .08, D.. 033. Broken off at transition
B 17. L. .07, W .. 002, Th .. 01. Rectangular in section; to bar (P).
point missing (G). B 35. Fig. 100. L. .10, \AI .. 03. One side missing; broken
B 18. Fig. 100. L. .O??, Vl .. 032 tapering to .022. Rec- off at transition to bar (1959).
tangular in section; point missing (1959). B 36. Fig. 100. 1.. .078, W .. 025. Chipped (G).
B 19. Fig. 100. L.. 08, D.. 023. Unusually thick, blunt B 37. Fig. 101. L. .05. Mashed or hammered together
point (1959). (G).
B 20. Fig, 100. L. .078, D .. 015. Central groove on one B 38. L.. 078, W.. 03. Very fragmentary (M).
side; point missing (P). C/. Cypr. Mus. 1958 B 39. 1.. .05, W .. 03. Very fragmentary (M).
VI-24 13, with groove and pointed end, from B 40. 1.. .04, W .. 03. Very fragmentary (M).
Stylianou Hoard. B 41. 1.. .04, W .. 03. Very fragmentary (G).
B 21. Fig. 100. L . .15, W .. 025. Central groove, blunt B 42. L .. 045, D.. 03, Th .. 003.
point (G). Cf. Cypr. Mus. Met. 2174, with groove, B 43. Fig. 100. 1.. .089, D.. 034, Th .. 003. Broken off at
and with socket preserved. transition to bar (G).
B 22. 1.. .07, W .. 016 to .01, Th .. 012. Rectangular in B 44. 1.. .061, W .033. Very fragmentary (P).
section, but with rounded edges, and blunt, rounded B 45. Figs. 100 and 101. 1.. .07, W .. 035. Part of socket,
end. hammered into flat fold (G).
 VOL. 57, PT. 8, 1967] BASS:.JHE BRONZES 87
B 46. Fig. 100. L.. 083, W .. 025. (1959). that there are three categories into which the pieces fall :
B 47. Fig. 100. L. .065, D.. 035. Broken off at transition nearly complete picks, bars, and sockets; most of the
to bar (G).
sockets are nearly complete, but in no case is there
That all, or very nearly all, of these tools were carried a socket with part of the bar preserved. The smith ob-
for their scrap value is evident. It is of some interest viously cut the bars off near their sockets, perhaps be-

•
......-_..-
Bll

'(.., .....
B14

......
B 15

B37

.,' .....
B6
r~__
B45

FIG. 101. Picks and pick fragments.

88 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK lTRANS. Ai\lER. PHlL, soc.

cause plain bars were easy to work with, and then prob- have been complete in antiquity:
ably, as is twice seen, hammered the sockets flat; the B 48. Figs. 102 and 105. L. .215, blade W .. 047, max.
only explanation for the latter action would be for blade Th .. 006. Socket damaged and tip of blade
economy of space in packing the material. cracked (1959).
\Ve have seen that picks with both rectangular and B 49. Figs. ]02 and 105. L. .21, blade W .. 052, blade Th .
.006, socket Th .. 005. Part of the socket is missing,
round bars are found in Cyprus, and the extremely close
but at least part of this is newly broken (1959).
similarity leads to the conclusion that the Gelidonya B 50. Fig. 105. L. .19, blade W .. 065 to .03, socket W .. 05,
picks are Cypriot in origin (cf. Catling. pl. 53:b and blade Th .. 005. Heavily concreted; socket broken
c). Cypriot picks with wide bars (Catling, 82, 110. 3, (1959).
with fig. 7:7; pl. 4 :j) did not appear on the wreck. B 51. Blade only. L. .111, max. W .. 05, Th .. 008. Similar
to B 53. (E).
Carling (82-83) points out that such picks, which B 52. Fig. 102. Blade only, similar to B 51. L. .08, W.
may have been used in farming or mining, are not .033, Th .. 005 (1959).
known in the Aegean, and suggests that they may be B 53. Fig. 102. Blade only, similar to B 51. L. .075, W .
a Cypriot development related to Near Eastern plow- .04, Th .. 005 (P).
shares. Dating depends on the dating of the Cypriot
hoards (see chap. XIV), which contained the examples TYPE 2
from that island, but a parallel (only .15 long) from
Tell el Fara (south) has been dated to ca. 1300 B.C.' The blades of Type 2 swell, rather than taper, toward
broad, rounded ends. This shape does not seem to cor-
respond to any listed by Deshayes, but one of his hoes
II. HOES AND SIMILAR SOCKETIZD BLAD!·:S
of Type B 3b is similar; 17 it is part of the Enkomi
TYPE 1 Founder's Hoard and its date is, therefore, uncertain.
'Two parallels found in the Stylianou Hoard are prob-
Type 1 corresponds closely to Deshayes' open-socketed
ably from Enkcn-i;" and an example excavated at
hoe type B 4,1 and Carling's agricultural tool C(a),8
Enkomi Illay be dated to the early twelfth century B.C.19
with its long blade tapering toward a rather pointed
All except B 55, from area P, were found in area G
edge. It finds its closest parallel in Palestine, at Beth
and were, therefore, from near the stern of the ship in
Shemesh, in the twelfth or eleventh century B.c.9 An-
the "cabin area." Only one may have been complete
other Palestinian example, from the end of the second
in antiquity:
millennium at Tell Beit Mirsim," is fairly close in
shape. Less alike, but perhaps related, is a hoe in the B 54. Figs. 102 and 105. L. .135, blade W .. 042, blade Th.
Cyprus Museum." A piece from the Founder's Hoard .004 tapering to edge, socket Th. .0015. Socket
at Enkorni;" which Deshayes has listed under his Type chipped (G).
B 3b,13 seems to be the same but worn, and Catling lists
Blades and Blade Proaments :
another fragmentary example from Cyprus.!"
It is evident that the origin and dating of this type are B 55. Figs. 102 and 105. L. .16, W .. 06, Th .. 006. Com-
uncertain. Albright states that iron plow points were plete blade, broken at socket (P). Larger than B 54,
but very similar in shape.
used in the eleventh century in Palestine," which sug- B 56. Fig. 102. L. .085, W .. 042, Th .. 003. Complete
gests an earlier date in the last quarter of the second blade with bit of socket preserved (G).
millennium for bronze examples. \AIe may, on the basis B 57. Fig. 102. L. .158, W .. 058, Th .. 007. Sides of socket
of the Beth Shemesh example, tentatively place the missing, blade chipped (G).
B 58. Fig. 102. L. .06, W .. 07, Th .. 004. Rounded end
Palestinian pieces in the twelfth century. Catling be- of very wide blade (G).
lieves the Enkomi Founder's Hoard to be probably not
prior to 1200 B.C.'" TYPE 3
All of the Gelidonya examples were probably in the
forward half of the ship (areas P and E). Two may Blades of Type 3 are short and approximately the
same widths as their sockets throughout. They end in
G Petrie. Beth-Pelct I, 9, with pl. XXVI, no. 98, called a blunt, almost straight edges. It is possible that some
"spear butt." of these may be only very worn examples of Type 1
r Deshayes I, 140, 142.
B Catling, 80.
which have been continually resharpened.
9 Grant, Ain Sbenis II, 33, no. 49, with pl. XLVII: 4l. The closest parallels are from mainland Greek hoards
10 Albright, Tell Be-it j\f.irsim III, 32, with pl. 62: 4. found on the Athenian Acropolis." and at Anthedon in
11 Nicosia Mus. No. 164; Gjer-stad, St. Pt'eh. C'}lpr., 237;
Catling, 80, no. (a) 1, with fig. 7: 5 and pl. 4: a; Deshayes II, 17 Deshayes I, 139, no. 121i; Catling, 80, no. (b) 2-8, with
no. 1225. pl. 4: c.
12 Schaeffer, Enkomi-Alasia, 29, with pl. LXV: 9. 18 Catling, 80, nos. (b) ]3-14, with fig. i: 3 and 4, and pl.
13 Deshayes I, 139, no. 1218, 4: e and f; BCH 83 (1959) 338, FIg-.2, shows one of the
14, Carling, 80, no. (a) 3.
pieces.
15 Albright, Tell Bcit Mirsill1 III, 32-33'. 19 Catling, 80, no. (b) 1, with pl. 4: h.
20 Moutefius, Grice Precl., 155, with figs. 494 and 495.
]6 Catting, 281.
 YOL. 51, PT. 8, 1967J BASS: THE BRONZES 89
-_C.'.,.
• c. ' ~"

"
"
} ~
(
""
:,.'/ , j -11--
;"'.
UU 52

49 53
B 48
r
o "
r~\
-

~_I
56 58

54 55
• 57
F

o 10 20

FIG. 102. Bronze hoes and socketed blades: B 48 to B S8.

Boeotia; ~1 another similar example, from Cyprus, is sarily that of the bronzes." Catling would place that
shown by Gjerstacl." Deshayes has included all of hoard after 1200 B.c.,z~ and Desborough shortly before,"
these under his Type B 3b~but as we have seen above, but neither states his case with complete confidence. The
the B 3& pieces from the Enkomi Founcler's Hoard are Anthedon Hoard is elated by Catling in the twelfth cen-
better fitted into our Types 1 and 2, and his B 3b ex- tury or later on the basis of a tripod fragment, but we
ample from l'vlegiddo 23 is quite unlike those from shall see that such a late date is uncertain. ~7
Gelidonya. The examples at Gelidonya were found in all areas
As usual, the elating of the hoards is difficult. Only except area G.
two sherds were published with the Acropolis Hoard,
B 59. Figs. 103 and 105. L. .165, blade W .. 06, socket W.
but their date of LH III A2, by Furumark, is not neces-
.05, blade Th. .0025. Complete, but blade worn
21]. C. Rolfe, AlA 6, 1st series (1890) 105, with pl. 15: 5. through use. Impressions from casting on upper
side of blade and on side of socket (P).
~2 Gjerstad, Stud. Prell. Cypr .. 237, lower right; Catling, 80,
no. (a) 1, with fig. 7: 5 and pl. 4: a, identifies it as being from
---
Furumark,
::'4 Cbronoloo», 95, n. 1.
the Gunnis Hoard. 25 Carling, 297.
23 Deshayes II, no. 1220; Schumacher, Tell el M1Itesellim I 2G Desborough, Last Mycenaeans, 49, n. 7.
(text), 86-!l7,fig. 120. 27 Ibid., 48, 11. 6, but see 49, n. 7. Catling, 297.
 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. soc.
90

o \
rv .-:? r-
\ /~

•
• r
62 63
& 60
B 59 p,

O 'J-~~1~
.•" ilL.·•....../.
-

~ .':-.'
--.-.--
'~:-
.~- -,
.

.. '-. ~
..
::.
-"

"
IfJi- ...' .'.
rV :

67
64 66
oems 15

FIG. 103. Bronze hoes and socketed blades: B 59 to B 67.

B 60. Figs. 103 and 105. L. .145, socket W .. 048, socket slightly rounded edge, but does not swell out with curved
Th .. 003. Shattered, but possibly crushed during sides as do blades of Type 2. The most similar blade is
sinking of ship by weight of ingots above. Piece of on a tool shown, but not identified, by Petrie; 28 the
edge missing (P). sides of the socket in that case are of the type more
B 61. L. .13, blade W .. 045, socket Th .. 004. Impression
from casting 011 outer side of blade (E). usually found in Egypt, with one side passing over the
B 62. Figs. 103 and 105. L. .145, socket W .. 059, blade other. It is possible, if not probable, that the two ex-
Th .. 005, socket Th .. 003. One side of socket miss- amples from Gelidonya did not serve the same purpose,
ing. The blunt edge is very rounded and worn; this
as one is nearly twice the size of the other.
is possibly a worn example of Type 1 (M).
B 63. Figs. 103 and 105. L. .135, max. W .. 068, blade Th. B 66. Figs. 103 and 105. L. .124, blade W .. 026 to .03,
,007, socket Th .. 003. Most of socket missing (M). socket Th .. 003. Probably complete on ship; socket
B 64. Fig. 103. L. .lIS, W .. 06, socket Th .. 003. Most of chipped during removal of concretion (G). The
socket missing (1959). small size suggests that it was a socketed spatula.s"
B 65. L. .076, blade W.. 044, blade Th .. 005. Almost all B 67. Figs. 103 and 105. L .. 23, blade W .. 065, socket W.
of socket is missing, .057, blade Th. _008 to sharp edge. Very fragile,
found in several fragments, but possibly complete on
TYPE 4 ship (G).
---
These hoes, both from area G, are very long in rela- 28 Petrie, Tools and Wlnis., pl. XX: 46.
tion to their widths. The blade widens toward a very 29 Cj. Catling, 105-106, with pl. 9: i and j.
-

VOL, 57, PT. 8, 1967J BASS: THE BRONZES 91

o o
.. • 72
70

69

, '
_ :J ....
o 80
l :' .

B 68 81
84

78
oems • 10 20

FIG. 104. Bronze hoes and socketed blades: B 68 to B 84.

TYPE 5 the piece, I cannot say whether it was merely worn

The single example at Gelidonya is by far the largest down through use. Another Cypriot parallel appears in
of the hoes. Its long, broad blade has straight sides the Mathiati Hoard." The only example at Gelidonya,
leading to a blunt, almost straight edge. It is in frag- from area G, is missing only chips from the blade and
ments, but the pieces indicate that it was complete in part of the socket, and may have been complete on the
ship.
area G. Its closest parallel is a very sharp tool in
Nicosia (Cypr. Mus. 1958 VI-24 8) from the Stylianou E 69. Fig. 104. L . .195, max. W . .104, socket W .. 05,
Hoard.30 socket Th .. 004. (G). The blade is incised, near its
junction with the socket, with a sign which is found
E 68. Fie's. 104 and 105. L. .316, blade W .. 084, socket W . in Cypro-Miucan.
.05°6,blade Th .. 005, socket Th .. 005. (G).
TYPE 7
TYPE 6
The blade is spoon-shaped and quite 5111all in relation
The large blade flares out to form a wide, rounded to the size of the socket. The example from Gelidonya
edge. A similar hoe £ro111 the Founder's Hoard at
may be only a badly worn and often resharpened exam-
Enkomi lacks the convex edge; 31 without examining
ple of Type 2, but this seems doubtful. Its closest
30 Ibid., 80, no. 12, with pl. 4: d. parallel seems to be from thirteenth-century Tell el
31 Schaeffer, Enhomi-Alasia, 29, pJ. LXV: 2; Catling, 79,
no. 1 ; Deshayes II, no. 3036. 32 Catling, i9, no. 2, with pl. 52: 8.
 l;"~ ••••
B60

"". -=-~-=-
B49
,'.....
B62

em,•• _, •
B63

{",'- - -
B54 -. _'M' __
B66

(.;B-. ~- ••
B55

_~C!"'._ B68
-
-- ...B59
FIG. 105. Bronze hoes and socketed blades.
_e"'._
879

92
 VOL. 57, PT. 8, 19671 BASS: THE BRONZES 93
Hesy, although no dimensions are published for that
piece."
B 70. Fig. 104. L. .13. max. W .. 062, socket W .. 045,
blade and socket Th .. 004. Blade and socket chipped
(P).

Miscellaneous sockets ond socket frag/llents) probably from

hoes:
B 71. L. .142, W .. 058, Th .. 005 (P).
B 72. Fig. 104. L. .122, pres. vv ..
035-.038, Th .. 0015 to
.006. Shows no trace of curving out to form blade;
extremely long (S).
B 73. L. .10, W .. 044, Th .. 004 (G).
B 74. L. .08, W .. 05S. Very heavily concreted (M).
B 75. L. .075, W .. 048, Th .. 0025 to .007. Fragment (G).
B 76. L. .085, W .. 045, Th .. 006. Fragment (E).
B 77. L .. 035, W .. 04, Th .. 004. Fragment (P) .
•
Miscellaneous blades and blade fragments, probably from
hoes:
B 78. Fig. 104. L. .18, W .. 094, Th .. 005. Flat; rounded
at one end, but broken off just at socket at other end.
Possibly from an extremely large, but relatively
short, hoe of Type 2 (E).
B 79. Fig. 105. L. .033, 'vV.. 034, Th .. 005 thinning toward
edge. Rounded edge of very small hoe-like tool (P).
B 80. Fig. 104. W .. 07, Th .. 008 thinning toward edge.
Rounded tip of blade (P).
B 81. Fig. 104. L. .103, W .. 065, Th .. 006. End of large
blade, similar to those of Type 1, but with blunt end FIG. 106. Possible use of socketed "hoes."
(P).
B 82. L. .03, 'V .. 061, Th .. 007 thinning to edge. Similar vVe have noted parallels for the various types from
to B 81 (P).
B 83. W .. 056, Th .. 005. Rounded edge of blade, probably Greece, Cyprus, and Palestine. The hoes in Palestine
of Type 2 (M). are related to Mycenaean deposits," but at a time when
B 84. Fig. 104. L. .088, W .. 075, Th .. 005. Probably blade Ugaritic influences were strong in Palestine, so and cer-
of hoe as large as B 68; trace of socket preserved tainly Ugaritic hoes, whether of the fourteenth century 31
(1959).
B 85. L. .045, W .. 04, Th .. 007. Probably from junction of or later," are similar to those notecl above. Deshayes
socket and blade of large hoe, similar to B 68 (G). suggests Syrian influence on the manufacture of such
B 86. L. .027, pres. W .. 05, Th .. 007 thinning to edge. hoes, for the open-socketed method of hafting is almost
Very fragmentary, but possible blade from hoe of unknown in the pre-Hellenic repertoire;" and Carling
Type 2.
also sees a Near Eastern origin for the hoes.!"
B 87. L. .021, 'vV.. 065, Th .. 005 thinning to edge. Prob-
ably blade of Type 2 (G).
B 88. L. .135, W .. 06-.077, Th .. 001 to .005. Large blade, III. SOCKETED TOOL
worn on one side of edge, possibly of Type 4.
B 89-B 94. Six fragments of curved metal preserve so Although this is similar in shape to several of the
little of their original shapes that one may judge socketed tools catalogued under Hoes, it differs in that
only that they were parts of socketed tools which the socket tapers toward the blade, ancl the handle was
were neither unusually small nor large when com- riveted; both would have preventecl the handle from
pared with the major-ity of the hoes. Two were fr0111
area, G, three from P, and one from M. slipping forward, and indicate that some pressure was
put on the handle during use of the tool. A somewhat
CONCLUSIONS similar too! of the same size is from Gaza. H

There can be little doubt that most of these were B 95. Figs. 107 and 108. L. .l l, blade W .. 02, socket D.
.025. Rivet holes (D .. 005) pierce socket. End of
agricultural implements, but whether they were plow- __ --'b:.:l"'ade
bent and chipped (G).
shares or hoes remains a question." At least some
S5 Deshaycs I, 139-140.
must have been hoes, mounted on crooked sticks (fig. 361b-id., and Albright, Archae. of Potcst.ne. 187.
106). 37 Schaeffer, Ugaritica III, 254-255; 263, fig. 227; and 267,
fig. 232.
33 Bliss, Tell el Hesv, 105, with pl. 4, no. 187; Deshayes II, 38 Catltng, 82.
no. 1195. an Deshayes I, 139.
34 Desborough, Last Mycellaealls, 48; Catling, i9--80; G. E. au Catling, 81.
Wr-ight, Bib/Arch 6 (1943) 35. 41 Petrie, Gasc II, pl. XIX: 265; Deshayes II, no. 1213.
94 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. AillER. PHIL. SOC.

/ 1-
j 0 (~
( -

-
~\

t -
B 95
~
)((
\~lJ
r ...
~

I
J
,J,
!
LJ \1

o
96 99

C~···:JI 101

96 ~
I

<I ,

I 102

I-
I

/
97

o<ms 103
FIG. 107. Bronze tools: B 95 to B 103.

IV. SHOVEL V.MATTOCK

The chipped edges give the blade a rounded appear- The large blade is nearly duplicated in iron at Gerar,"
ance, but it seems likely that it was originally more which indicates that the shape was known in the Near
rectangular and was, therefore, quite similar to an un- East at least at a later date; I know of no bronze
dated shovel from Cyprus." parallels.

B 96. Figs. 107 and 108. L. .185, socket D .. 063, Th .. 004 E 97. Figs. 107 and 108. Pres. L .. 145, W . .123, Th .. 003.
(G). Socket and shoulders missing; slightly curved blade,
nearly rectangular but with rounded corners (M).
42 Deshayes II, 110. 3035; Catting, 78, no. 4, with pl. 3: c; ---
Gjerstad, sw« Preh. Cypr., 241,237. 43 Petrie, Cerar, pl. LXVI.
 VOL. 57, PT. 8, 1967J BASS: THE BRONZES 95
VI. PRUNING HOOKS B 102. Figs. 107 and 108. Pres. L. .115, blade W .. 06, Th.
The small size and delicacy of the pruning hooks, or at socket .055. Half of double axe, broken through
socket; small depression inside socket runs into
bill hooks, indicates that they were used for cutting blade (E).
brushwood. They may also have been used on vines, The shape conforms with that of Deshayes'
but the pointed blades of the two complete tools from double-axe type A,51 with nearly parallel top and
Gelidonya lack the curved "gathering-in shape" 44 found bottom, only slightly rounded blade edge, and
round socket; the abrupt thinning of the blade
on later vineyard hooks; the unfinished blank (B 219), just after the socket further places this under his
however, has a slightly curved pointed blade. Each has sub-type Ab.52 The closest parallel, published after
a wide chopping blade perpendicular to the pointed Deshayes had compiled his list, is from the twelfth-
blade. century Weapon Hoard at Enkomi; 53 also similar,
but less alike, is an axe "of Minoan type" in the
B 98. Figs. 107 and 108. L. .115, socket D.. 022, Th .. 0045. British Museum.s-
Heavily concreted and very fragile; damage to B 103. Figs. 107 and 108. Pres. L. .105, W .. 048, Th. at
socket may be modem. Almost rectangular chop- socket .042. Half of double axe, broken through
ping blade; straight pointed blade (G). socket; small depression in socket as in that of
B 99. Figs. 107 and 108. Pres. L. .148, Th .. 004. Both B 102. Similar to B 102, but slightly smaller and
blades and socket badly chipped. Rivet holes in with more rounded, Raring edge of blade (1959).
• socket (G). B 104. Fig. 108. This is one of the two objects from the
wreck which have disappeared since being carried
There can be almost no doubt that the pruning hooks to America as souvenirs in 1959; the other pieces
were made in Cyprus. Not only are their best parallels were returned to Turkey in 1962 by the Ccchrans.
found there," but also two molds for making them; 46 It is not possible to identify its type without know-
one of these molds would have produced blanks almost ing the shape of the socket.t"
exactly like Gelidonya B 219. Study of previously found axes is of little help in
assigning dates and places of origin to the Gelidonya
VII. SICKLE tools; axes of Deshayes' type Ab were known only from
B 100. Sickle fragment. Figs. 107 and 108. Pres. L. .085, undated examples from Naxos 56 until the recent pub-
W .. 024, Th .. 002. Recurved tip broken, socket or lication of the Cypriot piece. Axes similar to B 101,
tang missing (G). from the Athenian Acropolis Hoard" and the My-
Whether this was attached to a handle by rivets, a cenaean Poros Wall Hoard," are neither exact dupli-
socket, or a plain tang, its closest parallels may be found cates nor of certain date.
in the Mathiati and Styliauou Hoards from Cyprus."
The recurved blade is also found on a narrower My- IX. ADZES
cenaean example from the fourteenth to the twelfth
Of the complete adzes, or flat axes, only B 105 is not
century."
lugged. All were originally quite similar in size, except
VIlI. DOUBLE AXES for B llO, B 112, B 114, and B 124, which were larger;
Two complete double axes and halves of two others if B 124 is truly part of an adze, it would have been the
were found: widest.
B 101. Figs. 107 and 108. L. .175, blade W .. 06, interior
of socket .045 X .017. Complete except for modern Plain adzes:
chips on edges. Blade tapers evenly from max- BIOS. Figs. 109 and 110. L. .15, W .. 048 tapering slightly
imum thickness to the two wide, rounded cutting to head, max. Th .. 008. Complete; blade may have
edges; blades flare out in even curve on top and been worn and resharpened. and it does not widen
bottom; central groove on one side of blades; as do most of the others (E).
biconvex socket (P). B 106. Pres. L .. 072, W .. 04, Th .. 0055. Head missing
This is of Deshayes' double-axe type BIb, which (E).
may have originated in Crete, but which is espe-
cially prominent in mainland Mycenaean Greece.w
I find no exact parallels, for axes which are identical Lllgged adzes:
in profile usually have oval, rather than biconvex, B 107. Figs. 109 and 110. L . .178, max. W .. 047, W. at
sockets; biconvex sockets appear most often on lugs .043, W. at head .025, Th .. 005. Complete
axes of Deshayes' type B la,50 whose tops and (M).
bottoms are less concave.
------'=
44 Petrie, Tools and Wims., 47. " iu«, 255,II, p. 106.
4:;Catling, 85, with fig. 8: 5 and pl. 5: 11,with riveted handle. 52 Ibid., 256, II, nos. 2042-2044.
46 Ibid., 272, no. 2, with pl. 50: b; SGE III, 665, fig. 374, and 58 Catling, 88--89, no. 2, with fig. 9: 1 and pl. 6: 111.
Catling, 274, no. 1, with pl. 50: f. "c. F. C. Hawkes, BSA 37 (1936-1937) 145, fig. 2,2.
47 Catling, 83, no. (b) 1, with fig. 8: 2 and pl. 5: c; 84, no. 1, 55 S. Waterman, Explorer's Lowmol 38, 3 (Oct. 1960) 31,
with fig. 8: 3 and pl. 5: f. with figure.
"BSA 48 (1953) pl. 2: b ; Deshayes II, no. 2796. 56Deshayes, loco cit. (supra, n. 52).
49 Deshayes I, 257; II, pp. 107-108. 57Montelius, Grice precl., 153, figs. 485--486.
50ibid. "BSA 48 (1953) pl. 2, c; Stubbings, BSA 49 (1954) 296.
_CMS ....

895

,c..... _. _CMS_

899 8100

t,••• ~
__ ••
896

8104

_ ....- - -
8102

8103

FIG. 108. Socketed tool, shovel, mattock, pruning hooks, sickle, and double axes.
96
 VOL. 57, 1''1'. S, [967] BASS: THE BRONZES 97

c:==========-=.
c J B 105
110
ID
123

107
•

108 118

c:::::::===;========
C0
~
109 122 127

o cms 10 20

FIG. 109. Bronze adzes and axe-adzes: B 105 to B 127.

B 108. Figs. 109 and 110. L. .17, max. W .. 045, Th .. 004. B 119. Pres. L. .032, W .. 042, Th .. 004.
Complete; end broken off and found separately (M). B 120. Pres. L. .056, pres. W .. 054, Th .. 007 tapering to
B 109. Figs. 109 and 110. L. .21, W. at edge .06, W. at sharp edge.
head .02, Th .. 006. Complete. Wood attached to B 121. Pres. L. .047, pres. W .. 047, Th .. 009 thinning to
head is not necessarily remains of a handle. Head sharp edge.
bent; blade edge slightly chipped (P). B 122. Fig. 109. Pres. L. .075, W .. 05, Th .. 008. Ragged
B 110. Figs. 109 and 110. Pres. L . .15, max. W .. 064, Th. edge of blade (P) .
.008. Missing head (G). B 123. Fig. 109. VIf.. 028, Th .. 003. Head broken from
adze of same size as B 108 (IV1).
Fragments of either lHgged or plain adzes: B 124. Fig. 109. Pres. L. .04, W.. 067, max. Th .. 006.
Broken at both ends; this is the widest of the blades
B 11L Figs. 109 and 110. Pres. L. .05, W .. 033 to .04, Th. (G) .
.007. Probably head of adze, bent as B 109. Sign
inscribed on one side (5). The adze, or flat axe, was widely spread during the
B 112. Fig. 109. Pres. L. .06, W. at edge .065, Th .. 008. Bronze Age;" but the lugged adzes from Gelidonya are
Blade fragment; edge chipped (P). all of Maxwell-Hyslop's Type II, which is most com-
B 113. Pres. L. .055, pres. W.. 05, Th .. 0074. Body frag- 1110n in Palestine." Of lugged adzes of that type, how-
ment of plain adze (P).
B 114. Pres. L. .088, pres. W .. 059, Th. .004 to .006. ever, the closest parallels of the Gelidonya adzes are
Edge unevenly burred, as if it had been used on Syrian, front the fourteenth or thirteenth centuries."
metal rather than wood. with but one Palestinian parallel, possibly of the same
B 115. Pres. L. .063, W .. 036, Th .. 006. Blade fragment
with very little spread at edge (G). 59 R. Maxwell-Hyslop, Iraq 15 (1953) 70, map; Schaeffer,
B 116. Pres. L. .054, W .. 048, Th .. 008 (M). Uqaritica III, 274.
B 117. Pres. L. .105, pres. W .. 045, Th .. 0035. Chipped on 00 Iraq 15 (1953) 74-75,81-82.

both sides (E). 61 Schaeffer, Uoaritica III, 261, with figs. 234, 235, 236.
B 118. Fig. 109. Pres. L. .08, W. at edge .05, Th .. 007. Deshayes II, nos. 1080 and 1081 are undated, from a collection
Blade broken obliquely (1959). in Beirut.
 • •
I • • •
~

I • • II
Ic •• • •
•
•s
e •
•s •
c
•s
c
•s

•
Bl05 11107 Bl08 Bl09

_CMS _

B 127

~~ - -"
B110 B 111

c., -._.- r- _

CMS -~

B 126 B 125
FIG. 110. Adzes and axe-adzes.
98
 VOL. 57, 1'1'. 8, 1967] BASS: THE BRONZES 99

date." The appearance of the two more ubiquitous

plain adzes (B 105 and B 106) is not unexpected, for
similar pieces are found together with the Syrian 1ugged
adzes."
The possible use of these tools has been discussed.?'
and they have sometimes been called flat axes. It seems
most likely that at least those with lugs were adzes, and
that these lugs facilitated their being attached to handles
in the manner shown on Egyptian tomb-paintings 6~

(fig. 111).

X. AXE-ADZES

A complete axe-adze came to light 111 area P, with

fragments of two others in area G. All are noteworthy
in that in each case the base of the adze blade seems to
• be wrapped around the socket, as if it were attached to a
single-edged, shaft-hole axe.

B 125. Figs. 109 and 110. L. .108, ext. D. of socket .023, FIG. 111. Probable lise of lugs in attaching handle to adze.
into D .. 019. Small, complete piece, with edges still
sharp. The adze blade flares slightly at its edge, The method of attaching the adze blade, typical of
and the bottom of the axe blade curves downward Deshayes' subtype B 4,~1appears earlier at Vaphio, on
(P).
B 126. Figs. 109 and 110. Pres. L. .065, ext. socket D. an otherwise unsimilar piece," which may suggest a
.025. Axe blade almost entirely missing; adze blade Greek mainland influence. The axe-adze is, in general,
reconstructed from fragments. The adze blade far too widely spread 73 for further suggestions. but it
flares slightly at its edge, and is attached to the does seem singularly rare in the Syro-Palestinian region.
socket at the same angle as is seen in B 125 (G).
There is the doubtful possibility that this originally Schaeffer has shown that their spread covers areas
lacked an axe blade, as does an example in the rich in mineral resources." and variations were cer-
T'resor de Bronzes at Enkomi.w tainly used in mining; the delicacy of the Gelidonya
B 127. Figs. 109 and 110. Pres. L .. 078. Adze blade and tools would indicate rather that they were for the less
most of socket missing, and axe blade is chipped
heavy work of carpentry, in which the axe chopped and
(G). Socket seems to be the same size as that of
B 125. The axe blade, although spreading slightly split, and the adze squared planks."
toward its edge, does not curve downward as did
the two preceding. XI. CHISELS
B J28. Pres. L. .029, pres. W .. 048, max. Th .. 008. A frag-
ment of: the edge of an axe or an adze, from an Four chisels, of four different types, were found; two
unknown part of the wreck , it is possibly from a small fragments are probably also from chisels, and at
double axe. least one of these might represent a fifth type:
Although larger than B 125 and B 126, the closest
B 129. Figs. 112 and 113. L. .085, W .. 024, Th .. 008.
parallels are all from Cyprus; at Enkomi, the Gunnis Burred head; sides converge slightly toward cutting
Hoard,": the Foundry Hoard." and the Tresor de edge (M). This is a cold chisel or wedge 7G of
Bronzes 09 contained very similar axe-adzes, and an- Deshayes' sub-type B 2,'7 of which there is an
other was excavated at Pyla (Kokkinokremmos ) along undated example at Byblos.t" Deshayes' type B 1,
with parallel sides, is so similar that we should also
with Myc. IIlCl pottery. TO The differences, such as consider its widespread examples which are con-
the higher attachment of the aclze blades to the sockets centrated in the eastern Mediter-rauean : Syria (Ras
on all but the piece from the Gunnis Hoard, are minor Shamra) 70 and Palestine (Megiddo ) so in the four-
~~
when compared to those of other k110V'111 axe-adzes. 71 Deshayes I, 284-285.
G:! Guy and Engberg, Mcgiddo Tombs. 168, fig. 173: 6 and pl. 1"2 T'sountas, ArchIipt, (1889) col. 155, with pi. VIII: 2.
90: 1. (These are either from eighteenth-seventeenth or four- 73 Schaeffer, Enkonn-Atasia, 46-57, with map (fig. 5, p. 45);
teenth-thirteenth centuries B.C.) Deshayes J, 279-280; Catting, 92.
G3 Schaeffer, UgG.'l'itica III, toe. cit. (supra, n. 61). 74 Enkonn-Alasia, 48-49.
"Iraq 15 (1953) 69, 71; Deshayes I, 126-129. 75 Deshayes I, 289; R. \V. Hutchinson, PPS n.s. 16 (1950)
05 Norman de G. Davies, Tomb of Rekh-mi-rec, pis. LII, LV. 62; Catling, 91.
OG Schaeffer, Enkonii-Alasia, 39, no. 3, with fig. 1, no. 3, and 76 Catling, 96.

pl. III: 1 and 2. 7'{ Desbaycs I, 87.

et Catling, 91, no. 3, with fig. 9: 9 and pl. 8: c. 18 Duuand, Bvblos II, 169, fig. 173, no. 8091; Deshayes II,
6S Ibid., 110S. 4 and 5, with pl. 8: d and e. no. 717. '
60 Schaeffer, Enhonii-Atasia, 39, no. 2, with fig. 1, no. 2, and jl) Schaeffer, S'jwia. 18 (1937) pl. XIX; Deshayes II, no. 71I.
pl. III: 5 and 6; Catting, 91, 110. 2, with pl. 8: b. 80 Guy and Engberg, M cgiddo Tombs, pl. 127:7; Deshayes II,
70 Catling, 91, no. 1, with fig. 9: 8 and pl. 8: a. 110.710.
 100 CAPE CELIDONYA A BRONZE ACE SI-IIPWRECK [TRANS. A~JER. PHIL. SOC.

i-_ .- ~i 132
•
rz:
~l.~ .

I
~
~"'.-.

r
-U I
(

II'~i'il.'
- .
I

B 129 U r
I I
135

130 131 ~
133

;)
~' ~ ·'fr
,,0
• ~ • • v-a •
).
~~_../"-~

I ~
.~
~-~
Gil

OJIIJ i
~ ~
139
141 143
@
144
146 147
140 142

136 138
137
Oems 10 20
FrG. }]2. Bronze tools: B 129 to B 147.

teenth and/or thirteenth centuries, Cilicia (Mer- of the square bar soon led to deepening it, to gain
sin) 81 in the thirteenth century, Cyprus (En- strength for levering out the chips from the narrow
komi) 82 in the twelfth century, and Crete (Agia slits of the mortise holes." 86 Pieces similar to the
Tr-iadha ) 83 of uncertain elate; a similar chisel, but Gelidonya example are widespread, appearing in
with splaying edge, appears in fourteenth-century fourteenth-century Egypt 87 and Syria,88 twelfth-
Egypt.s" century Cyprus.w and Crete; 90 the Cypriot parallel
B 130, Figs. 112 and 113, L .125, W .. 015, Th .. 004. seems to be the most similar.
Very corroded, but probably with near-ly parallel B 132. Fig, 112. Pres. L .. 045, W .. 01, blade Th .. 002.
sides coming to blunt point at head, which may have Probably tip of small socketed chisel (M). It is
been inserted into a wooden handle (G). Com- similar in size to an undated Cypriot chisel."!
parisons for such a simple shape are almost mean- B 133. Figs. 112 and 113. L. .034, V';.. 019 widening to
ingless, but a similar tool is from Megic1do, in the .021 at edge, Th. .01 tapering to .002 at edge.
thir-teenth or twelfth century.ee Wedge-shaped fragment; probably chisel end
B 131. Figs. 112 and 113. L. .115, W, .012, ri, .004, pres. (1959),
L. of wood .04, max. pres. W. of wood .014. Head B 134. Fig. 113. L .018, W .. 016, Th .. 006. Probably
of chisel was inserted into wooden handle, part of rounded edge of some type of small chisel (S).
which is preserved (G).
This is a deep-bar chisel, in which the cutting Vle may conclude that the dating and place of manu-
edge is perpendicular to the plane of the wider facture of these chisels is uncertain, but that the good
faces of the blade. Petrie has described its usc; Cypriot parallels for B 131 and B ] 32, both rather dis-
"The chisel was not only used for cutting but for tinctive in shape, point to Cyprus.
levering out the pieces cut; this was especially
needed in the large amount of mortise cutting. done So Petrie, Tools and WpIiS., 20. See also Deshayes I, 101-
for the joining of planks edgeways. The weakness lOS.
tl7 Tbid .. pI. XXII: 85 and 86; the latter still with wooden
81 Car-stang, Preb: Mcrsin, 250, fig. 158: 10; Deshayes II, handle.
no. 713. sa Schaeffer, Ugaritica III, 261, fig. 233: 12; Deshayes II,
82 Schaeffer, Enkotni-Atasia, 44, fig. 3; 27; Catling, 96, no. I, 110.999.
with pl. 9: e; Deshayes II, no. 714. 8G Schaeffer, Enhomi-Alasia, 43, fig. 3: 2S; Deshayes II, 110.
S3 Deshayes II, no. ilS. 1000, with pl. XIT: 4, which is not an accurate drawing.
8+ Petrie, Tools and Wbn«, 20. no. 88, with pl. XXll: 88. uo Desbayes ] I, no. 998.
so Loud, M cgiddo II, pl. 184: 16, i!l Carling, 98, 110. 3, with fig. 10: 9.
 I
f
I

I -
I il I
I t
J

I I
,I
I t
eM>
I
eM'

• I B 133 BI34
I:t
CM>
tM$

•
BIt,l9 BI30 B 131

____ CMI

B 135

____ CMS

B 136

I I I
I
~
I
! I
CM.
CM'
I
I
CM'

B 143 B 144 B 146 B 147

FIG. 113. Bronze chisels, anvil, swage block, awls, punch and needle.
]01
102 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRA},·S. A:'1ER. PHIL. SOC.

XII. HAMMER OR ANVIL B 138. Figs. 112 and 113. Pres. L . .10, D .. 003. Squared
tang. Point broken (M).
The single fragment, which has been called an anvil B 139. Fig. 112. Pres. L. .03/, .004 X .004. Fragment;
by Carling (p. 107), is approximately the same size as square in section (G).
a hammer with round socket from the Enkomi Foundry B 140. Fig. 112. L. .068, .004 X .005. Rectangular in
Hoard." It is so similar to some types of modern ham- section (P).
B 141. Fig. 112. Pres. L. .043, D, .005. Round section
mers that 1 would find that identification irresistible (M).
were it not for the socket. Although the socket is partly B 142. Fig. 112. Pres. L. .065, D .. 01. Round section
missing, enough is preserved to show that only an ex- (M).
tremely thin and, therefore, weak handle could have B 143. Figs. 112 and 113. Pres. L. .05! socket D .. 01.
Socketed awl with sharp point; socket chipped and
been inserted. It is possible that rather than a handle, partly missing (P). Cf. Cypr. Mus. Met. 800.
a thin piece of wood ran completely through the socket B 144. Figs. 112 and 113. Pres. L. .053, W .. 011. Socketed
in order to prevent the tool from sinking into the ground point; socket oval in section (·~vr).
if it were used as an anvil." B 145. L. .043, .006 X .006. Fragment of square nail (1V1),
B 135. Figs. 112 and 113. Pres. L. .084, H .. 032, W .. 022.
Broken through socket; half missing. Oval, burred X.V. PUNCH
face preserved; rectangular socket for very thin B 146. Figs. 112 and 113. L .. 05, D .. 012. Hollow,
haft (P). cylindrical punch with solid, burred bead; opening
for cutting is square (.005 X .005). Small holes
XIII. SWAGE BLOCK corroded through sides; open end heavily concreted
B 136. Figs. 112 and 113. L. .10, W.. 038, H. .Q38. Four (G).
grooves of graded sizes 011 each of two opposite Powdery material removed from the interior was
sides. Plain sides are pierced by two tapering holes analyzed in the hope that it might give a clue as to the
which run through the block; the larger hole is
.015 in diameter, tapering to .006, and the smaller material on which the punch was used (possibly leather
is .10 tapering to .004. One end is in the form of a or metal foil), but it proved to be corrosion products
very large groove (E). fr0111 the punch itself. I cannot suggest for what pur-
The variously sized grooves were used for drawing pose such a large hole was needed,
out pins and similar objects by hammering." The con-
cave surface at one end could have been used to form XVI. NEEDLE
open sockets, and the size would have produced small B 147. Figs. 112 and 113. L .. 06, max. Th .. 0015. Hook-
sockets of the size found on the two pruning hooks ended pin, pointed at one end; very corroded (M),
(B 98 and B 99). The use of the tapering holes pre- It seems probable that the hooked end is the broken
sents more of a problem; I originally stated that they remains of an eyelet fanned by bending the top
dOWTlas mentioned by Carling (lOS, with fig.
were for hammering heads on rivets, but there is the
10:14).
possibility that they were dies for drawing out wire of
soft metal. Cyril Smith has written 95 of the great XVII. KNIVES
power needed for drawing out wires of even soft copper
through holes of such small diameter, but I would sug- The knives were in very poor condition and almost
gest, without experimentation, that the copper might all that may be said of them is that 1110Stwere single-
have been hammered through the holes; if this is pos- edged:
sible, the tapering holes would be the earliest known B 148. Figs. 114 and 115. L. .17, W .. 015, Th .. 003 to .001
dies for such work. A stone swage block comes from or less at tip. Blade chipped slightly. Flanges on
a late twelfth-century level at Enkomi." both sides of haft; blade slightly wider than haft,
with nearly straight back, but slightly convex cut-
ting edge (M).
XIV. AWLS AND NAILS(?) Although the flanges are not deep, the absence of
B 137. Figs. 112 and 113. L . .125, D .. 004 to .002. Cor- rivets indicates that the handle was held in place
roded in middle at point of smallest diameter; tapers by the flanges; the knife is, therefore, of Sandars'
__ --=sc...lightly
at both ends (1959). Type 2,91 which is especially prominent at Mycenae.
B 149. Figs. 114 and 115. Pres. L. .082, W .. 009, Th.
ua Schaeffer, Enkomi-Atasia pl. LXIV: 5; Catling, 100, pI. .0005. Back curves downward to meet straight,
11: c. sharp edge at tip. On either side of blade two in-
93 Petrie, Tools and WpIiS., pl. LXXVrIT, nos. 49-52 (Roman cised lines follow curve of back, ,003 below top and
examples). .001 apart. The haft is missing, making the knife
!H The object was identified by Paul Shaw after I had guessed difficult to classify; similar decorative bands are
that it was used for metal-working by its similarity to a small found On knives of Sandal's' Types 1, 2, and 3.
anvil from Great Britain, described by John Evans, Anc. B 150. Figs. 114 and 115. L. .127. W.02. Th .. 004 at haft
Bronee implelJlcnts, 182-183, with figs. 217 and 218. See also to .002 at pointed tip. Tip broken. Triangular
Coghlan, Prehist Mct. of Copper and Bronec, 77 and fig. 13. haft; straight back; concave blade. No rivet holes
9:> Letter of 12 March, 1964, in which he suggested the possi-
__ ("--M).
ble use of the concave end of the block for forming sockets.
DO Catling, 275, with pl. 51 : d and e. "N. K. Sandars, PPS 21 (1955) 179, with n. 1.
 -~\
VOL. 57, PT. 8, 1967] BASS: THE BRONZES 103

'l -
11,,--
- -
I
r\_~ -elll 'l )

-- •

iT ct~
i

(:'1
I
i

I
r
I
U < .'
\UiI \i ..
~i
!
I
lJ
1 0 Q
• I

0 -i~-/'--
I,

) \ II
U
,
lSi
\I
J
152
~!
153 154 155 ,I
156
~
I
157 158 159 ~ IJ
160
B 148 149 150

• o
.. I ,
.. {17
j

":",'1
:1
-
- --- -- /- -)
~
;1

163
I
164
165
167 168

161 162
Oems 20

FIG. 1I4. Bronze knives, spearheads, razor and spatula: B 148 to B 168.

B 151. Figs. 114 and 115. Pres. L. .075, W.. 022. Very nussmg. Possibly double-edged, but very worn
short blade tapers, with slightly convex sides, to (1vI).
rounded tip. Pair of rivet holes just above shoulder, B 159. Fig. 114. Pres. L. .052, W .. 022. Th .. 0015. Pos-
with possible third hole at break higher up; end of sibly worn blade fragment; slightly thinner on one
butt missing. Flattened rhomboid in section (P). of parallel edges (P).
Although the shape is basically that of Catling's B ]60. Figs. 114 and 115. Pres. L. .10. Seven fragments
daggers of type (a) from Cypms.w the extremely of ribbed knife; one r-ivet preserved (P).
short blade suggests that this was a knife for house- B 161. Figs. 114 and 115. Pres. L. .14, pres. W .. 02.
hold use rather than a dagger.?" Very badly damaged, but similar to B 160. with
B 152. Figs. 114 and 115. L. .105, W .. 013, Th .. 002 at hollow rib (G).
haft to .001 at tip. Midsection slightly thicker than
back and edge. Badly worn (M). The knives are not so distinctive as to be assigned
B 153. Fig. 114. Pres. L .. 073, W .. 014, max. Th .. 004. any certain place of origin.
Almost straight back, convex edge. Very worn
(P) .
B 154. Fig. 114. Pres. L. .076, max. W .. 025, Th .. 002 XVIII. SPEARHEADS
to .003. Short tang; back nearly straight, but
curves downward near tip; convex cutting edge The spearheads at Gelidonya were so poorly pre-
(1959). served that some of those catalogued below may be frag-
B 155. Fig. 114. Pres. L. .062, W .. 016, max. Th .. 0025. ments of daggers or knives. vVe may see, however, that
Tip of blade with downcurving back meeting both open and closed sockets were used:
straight cutting edge.
B 156. Figs. 114 and 115. Pres. L. .15, W .. 023, Th .. 005. B 162. Figs. 114 and 115. L. .147, W .. 02. Nearly paral-
Straight back. Blade very worn; made of four lel, slightly concave sides; rounded midrib; flat,
fragments, but missing both ends (G). triangular tang with two rivet holes at shoulder
B 157. Fig. 114. Pres. L. .048, W .. 02, Th .. 002. End (M). Such objects are usually called daggers, but
fragment; very worn (P). I feel that they were more probably spearheads.t''''
B 158. Fig. 114. Pres. L. .065, W .. 018, Th .. 022 at mid- B. 163. Figs. 114 and 115. Pres. L. .105, W.. 018, socket
dle, thinner on both parallel edges. Haft and tip D. .009. Badly concreted and chipped. Open
--- socket formed by wrapping butt of blade; no mid-
98 Catling, 125-126. ---"
Petrie, Tools asui Wbns., 30, mentions small daggers from
91l 100 Spearheads of similar shape are socketed, rather than
Spain which were found only in women's graves, indicating a tanged, but either method of attachment would have been used
domestic use f or the blades. on spearheads.
 - - - -- 1148

~.-- 1149
_CM'__
BI63

_CM'- __
B 150

_CMS __

B 151

_ 'M' __
BI52
_CMS _

BI66 -

_CMS _

BI60

tiir _
_ c.".
B 161
_ _ eMS _

BI67
FIG. 115. Bronze knives, spearheads, razor and spatula.
104
- _'1"
BI68
_
 voi.. 57, PT. 8, 1967] BASS: THE BRONZES 105

8 169
) .a.:;."'..?
...•.
44A"i::-'
'

183

170

18. ~-{
177

173
, 171

175
/ 185 )-~

178
•

17'
CD l]J
179
180
18'

187

Oems 20
FIG. 116. Bronze vessel, tripod and offering stand fragments, and spit: B 169 to B 187.

rib (S). Ct. Calling's spearheads 1 and 2 of type Classical times in Greece;':" but its thinness suggests
c, undated, from Cyprus (118, with fig. 13:8 and that it was a razor of unusual type. lOS
pl. 13 :d).
B 164. Fig. 114, Pres. L. .144, pres. W .. 023. Large,
hollow midrib leads into socket, which is missing; xx. SPATULA
blade edges badly chipped (G). Too badly dam- B 168. Figs. 114 and 115. L. .074, W .. 01, Th .. 003.
aged to be assigned to any specific type; socketed Small spatula with thickened handle; constriction
spearheads existed in both the Near East and the above oval blade (G). I find no parallels for this
Aegean earlier than on Cyprus, but they were being piece, which may have been an unguent spoon.
made on Cyprus by the time of the shipwreck.
B 165. Figs. 114 and 115. Pres. L. .105, D .. 022. Socket XXI. VESSELS
only; beginning of blade (Th .. 005) shows no mid-
rib (P). The vessels were extremely fragmentary; even if we
B 166. Fig. 115. Pres. L. .07, D .. 018. Similar to take into account the extreme thinness of their walls,
B 165. Divided socket ;101 fragment of wood pre-
which have in no case been substantially preserved, we
served inside. Blade is solid at break, with no
evidence of hollow midrib (S). must doubt that the rim and handle fragments repre-
sent complete bowls on the ship. The sizes of the frag-
XIX. RAZOR ments have made estimates of original diameters impos-
sible in most instances:
B 167, Figs. 114 and 115. L. .084, max. W .. 046, max.
Th between .001 and .002. Crescent blade with
Rim and rim. fragments:
concave sides, sloping shoulders, and tang; tang
chipped very slightly during excavation (G). B 169. Figs. 116 and 117. D .. 135, pres. H .. 01. rim Th.
.0015, wall Th .. 0005. Almost complete rim, proba-
The shape is similar to that of knives used in cutting ----cc=---c-
leather, from at least the fifteenth century in Egypt to 102 Petrie, Tools and Wpns., pl. LXII; D. B. Thompson,
Archaeology 13 (1960) 239; Water-er, Hist. of Tech. II, figs.
101 Catling, 125, 11. 1, explains that such sockets were cast, 111,125,166,167.
unlike the sockets of many of the agricultural tools, and were 103 Petrie, Tools and TIVpJlS., pls. LX and LXI, for razors
not rolled. of other types.
106 CAPE GELTDOJ\iYA: A BRONZE AGE SHIPWRECK [TR,ANS. AMER. PHIL. SOC.

&.,- • •
8169

_CMS_
8 177
,.-
8170

Q s

----_/
_5,,,,_
.eM'" _ •• 8 187
-
8 176
FIG. 117. Bronze bowl rim and handle fragments, and spit.
 VOL. 57, PT. 8, 19671 BASS: THE BRONZES 107
bly from hemispherical bowl; rim is slightly thick-
ened .St-curve at top of wall (P).
B 170. Figs. 116 and 117. Pres. L. .06, rim H .. 018, max.
rim Th .. 012, wall Th .. 001. Wall of vessel fits
into groove in bottom of thick rim, which was
separately made and which is oval in section (P).
Catling (p. 125) compares a similar piece from
Cyprus.
B 171. Figs. 116 and 117. Pres. L. .11, pres. H .. 018, rim
Th .. 003, wall Th .. 00075. Rim fragment curves
outward at one end as if to form a spout, but this
curve could represent damage to the fragment (M.).
The flat rim is thickened on its inner side and can
be duplicated in Cyprus 104 and the Near East.w-
B 172. An irregular fragment (.03 X .022) of a rim iden-
tical to the preceding one comes from area P; it is
possibly part of B 171.
B 173. Fig. 116. Pres. L. .07, Th .. 001. Fragments of '-------Hr- INNER
bowl with outcurved rim; badly mashed, but pos-
• sibly of larger diameter than B 169. Matting
RING
adheres (G).
B 174. Pres. L. .035. Possibly rim fragment (P).
B 175. Fig. 116. Pres. L. .03, pres. H .. 025, rim Th .. 003, FIG. 118. Rod tripod from Kourion in University Museum.
wall Th .. 0005 (P).

Handle fragments: B 182. Roughly .085 X .07, Th .. 005. Very irregular frag-
ment; slightly convex bottom of concave vessel
B 176. Figs. 116 and 117. Rivet plates for horizontal (P). This is possibly part of a mesomphalic bowl.
loop handle, which was round (D .. 008) in section.
Each plate is .05 long, .015 wide, and from .003
XXII. TRIPOD AND OFFERING STAND
thick in the center to .002 thick at its edges. Two
rivet holes in each plate. CU1've of plate indicates FRAGMENTS
vessel was larger than the bowls mentioned above Small fragments of one or two tripods and an offer-
(P).
B 177. Figs. 116 and 117. Pres. L. .09, W .. 025. Double ing stand were scattered in areas G, M, and P. As
bar handle whose divided end forms attachment these could have an important bearing on the date of
plates which are mostly missing (P). Cf. a simi- the wreck, they will be discussed in some detail. Fig-
lar, but complete, handle from the Cypriot Stylianou ures 118 and 119 show, respectively, a later rod tripod
Hoard (Cypr .. Mus. 1958 VI-24 24), which Cat-
ling has compared to cauldron handles from Tiryns from Kourion and an offering stand from Megiddo,
and Tell Abu Hawan1.10U which should allow the reader to visualize the positions
B 178. Figs. 116 and 117. Attachment plate. L. .147, of the Gelidonya fragments on complete stands.
\-\1.. 025, max. Th .. 006. Single rivet hole (D .
.0045) at either end. Only stump of handle pre-
served; oval in section (.02 X .015). Made of
several joining fragments, most of which were
found in area G, but one of which was in area P.
This may be compared to B 177 in form, but the
attachment plate shows no curve; this suggests
that if the plate had not been hammered straight
at a later time, it came from a rectangular con-
tainer (the sponge divers reported a metal "box"
on the 'wreck; p. 15) or was attached vertically to
a cylindrical vessel.
B 179. Fig. 116. Pres. L. .034, W .. 016, Th .. 003. Curved
strip, possibly fragment of a very thin handle (G).
B 180. Fig. 116. Pres. L. .025, W .. 017, Th .. 002. Curved
strip, possibly fragment of handle of small cup
(M).

Bod')l [raqme-us:
B 181. VI/.. 02, Th. .002. Roughly triangular fragment
pierced by many pinholes (G). Possibly part of a
strainer.t'"

104 Catling, 147, with fig. 17: 5.

105 Loud, Megiddo II, p1.190: 9, from 1150-1100 B.C.
100 Catting, 164, no. 2. FIG. 119. Offering stand from Megiddo (after
107 tu«, 161. M egiddo Tombs).
108 CAPE CELIDONYA: A BRONZE ACE SHIPWRECK [n<ANs. "\lER. PHIL. soc.

B 183. Ring fragment (fig. 116). H. .04, L. .036, Th. B.C.).'" Lastly, a stand from a XIX Dynasty building
.003. Short strip of cast ring. Three rope-pat- at Beth Shan has been given a tentative date in the
terned bands, framed and separated by four plain
bands; plain back (P). This finds its closest paral- first hall of the twelfth century by its excavator-s.t'"
leI in a rod tripod from Beth Shan which is tenta- Vle may conclude, therefore, that the surest evidence
tively dated to the first half of the twelfth century places the offering stands, which supported pottery
by its excavators.t'" bowls.>" in the thirteenth century B.C.
B 184. Volutes and ring fragment (lig. 116). H .. 035,
W .. 038, Th .. 002. Spacer and small parts of ring The evidence for rod tripods and cast or strut tripods
and leg preserved; one volute broken. Two verti- is less secure. In the most recent studies, Benson 120
cal rope patterns framed by pair of plain bands on feels that the rod tripod developed, in the twelfth to tenth
leg, which is bent in sharply from the volutes. This centuries B.C., from the earlier strut tripod, which he
is very probably part of B 185; both were in area
M. places in the Late Bronze Age /21 but Catling believes
B 185. Leg fragment (fig. 116). H. .05, W .. 017, Th. the rod tripod to be the earlier."? The unsure footing on
.0025. Double twist or double rope pattern on one which we tread is apparent in Catling's table of chronol-
face, as on B 184, of which this is 1110stprobably ogy and distribution of tripods and stands.F" Of the
a part (M). This is very similar to a rod tripod twelve Cypriot examples which he has placed in the
of uncertain date from the Pnyx in Athens.t'"
B 186. Offering stand fragment (fig. 116). H. .038, L. twelfth century, I feel that only one (Catling, no. I)
.085; ring oval in section: H. .006, W. .004. can be dated with complete confidence. Two of the
Volutes are attached to ring, with double leg rod tripods (Catling, nos. 7 and 9) are placed in the
rising above; molding of leg and volutes same on twelfth century because they are probably from the
both faces of fragment (G).
This was listed as a fragment of a rod tripod Enkomi Foundry Hoard, but his best evidence for
with a plain ring by Catling,l1O but the error was placing that hoard in the twelfth century 124 is the
based on the insufficient information with which I occurrence of the stands themselves and of "the latest
had supplied him. Later, in restudying the piece, form of ox-hide ingot" ; we have shown, in the chapter
I noted that if the ring was kept horizontal, the leg .
was directed at a fairly sharp angle toward the on ingots, that this form appears as early as the four-
center of the ring; this angle was not caused by teenth century. The dating of the three cast (strut) tri-
accidental bending. and could mean only that the pods from Myrtou (Catling, nos. 21,27,28) depends on
Gelidonya fragment was from the base of an a reinterpretation of the excavation report.':" in which
offering stand. they were believed to have been of the thirteenth cen-
We have, it seems, fragments of one or two tripods tury; Catling's reinterpretation is convincing, but dates
and one offering stand. The offering stand poses fewer reached in such a fashion must, necessarily, be subject
problems and will be discussed first. Such stands are to some question. The fourth cast tripod (Catling,
almost certainly Syro-Palestinian for, as Catling points 200, no. 23; Benson, 12, no. 15) he dates on the evi-
out,111 eight are known from the Near East, with but dence of three unseen Mycenaean sherds. Of the five
one in Cyprus and none in the Aegean. The Cypriot Cypriot four-sided stands which Catliug places in the
example cannot be accurately dated,112 but two very twelfth century, one (Cat1ing, no. 32) is dated by the
close parallels for the Gelidonya fragment, from Ras same unseen sherds ; one (Catling, no. 34) by a repre-
Sharnra 113 and Megiddo,'>' are elated to 1300--1200 sentation on it of an cxhide ingot, which we have
B.C. and 1350-1200 B.C., respectively, by the contexts shown cannot be assigned to the twelfth century alone;
in which they were found. Five other examples from two (Catling, nos. 38 and 39) because they are from the
Megicldo, only one of which has volutes.v" were origi- Enkorni Foundry Hoard, which we mentioned above as
nally published as being from the tenth century,"?" but not able to provide conclusi ve dating; and one (Catling,
these may also fall in Late Bronze II (i.e., 1350-1200 no. 37) because it is from the Stylianon Hoard, which
109Fitz Gerald, PEFQ (1934) 133 f., with p1. 7: 3; ]. L. has no independent means of being dated.
Benson, Grk, Rom. and Bvz. Studies 3, 1 (1960) 12, no. 14, For the Near East, the evidence is again uncertain.
with fig. 6, no. 14; Catling, 196, 110. 13, with pl. 29: b. Catling The dating of the rod tripod (Catling, no. 13) and the
has made the same comparison with the Celidonya fragment on offering stand (Catling, no. 50) from Beth Shan in
p.?ol/B~~~:l~:op. cit. (supra, 11. 108) 11, no. 12, with fig. 6, the twelfth century B.C.was considered uncertain even
no. 12; Catling, 194, no. 6, with pl. 28: a. by the excavators. To date the cast tripod from Ras
110 Catting, 192-193, 110. 2, with pls. 27: d and 37: b.
1.11.1bid 212 117 May, ~Megiddo Cutt, 19, n. 73.
11.2Ibid:: no. '48, with pl. 37: a. 118 FitzGerald, PEPQ (1934) pl. VII, fig. 2, and pp. 133-134;
113 Schaeffer, Enkomi-Alcsia, 65, fig. 18; Catling, 212, no. Catling, 212, no. 50.
49, with p1. 37 : b. 119 May, Megiddo CIt1t. 19.
114 May, Megiddo Cnit, 19 f., pI. XVII; Guy and Engberg, 120 Benson, op, cit. (supra, n. 108) 16.
M eqiddo Tombs, 188, fig. 186: 4, and pl. 119: 1; Catling, 212, 121 tu«, 13.
no. 51, with pl. 37: c. 12:lCatliug. 222.
115 Schumacher, Tell el-Muleselfim I, 86, fig. 118c; Catling, 123Ibid., 223.
213,110. 53. 121Ibid., 281.
116 \iVatzinger, Telf el-Mu/escfh'm II, 91. 1~5Ibid.} 200.
 VOL. 57, PT. 8, 1967] BASS: THE BRONZES 109

i>. ~
;, '< 'iO\
.-~V' []' ~~'·a"
,I
-
4

l111
"" ,/~)!
199 200 201 ~
'~:".. ::;)( "'~ "9" 193 194 197 202

',.7 J
.~J.'- ~ -
• 188 • 190' (:~. '.:.' • ()
O<m, 20 t:: •.• 203
~
FIG. 120. Bracelets, rings, hooks, and unidentified bronzes: B 188 to B 204 .
•
Shamra (Catling, no. 30) to the twelfth century, in- XXIV. BRACELETS OR ANKLETS AND RINGS
stead of the fourteenth, means reinterpreting Schaef-
B 188. Figs. 120 and 121 a and b. Exterior D .088, Th.
fer's excavation report.126 Only the four-sided stand .004. Approximately one-third missing. Round in
from Megiddo (Catling, no. 33) seems safely to fall section. Twisted wire knot at one end (P).
after 1200 B.C. on the basis of its context.':" This was found imbedded among the beads and ..... vas
The eleventh- and tenth-century tripods and stands undoubtedly carried in the same jar with them.
It is almost identical to a bracelet of adjustable
in Carling's table are all well dated by the contexts in
size from the Tresor de Bronzes at Enkomi, which
which they were found. would indicate that a second knot or terminal spiral
'iV e may conclude, therefore, that even the most thor- was on the missing portion.w- Schaeffer 132 and
ough study of bronze tripods and stands, of the types Herzfeld 183 have remarked on the great chrono-
under discussion, cannot lead to strong conclusions in logical and geographical spread of this type, but
of particular interest are examples from thirteenth-
our present state of knowledge; there is simply not century Megiddo 13.. and Ras Shamra,135 and
enough material which may be dated by stratigraphy twelfth- to tenth-century Troy.136 Catling points
and carefully documented tomb groups. Only the out that the type is not found in the Aegean.w"
thirteenth-century offering stands may offer a clue to B 189. Figs. 120 and 121. Fragment. L. .105, Th .. 005 X
.004. One end is shapeless and corroded, the other
the date of the Gelidonya ship, and that wreck itself tapering to a point. The tapering end suggests that
may provide new evidence for the dating of tripods. this may have been similar to B 190.
B 190. Figs. 120 and 121. L. .105, W. .06, Th. .008,
XXIII. SPIT Open bracelet with possible, but worn, serpent or
duck heads at ends; crushed into oval (M). Cf.
B 187. Figs. 116 and 117. L .505, W .. 015, W. of head Cyprus Museum Met. 1992 .
.021, Th. .003 to .002 at pointed end. Strip of
metal with parallel sides, pointed at thinner end, In addition, there were small fragments of bronze
with flat, roughly oval head at thicker end (1959). rings which may have been either open or closed; as
This has been called a blade from a pair of smithing in the Tresor de Bronzes at Erkomi.s'" these are round,
tongs by Carling;':" and Harold Burnham of the Tex- oval, or flattened on one side in section. It is not
tile Department of the Royal Ontario Museum has possible to say whether they were originally ornaments
pointed out its similarity to "a weaver's sword for or simply bullion shaped in a form common in the
use with a vertical warp-weighted loom of classical Bronze Age: 139
type. "120
B 191. Fig. 120. L. .07, Th .. 01. Tapered and flattened
r think that there can be absolutely no doubt that (M).
this is simply a spit, as I first identified it.1SO It is a B 192. L. .03, Th .. 009, est. original D. .042 (P).
well-preserved, unbroken piece, identical in all respects B 193. Fig. 120. L. .022, Th .. 014 X .009. Flattened on
one side (S).
to the spits used for making shish kebob in modern
Turkey. The metal is still quite pliable, and the bend 131 Schaeffer, Enhonu-Atasia, 43, 110. 21, with fig. 2, no. 21
is meaningless and probably accidental. on p. 41.
1$2 tu«, 63, 65-66.
1'20 iu«, 202. 133E. Herzfeld, fran in the Anc. East, 148.
121 May, Megiddo C1dt, 19 f., with pl. XVIII; Catting, 205. 13+ Guy and Engberg, M egiddo Tombs, 8, with fig. 179.
128Catling, 99 and 292. 1.~5 Schaeffer, Enkonu-Alasia, 65.
129 I thank 1',111'. Burnham for this suggestion, which he wrote 136 Ibid.,~alJd H. Schmidt, H. Sclitienianws Traj. Altert., 261.
to me after having quickly seen a picture of the object in an 137 Catling, 231.
illustrated lecture. 138Schaeffer, Enlw1II£-A kisia, 41-42, with fig. 2.
i so Ai A 65 (1961) 274. 1.".9 Herzfeld, op, cit. (supra, ll. 133) 147.
110 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK (TRANS. AMER. PilIL. SOC.

_C.' __ B 200

__ eMS __

B202

8 IBB b.

_cM
,

B IB9
_

_CMS __

B 203

.CM' •••
8190

_CM' __
8 199 ~

FIG. 121. Bronze bracelets and unidentified objects.

 VOL. 57, PT. 8, 19671 BASS. THE BRONZES 111

• III 1 C>
[2 -
--1 ,

I-
. ,

!;
!
~
~'/
I

I , 214

[J ~J
8206

I
Oems
207
208
~

I
20
U 210
< -/

212
I
V
213
~
215
•
21.

FIG. 122. Bronze attaching devices and unidentified objects: B 206 to B 216.

B 194. Fig. 120. L. .04, Th .. 008, est. original D .. 051. out much conviction, that it may have been some
• Round section (P). S01-t of locking device.
B 195. L. .032, Th .. 009, est. original D .. 048. B 204. Figs. 120 and 121. L. .056. W .. 035, Th .. 007
B 196. L. .048, Th .. 01 X .006, est. original D .. 074. Oval tapering to end, Roughly triangular object, with
in section, rounded point. A layer of even concretion was
B 197. Fig. 120. L. .039, Th .. 008, est. original D .. 057. cracked off, inside of which was a layer (Th .. 002)
B 198. L. .04, Th .. 014. Except that it is bent, this is of hard green material which evenly encased the
very much like a pike end fragment (G). object itself, which is very well preserved, with
B 199. Figs. 120 and 121. L. .061, Th .. 009 X .006. Oval end beveled up hom one side. The green material
in section. Tapers at one end; other end Hat-es has not been analyzed, but it was unlike any of
slightly just at break (M). the natural corrosion products which were found
on other objects.
XXV. HOOKS B 205. Fig, 123. Pres. dimensions .13 X .12, Th .. 0075.
Fragmentary pierced disk (E).
B 200. Figs. 120 and 121. Three hooks (?) tangled to- B 206. Fig. 122. L. .058, D .. 008. Rod, slightly thickened
gether. Each piece is metal about .005 in diameter, at either end.
and becomes thinner at either end, although the B 207. Figs. 122 and 123. L. .085, \IV .. 025 tapering
ends cannot be called pointed (P). I can suggest slightly toward one end, max. Th. .015, Thick
no use for these, but recall a similar bunch of strip of metal with rounded back; perhaps unfin-
hooks, each sharpened at both ends, from a much ished casting (P).
later date at Dcdoua.v-v B 208. Figs. 122 and 123, L. .06, max, Th .. 007, Semi-
B 201. Fig. 120. D. of loop .025, max. D. of metal .004. circular; hollow, with opening on straight side (P).
Small hook, very sharp at one end (P), B 209. Fig. 123. L. .043, .013 X .012. Bar, nearly square
in section, with oval knob on one side (G).
XXVI. UNIDENTIFIED BRONZES B 210. Figs. 122 and 123. L. .088, .03 X .04 at larger
end. Bar, rectangular in section but with beveled
These differ fr0111 unidentifiable scrap (group XXX) edges, wider at one end; both ends broken (1959).
in that they represent complete or fragmentary objects B 211. Fig. 123. Original D. approx .. 12, Th .. 002.
'which I cannot identify, but which have distinctive Fragmentary. Slightly concave, round plate, with
chips from edge of preserved portion (P). It is
shapes. possible, but far from certain, that this is a scale
pan of the type found in Cyprus, Greece, Crete, and
B 202. Figs. 120 and 121. L. .067, max. W .. 01, Th .. 006.
the Near East (Carling; pp. 162-163, 185); the
Probably bronze, but possibly lead. One straight
chips may represent suspension holes which have
edge with notch (slightly more than half a circle
broken through the edge.
of .0045 in diameter) in center; other edge curves
convexly to meet straight edge. No evidence of
XXVlI. ATTACHING DEVICES (TANGS
break; sides are even.
AND RIVETS)
B 203. Figs. 120 and 121. L. .065, W .. 037, D. of rod .01.
A short rod is attached (cast on?) at an oblique B 212. Tang. Figs. 122 and 123. L .. 14, Th .. 01 to .003
angle to one end of a roughly rectangular piece at wider end. Tapered. Although the tang is
which is deeply depressed with a /iV-shaped notch similar to those found on bronze mirrors, the pre-
all the face opposite the "handle": there is no served portion suggests that the blade was spatula-
evidence of a break, but there is no certainty that shaped (P).
it is complete (1959). I have found 110 explana- B 213. Tang. Figs. 122 and 123. L. .053, max. W .. 02,
tion for this object, although it resembles slightly Th. ,005. Same as preceding, but much smaller;
a piece in the Cypriot Mathiati Hoard which Cat- broken off obliquely at wider end. Wider faces
ling has labeled a late intrusion and identified as slightly inset. Very possibly from a bronze mir-
a "badly modelled ear with dowel for attachment ror,142 especially if the identification of B 225-
--~
___ :eta a life-size statue." 141 I hesitantly suggest, with-
142 Ci. Cypr. Mus. Met. 2506; see also the Cypriot examples
140 These are, or were. on display in the National Museum ill in Catling, pl. 40, and Schaeffer, Enhomi-Alasio, 181, no. 199,
Athens, but with no further identification. with pl. XLII: 2 (pierced tang); and Egyptian examples in
141 Catling, 283, with pl. 52, no. 26; SCE III, 665, fig, 374. Petrie, ou. of Daily Usc, pl. XXVIII,
 -,,,,- -8209

,., .~__c. .~.

8205

-,,,,- - - 8 210

8207 --
-,,,, -
8208
C"".--'- .~C• ~
8211
III

I II
I I I
II I I
I Ica s
I
I
''''
,..I 8 214

8213 8216
8215
8 212
FIG. 123. Bronze attaching devices and unidentified objects.
112
 VOL. 57, PT. 8, 19671 BASS: THE BRONZES 113

B 228 as bronze mirrors is correct. Tangs are not

found on Aegean mirrors, except on Rhodes.t-"
but are associated with the Near East, having

\
appeared as early as the Old Kingdom in EgypUH
B 214. Pierced tang? Figs. 122 and 123. Pres. L. .037,
pres. W .. 018, Th .. 006. Rivet hole (D .. 005)
near end. Only parts of two sides preserved.
B 215. Tang. Figs. 122 and 123. L. .036. Very small,
square tang, with part of blade preserved. Almost
certainly an arrowhead fragment (G).
B 216. Riveted tang. Figs. 122 and 123. L. .071. max.
'vV .. 019, Th .. 004. Flat tang, pierced by two ri vet
holes, .016 apart and each .005 in diameter, tapers
into round bar (D .. 01), which must be part of
tool itself.
B 217. Riveted plate. PI-es. L. .039, pres. W .. 033, Th .
.007. Triangular fragment of plate with two rivet
holes: one is .004 in diameter, the other .002; the
former may have been enlarged through corrosion .
• B 218. Rivet fragment or stud. L. .012, max. D. 015.
Conical head; pin is .004 in diameter (G).

XXV!!!. UNWORKlm CASTINGS

FIG. 124. Hypothetical casting of pruning hook in open mold.
Many of the implements, such as the flat, open-
socketed hoes and pruning hooks, were cast either in B 224. Metal strip. Figs. 126 and 127. L. .28, W .. 038,
open or multiple 111olds,l1Gand sockets and edges were Th. .008. Thinner at one end, which is bent
formed by later hammering. When open molds were (1959)
used (Jig. 124), flat stones probably were used as The following pieces, which are the same size as the
covers to prevent rough surfaces, which are caused by disks from the Athenian Acropolis Hoard,"?" may be
oxidation (fig. 125).'" hemispherical bowl castings r.s i or, more probably, sim-
B 219. Pruning book casting. Figs. 126 and 127. L. .138, ply mirror "Fragments. Although they are thicker than
Th. 005. After the sides had been curled over to mirrors from Dendra lG2 and Cyprus/53 they are no
form a socket, this would have been similar to thicker than some mirrors from Egypt,154 Further, the
B 98 and B 99, although the pointed end curves
out more strongly in this case (1959). A metal straight cut on the more complete disk, B 225, suggests
one-piece mold which would have produced almost an arrangement for fitting a wooden handle; such
identical castings is from the Cypriot Mathiati handles were attached normally with either rivets pierc-
Hoard.wr ing the disk, or with a protruding tang.15.5 The slightly
B 220. Hoe casting. Figs. 126 and 127. L. .065, max. V\T .
.06, Th .. 006. T -shaped piece of flat bronze. with
rounded angles to ar111S. The end of the blade is
probably missing (S). Such a casting could have
been used in forming hoes of the type found at
Geliclonya. A similar piece is from the Mathiati
Hoard.t-" and a metal mold for producing such
castings is also from Cyprus, in the GUl111is
Hoan1.l-JO
B 221. Hoe casting. Figs. 126 and 127. L. .155, max.
w ..
09, blade W .. 055, Th .. 01. The sides of the
socket are more rounded than those found on
completed tools at Gelidonya
B 222. Rectangular sheet.
(1959).
Figs. 126 and 127. L. .25,
~,"-._------'.
, -,,
,,,_.... I
'vV.. 09, Th .. 01. Casting for unknown purpose; I / ...... J

broken into three pieces (P). I

\. __
I
1
B 223. Rectangular bar. Figs. 126 and 127. L. .135,
W .. 022, Th .. 014. Slightly bent (1').
---
143 Catling, 226-227. FIG. 125. Stone cover on open mold prevents oxidation
114 Petrie, n. 142) and 29.
loco cit. (slIpra" on surface of bronze.
145 Two molds from Cyprus, one single and one multiple, but
both for similar, pruning hooks, show that the caster had a 150 Montelius, Crece Precl., 156, with fig. 497.
choice. Catling, 272, no. 2, with pl, 50: b, and 274, no. 1, with 151 Catling, 147 and 293.
pI. 50: L 152PerSSG11, Royal Tombs at Dendra, 96, Chamber Tomb
14-5 Tylccote, Met. in Arclme., 111-112; Schliemann, Ilios, 434. no. 20, 1l0~20.
1<1-7 Catling, 272, no. 2, with pl. 50: b. 153 Catling, 224-225.
148 Ibui., 276, no. B 1, with pl. 51: g; SeE III, 665, fig. 374. 154Petrie, Obj. of flaif'}l Use, 31-32.
ith pl. 50: a.
149 Catling, 272, 110. 1, ,.... 155 Catling, 224-227, discusses methods of attaching handles.
114 CAPE GELlDO~YA: A BROJ\ZE AGE SHIPWRECK [mANS. AiIlER. l'HlL. SOC.

- ~
•
i·

220 <
B 219
221 223

222

225 246

[J
243
-
)
224
- I

226
228
232
-244
257

I
Oems 20

FIG. 126. Unworked bronze castings, casting waste, and unidentifiable scraps: B 219 to B 257.

convex, smoother surfaces, which would seem to make XXIX. CASTING WASTE
the disks impractical as mirrors, are also found on In casting an implement in a closed mold, molten
many Egyptian examples where the use is not in metal was poured from a crucible into a funnel-shaped
doubt.t'" opening (fig. 128); the metal was carried from this
B 225. Fig. 126. D . .16, Th .. 004 to .003 at rim. Disk opening through one or more tube-like passages so that
with squared rim. Surface beaten on one side; it spread evenly throughout the mold, After the casting
smooth, but slightly convex, on other. Two seg- had solidified, the metal in the opening and channels,
ments cut off straight across edge in antiquity (P).
known as jets and runner s.t'" "vas cut away and the
B 226. Fig. 126. .05 X .036, Th .. 003. Fragment of disk
of same diameter and thickness as B 225, also with surface of the implement was ground down and
squared edge (P). smoothed. One or more openings were also needed to
B 227.. 049 X .03. Fragment of disk originally the same allow the escape of gas from the mold, and the metal
size as B 225 (G). which rose in these had the rough, bubbly surface found
B 228. Fig. 126. Max. chard L. .09, Th .. 004 at break
and .002 at edge. The edge is rounded rather than on the ingots; these risers were also cut away from the
square, and the diameter was originally larger than
157 Tylecote, Met. in. Arcbce., 109, points out that the term
those of the preceding (G).
--- "jet," which is commonly used by archaeologists, is not used in
1513 Petrie, loco cit. (snpra, 11. 154). foundry practice; "gate" is the more accepted term.
 ••
..
••
B 220

B 219
• B 221

I
I
••
I

• ••• I
I
• ••
;
I
J I
.m'

8229

8222 B 223 8224 I

I
8230

I • I
•
I •.•~. I
8233 8235 8236 B 237
FIG. 127. Blanks for unfinished bronze tools, and casting waste.
115
116 CAPE CELIDONYA: A BRONZE ACE SHIPWRECK [TRANS. AllIER. PHIL. SOc.

FIG. 128. Casting scene from Tomb of Puyemre

at Thebes (after Davies). •

finished product and discarded. Figure 129, which

depicts a hypothetical mold for casting a solid figure
such as the horned male from Enkomi.l'" shows the
various features described.
B 229. Figs. 126 and 127. Conical jet with three runners.
H .. 05, D. 04. The three runners are broken off
quite close to the jet, but enough is preserved to
show that they ran off at different angles. A
slight ridge of metal protruding upward around
the convex top had adhered to the sides of the
funnel-like mold opening (P).
B 230. Figs. 126 and 127. Jet with two runners. I-I. .029,
top .03 X .021, runners approx .. 008 and .009 in
diameter. The pouring cup was not strictly conical,
resulting in an oval top on the jet; runners at
slight angles to one another (P).
B 231. Fig. 126. Conical jet with one runnel'.
D .. 045.
of metal; runner at slight angle (P).
H .. 02,
51ight ridge around top from overflow FIG. 129. Hypothetical mold for casting Horned God from
Enkomi, by Susan Womer. A. Funnel-shaped opening.
l
I
B 232. Fig. 126. Conical jet with one runner.
D .. 026 (M).
B 233. Fig. 127. Conical jet or riser.
H .. 022,

H .. 03, D .. 038.
B. Metal runners. C. Ail" vents.
•
Ridges, or seams, on either side resulted from metal
seeping between the two halves of the mold. Slight
ridge on one side of top from overflow of metal.
A lead jet, with runners
(p 131).
(L 23), also came to light
•
Top very rough (G). Such casting waste probably lasted only until it was
B 234. Nearly conical jet 01- r-iser. H .. 024, top .04 X .032.
Oval top slightly concave; bottom founded off
remelted and used again. It has been found in Cypriot
evenly (E). founder's hoards.t'" but was surely not limited to that
B 235. Fig. 127. Unfinished and broken tool with run- island.
ners. L.. 04, VV.. 025, Th .. 012. Possible pick end
from which runners had never been removed xxx. UNIDENTIFIABLE SCRAP
(1959) .
B 236. Fig. 127. Casting waste. L. .125, W.. 085, Th. Fragments of discarded utensils of unrecognizable
.032. Roughly oval lump of metal, rather flat 011 shapes were found scattered throughout the cargo;
one side, which may have hardened in a crucible.
B 237. Fig. 127. Dump. L. .033, W. .025. Roughly most, presumably, had been contained in baskets which
conical Jump of copper or bronze which probably have disappeared. Such scrap metal is usually found
spilled and hardened during casting (G). in ancient founder's hoards."?" At Gelidonya, a number
B 238. Dump. L. .021, W .. 015. Hardened lump of
spilled copper or bronze (M).
159 Catting, 276.
160 Chi Ide, The Bronze Age, 45, distinguishes "founders'
158 The illustration, by Susan Worner, is based on modern hoards" from "domestic," "traders'," and "votive" hoards by
casting techniques and uses the statuette from Enkomi (Di- the "presence of old and broken tools, obviously scrap metal
kaios, AA (1962) 1 L, with figs. J8-21; Catling, 255, with collected for remelting, and often too of metallurgical tools,
pl. 46) only as an example. moulds and ingots of l-aw metal."
 VOL. 57, PT, 8, 1967]
BASS: THE BRONZES
117
of nearly square or rectangular pieces of sheet metal
parallel objects from other sites, and they are the imple-
may have been cut from unworked castings,
ments which most likely accompanied the ship fr0111
B 239. Nearly square plate, .068 X .065, Th .. 005. Tapered its home port.
edges, very slightly curved (P). The cargo of scrap, however, we may judge to have
B 240. Nearly square plate .. 04 X .04, Th. 007. (1959)
B 241. Nearly square plate. .045 X .045 to .040, Th .. 005 been Cypriot. By far the best parallels for the picks,
to .008. (1959) shovel, pruning hooks, sickle and axe-adzes, as well
B 242. Roughly rectangular plate .. 06 X .05, Th .. 01. (G). as for several of the various types of hoes, chisels, and
B 243. Fig. 126. Plate fragment. .06 X.045, Th .. 01. bracelets, are from Cyprus; that at least Some of these
Three straight sides, one broken. (P)
parallels were actually made in Cyprus is proved by
B 244. Fig. 126. Plate frag. .045 X .04, Th .. 005. One
right angle, two broken sides. (P) the discovery of molds for pruning hooks and hoes on
B 245. Plate frag. .035 X .035, Th .. 003. (M) that island. Axes, harnrner or anvil, swage block, awls
B 246. Fig. 126. Plate frag. .047 X .04. Triangular and nails, punch, needle, knives, spearheads, razor,
fragment with one square corner. (P) spatula, vessels and stands, spit, and miscellaneous
B 247. Notched plate frag .. 048 X .033. (G)
B 248. Plate £rag. .062 X .047, Th .. 003. Broken all all bronzes could as well have come from Cyprus as else-
sieles. (M)
• where. Of the remaining objects, some of the hoes,
B 249. Plate frag. .05 X .044, Th .. 005. Broken on all the socketed tool B 95, the adzes, and possibly the
sides. (G)
mattock show strong Near Eastern relations. Almost
B 250. Plate frag .. 04 X .22, Th .. 008. (G)
B 251. Curved hag. .03 X .025, Th .. 002. (M) nothing points to an Aegean origin; the hoes of Type
B 252. Tool hag. .044 X .032, Th .. 006. Shaped shoul- 3 from Athens and Antheelon are so much the exception
der, perhaps spade fragment. that one must consider them as being imports into
B 253. Plate frag. .07 X .048, Th .. 01. (M) G reece. (See figs. 130 and 131.)
B 254. Tool £rag. .044 X .03, Th .. 002. Triangular tool
fragment. (M) A determination of the elate of the bronzes poses a
B 255. Plate frag. .075 X .045. (M) number of problems. -VVehave seen that many of the
B 256. Curved strip. L. .052, W .. 013, Th .. 0015. parallels are from hoards which may not be elated by
B 257. Fig. 126. Irregular frag. L. .055, \1\' .. 018. the contexts in which they were found. Properly exca-
vated parallels, on the other hand, are dated mostly to
COI'CLUSIONS either the thirteenth or twelfth century B.C. This dif-
ference in dating is important, for the closely similar
It is evident that most of the bronzes were frag- bronzes from Gelidonya formed a closed deposit of
mentary and were, therefore, being transported for contemporary material; three possible explanations for
their scrap value. Whether or not the whole tools, the difference corne to mind; styles of metal tools and
including adzes, hoes, an axe, and an axe-adze, were weapons did not cllange appreciably during a century
l ready for sale or were also to have been melted down
and recast remains a question. Some of the tools may
or more, Some of the Gelidonya tools were already
quite old when placed in the cargo, or the evidence [or
I have been for use on the ship itself, as a study of the
• distribution of the bronze cargo suggests .
Both complete and fragmentary bronzes were found
dating parallels found all Janel has been misinterpreted
by some of the excavators.

• in all areas of the site, showing that the cargo had been
packed in the ancient merchantman almost from stem
I believe that the fir-st two possibilities may be dis-
missed. A Ithough some shapes changed little if at all,
the large number of variations on such a simple object
to stern. Although some of the smaller pieces, espe- as the lugged adze, as shown in Mrs. Maxwell-I-Iyslop's
cially those carried in somewhat buoyant wicker bas-
study, and the very few close parallels in her catalogue
kets, may have shifted their positions as the ship sank, for the Gelidonya tools, indicates that subtle variations
the concentration of sma.l scarabs, weights, seal, lamp, were clue to the preference of an individual or of a
and whetstones in the "cabin area" of G (spilling into particular atelier. We may further note that some of
area 1\1) shows that even the lightest objects did not
the Geliclonya objects which are most perfectly pre-
move far. This is possibly significant in that the loca- served find their best parallels alllong the earlier,
tion of most of the u-nique and coniolete bronzes was thirteenth-century objects, and it is doubtful that tools
in the "cabin area": the razor, punch, spatula, needle, would have been preserved so well after years of use.
socketed tool B 95, and 'well-preserved chisels were all Almost perfect parallels must rather be mostlv from
from areas G and TvL In this category, only the swage the same century, I feel, but whether this was the
block (area E) and, possibly, tile spit (raised in 1959) thirteen tb century or the twelfth century must be
were found elsewhere. We have, therefore, the strong decided. -
possibility that some of these unbroken pieces were
Catling, who has collected and presented well all the
personal possessions of the crew and were not cargo;
direct and indirect evidence for dating Late Cypriot
most could have been used on board a ship.
bronzes, assigns most of the Cypriot material to the
Unfortunately, these "personal possessions" are ex-
twelfth century. The thesis of his book is that a
actly the bronzes for which it is most difficult to find
colonization of Cyprus from the Aegean after 1200 R.C
118 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

H
RCls Shamra

I Byblol

9 'gypl

DISTRIBUTION OF HOES: CHISELS:

type 1 D. type5 • I cold chisel
type2 ... type6 0 ~ deep bar chisel
type3 0 type7 •
type4
FrG. 130.

brought about the manufacture of a great number of tool forms are evidently of Near-Eastern origin. "163

new types of bronzes on that island, and it is these new Based on the material presented in the same chapter
types which are found so abundantly in hoards. I which contains these statements, "many" and "several"
believe that in stressing Aegean influence on the should be interchanged. Catling's own conclusions
bronzes, an influence that Desborough also sees.l'" about individual types of tools reveal that more than
Catling misinterprets some of the evidence which he twice as many find their origins in the Near East as
presents. in the Aegean. Similar Near Eastern origins also are
Catling concludes that "many" of the Late Cypriot found for many or 1110stof the Cypriot statuettes, tri-
tools "can either be closely matched by Aegean tools pods, personal objects, and miscellaneous bronzes. Cat-
of the same type or may reasonably be seen as the end ling's suggestions that some bronzes, such as charcoal
of an Aegean series the final stages of which have yet shovels and tanged mirrors, may have been introduced
to be found there." 1(l2 He also states that "several from the Aegean, where they have not been found,
rather than Irom the Near East, where they have been
1(;1 Desborouzh. Last Mycenoeans. 49.
1132 Catling, 109. ]G3 Ibid.
 VOL. 57, PT. a, 1967J BASS: THE BRONZES 119

•
•
~

DISTRIBUTION OF MISCELLANEOUS BRONZES:

, picks
• double axes
(jj) socketed tools , adzes
• shovels ~ axe- adzes
o mattocks m tripod stands
~ pruning hooks X offering stands
) sickles o bracelets
FIG. 131.

found, show the difficulty involved in finding Aegean assigned to the twelfth century by their excavators,
origins. 1M
both are atypical.':" The evidence for twelfth-century
A twelfth-century Aegean colonization of Cyprus did dates for the other hoards is not strong. Catling places
not influence the manufacture of bronze implements on the Enkomi Foundry Hoard after 1200 B.C. on the
the island, therefore, and we may consider the dating basis of stands and oxhicle ingots, which we have
of the bronzes independently of such an event. Most shown are not necessarily that late. He places the
of the Cypriot material which is relevant to the Geli- other hoards after 1200 B.C. by the similarity of the
donya cargo is from hoards. Only two of these hoards bronzes in them to "twelfth-century" types, but he has
(the Tresor de Bronzes and the Weapons Hoard) have dated almost all of these types on the evidence that they
been dated stratigraphically and, although hath are came from these same "twelfth-century" hoards. If we
164 tu«, 101, 227.
te tu«, 287-288.
(

120 CAPE GELlDONYA: A BRONZE ACE SI-J IPWRECK ITRAl\'S. AMER. PHIL. SOC.

do not begin with the assumption that the hoards are Palestinian coast by the Peoples of the Sea, at precisely
from the twelfth century, however, ..... ve see that the best the time the ingots disappeared, is the most likely
evidence does not point to that date at all. Of the event. Because the ingots and bronzes are so closely
stratified single parallels on Cyprus, more may be associated that they call scarcely be studied indepen-
elated before the twelfth century than to it, and good dently, .I would further suggest that the trade of the
Near Eastern parallels are also generally earlier than bronzes stopped at the same time; there is not a single
1200 B.C. hoard on the Greek mainland whose context would
\Ne have already seen that copper "oxhide" ingots suggest a elate after 1200 B.C.,16!J and it would seem
have been misinterpreted almost universally in a like 1110stunlikely, even without the evidence which we have
manner, for such ingots did not introduce "a wholly discussed above, that none of the Cypriot hoards, as
Aegean element into the management of the smelted Carling believes, comes from the extremely rich thir-
copper," lOG and no particular type of ingot suddenly teenth century on Cyprus.
appeared on Cyprus, as Carling believes, about 1200 The conclusion that Near Eastern trade with Cyprus
B,C. Representational evidence, indeed, indicates that and the Aegean ended at the close of the thirteenth
it was just at that time that oxhide ingots disappeared century has been reached by a consideration of the
forevcr"'" The only evidence for ingots existing any- copper ingots and bronze implements alone. A pre-
where after 1200 B.C., in fact, is from the later excava- ViOLlSstudy of non-metallic material had led Sjoqvist
tion by Schaeffer of the area which 'tvas swp posed to to believe that at this time (LeI II) there was "an
have produced the Enkorni Ingot Hoard, and the re- almost complete break in the peaceable relations with
cently published ingot-shaped base for a bronze statue the East," 170 and Cyprus "turned her face west-
found at Enkomi (1'.69). wards." l~l It seems, then, that increasing Aegean
Professor Dikaios, on the other hand, has kindly influence affected Cypriot bronze-work more in hasten-
informed me that at least in the parts of Enkomi which ing the end of established, 1\1ear Eastern types than in
he has excavated, copper smelting reached its peak in introducing those types.
the thirteenth century, and that it is possible that the vVe have arrived at the same question that arose lI1
bulk of ingots come from that period (Late Cypriot our conclusions concerning the oxhide ingots: are we
lIB) ; indeed, he has a fragment of an ingot from as sure that Phoenician sea power and trade did 110t play
early as LC IIA. The impetus which sent copper a large part in the Aegean before the first millen-
working to new heights in the thirteenth century almost nium ? 1.2 A second question also arises: is there any
stopped when Myc. IIIC:l pottery appeared late in reason to believe that Phoenician metal workers did not
that century (1230-1200 B.C.). Whether the smelting gain their great fame until after the close of the Bronze
ceased entirely or was merely moved to a new location Age?
is not yet known, but Dikaios feels that the industry The answers, again, are no. Even without the ship-
generally declined owing to the repeated disasters of wreck at Cape Gelidonya, a close study at the Cypriot
the twelfth century.':" copper and bronze industry should have revealed that
This reconsideration of the ingots, which have been it was more Syro-Palestinian than Aegean in character,
fully discussed in a separate chapter, is important for and that many of the metal pieces in mainland Greek
the following conclusions. If the ship at Gelidonya was hoarels were Near Eastern types made on Cyprus. The
Syrian, as I believe we can show with evidence inde- Gelidonya ship serves only to show that the metal was
pendent of the ingots or bronzes (chap. XIV), our traded by Near Eastern merchants as well. That the
earlier conclusion that oxhide ingots were dealt with bronzes were cast on Cyprus, as shown by the occur-
by Syrian merchants is proved. These ingots suddenly rence there of molds, suggests that, as in the case of
ceased to be manufactured around 1200 n.c., after two the ingots) Syrian or Phoenician smiths were them-
centuries of uninterrupted use in which not even the selves present on the island, nearer the source of the
slight variations on the shape had changed, That bronze copper which they used; at the least, the bronzes seem
tools, weapons, and other implements of mostly Near to have been made to meet Near Eastern specifications.
Eastern origin were dealt with by the same merchants,
The earlier occurrence, preceded by prototypes in some
w hich had been suggested by the presence of ingot
cases, of the bronzes in the Near East precludes any
fragments in hoards of bronzes, is further proved by
suggestion that the examples found in Syria and Pales-
the GeJidonya wreck.
I believe that some historic event brought about the tine were simply imports of native Cypriot handiwork.
end of this trade, and that the devastation of the Syro-
169 Desborough, 1oc. cit. (supra, n. 161), with n. 7,
lOU tu«, 302. 170 Sjoqvist, Problems, 190.
167 We have seen that the only sure twelfth-century repre- 171 tu«, 209.

sentations of ingots, at Medinet Habu, were probably copied 172 The prevailing view, that Phoenician maritime commerce

directly Irom the thirteenth-century Ramesseum. expanded only after the close of the Bronze Age, is discussed
168 Letter dated 19 February, 1964. in chap. XIV.
 VOL. 57, PT. 8, 1967J BASS: THE BRONZES 121
I suggest further that the Near Eastern smiths 'were the twelfth century: the Tresor de Bronzes from En-
making the bronzes for their own use, both on Cyprus komi. I have stated elsewhere my reluctance to reinter-
and, especially, in their homeland where examples are pret excavated material, or to accept the reinterpretation
more often complete than not. Most of the objects of others, for the excavator is in a much better position
found at Getielonya and in Greek hoards, on the other to evaluate the context of his finds. I must point out,
hanel, were broken, and there is nothing to suggest however, that the Tresor offers the only strong evidence
that ~1ycenaean Greece imported complete bronze im- against a thirteenth-century date for the Gelidonya
plements in any quantity. It seems, rather, that Greece cargo, and the distinctive axe-adze, which is Common
preferred to buy only raw materials and scrap metal to both the Gelidonya cargo and the Tresor, finds a
for their own smiths or for itinerant craftsmen who close parallel at Pyla (Kokkinokremmos) which may
would cast objects to suit Mycenaean tastes. be dated by the pottery found with it to around 1200
In concluding that bronzes of the type found at H.C. One cannot ignore such a deposit as the Tresor,
Getidonya were made on Cyprus under Near Eastern but, on the other hand, one cannot ignore the much
direction during the thirteenth century, one problem greater mass of evidence that the other hoards were
remains. We have passed lightly over the one impor- oE the thirteenth century.'!"
• tant hoard which has been dated stratigraphically to 173 See chap. XIV, n. 15, below.
 IX. MISCELLANEOUS FINDS

GEORGE F. BASS

l. LEAD II. METAL FOIL

Relatively few lead objects were found, and although Metal foil, folded double and then folded in an S-
some of these, such as the net sinkers and conical curve, was found under a stack of four slab ingots
weight, were not out of place on board a ship, the lead (SI 1, SI 4, SI 12, SI 15) in area G (figs. 140,33).
casting jet and pieces of scrap indicate that at least It proved impossible to remove the sheet, or sheets,
some of the lead was carried for future use in forming intact from the surrounding concretion, for the foil was
other objects only .00075 thick and was extremely brittle. One piece
The high silver content of one sample (append. 4) was approximately .35 long, with a shorter, probably
indicates that there had been 110 attempt to extract the separate sheet below. A width of .10 for the foil is
silver from the ore, and that, therefore, the lead was only an estimate and could be completely erroneous.
mined for its own use and was not a by-product of A. E. Parkinson, chemist of the University Museum,
• silver mining. It is not possible at present, however, has reported on analyses of samples of the foil as
to speculate on the source of the ore. Hopefully the follows:
further study of isotope distribution ratios in lead will
Spectrographic analysis of a sample by Mark Han
eventually offer a clue as to where it may have been showed copper and tin as major metall ic constituents, with
mined (see append. 4). Lead ores are widespread in silicon. silver, iron, manganese, nickle, cobalt and mag-
the Eastern Mediterranean, with a concentration in nesium.
Asia Minor, but with deposits in Greece and Syria; A quantitative analysis of another sample to determine
the amounts of copper and tin gave 52.8 per cent of tin
the Syrian deposits are small, but certainly sufficient
dioxide, 19.0 per cent of copper (cupric) oxide and 4.3
to have supplied the lead for the small pieces found on per cent of silica. In terms of metal these results are
board. Only Cyprus, of the regions of greatest interest equivalent to 41.6 per cent tin and 15.2 per cent copper in
in this study, may lack its own lead." the sample. Spectroscopic analysis of the non-siliceous
residue showed the presence of iron, calcium, magnesium
Ll-l8. Net weights. Fig. 139. Eighteen of these were and lithium.
found concreted together in a cluster in area P. The large amount of tin suggests that the original alloy
Each of these is a strip of lead, approximately .035 may have been largely tin and may have had a silvery
long, .03 wide, and .003 thick, which has been folded color, although caution must always be observed in draw-
over along its longest center line. The occurrence ing conclusions from the composition of a corroded speci-
of such weights, which are exactly like those used by men. When tin corrodes it usually forms the very insolu-
modern fishermen in Greece," among the personal ble tin dioxide, whereas copper may form compounds that
objects on a Byzantine shipwreck 3 indicates that may be leached out by subsequent solvent action; thus a
they were not necessarily part of the cargo; Homer specimen may become enriched in tin.
mentions the use of nets for fishing in the Odyssey
(22.386). A number of un catalogued duplicates are The fineness of the material suggests that it was
in the Cyprus Museum. used for making bowls, for it is approximately the same
L 19. Line sinker. Fig. 139. If. .032, max. W .. 017. thickness as two of the bowl fragments found on the
Pyramidal lead weight, pierced at top; top broken wreck (B 171 and B 172). That it was already
off through hole (area 1\1). This is similar to a
sounding lead found on the Byzantine shipwreck at extremely pliable, however, is shown by the manner in
Yassi Ada and to those still in use in the Eastern which it was folded, and it would have become even
Mediterranean, but it is much smaller and was thinner in the process of being beaten into bowl shapes.
probably only a line sinker. The use of lead on a This fineness and apparently high tin content suggest
fishing line is mentioned by Homer in the Iliad
(24.80) . another intended use.
L 20. Whorl. Fig. 139. D .. 05, hole D.. 01, max. Th .. 001. Sara Immerwahr has reported on the apparent use
Pierced disk, tapering to edge (area M). of tin foil to cover certain Mycenaean vases found on
L 21. Weight? Fig. 139. L. .065, max. D .. 01. Narrow the Areopagus, and possibly others from Mycenae,
rod, pointed at ends and flat on one side (area G).
Dendra, Knossos, and Ialysos.:' She and her collabo-
L 22. Disk. .085 X .07, Th .003. Irregular lead disk,
probably scrap (area P). rator, Mar-ie Farnsworth, have informed me that the
L 23. Jet with two runners (see p. 114). H .. 021, max. D . purity of the tin and the thickness of the foil are not
.025. Conical jet, concave on top, with two runners. known in those cases (continued investigation is at
L 24. Casting waste. L. .055, W .. 017. Oval dump, flat
on one side (area P). 4 Paper read at the 64th General Meeting of the Archaeo-
L 25. Scrap sheet. .07 X .07, Th .. 003 approx. Very logical Inst. of America, and summarized in AlA 67 (1963)
irregular. 212-213. Mrs. Immerwahr has since informed me that "the
---' vases I -mentioned in my paper from Dendra, Mycenae, and
1 Forbes, Met. in Ani., 169-201. Knossos were also sheathed. . However, my examples all
2 Gladys R. Davidson, Corinth XII, 190, with pl. 88 :1449. belong to a fairly restricted period (the early fourteenth cen-
3 Bass. AA (1962) 552. tury) and occur in wealthy chamber tombs."

131
132 CAPE GELJDONYA: A BRONZE AGE SHIPWRECK [TRANS. A"tIIER. PHIL. SOC.

LI
u-u-. Ll9

L20 l21

1i2\.'....
0-'0 n."'· @-o
ASTRAGAL

8EAD TYPES
o em. 5 10

FIG. 139. Lead, glass, and bone objects.

2 3 5
'7 / FIG. 142. Beads.

FIG. 140. Metal foil.

FIG. 143. Astragal.

present underway}, but she feels that the Gelidonya

sheets would have been rather thick for such sheathing;
it is possible, however, that the foil covered wooden or
ivory containers, for an ivory pyxis from the North
Slope of the Areopagus 5 showed traces of a metal
lining which was analyzed as tin by Miss Farnsworth.

Ill. GLASS BEADS

A lump of concretion in area P contained several

hundred beads (fig. 141). Pottery sherds around the
lump suggest that the beads had been contained in a
vessel which had broken during the sinking of the
ship; a bracelet, B 188, was imbedded in the beads and
was undoubtedly carried in the same jar. A few beads
were removed from the concretion with great difficulty
(fig. 142) and these, although they crumbled to dust
if allowed to dry, revealed three types:
Type 1. Ovoid. Fig. 139 a. Ca. .01 long, tapering at
ends; white spiral around dark body.
Type 2. Spherical. Fig. 139 b. Ca.. 004 in diameter, but
.. some are slightly piriform; greenish yellow.
FIG. 141. Beads under water. 'Hesperia 9 (1940) 286, fig. 27.
 VOL. 57, PT. 8, 1967J BASS: lVllSCELLANEOUS FINDS 133

• ••• eM'

Fie. 144. Fish bones.

Type 3. Discoid. Fig. 139 c. Ca... 005 in diameter and

.0015 thick. Solid dark color, almost black.

Analyses (append. Sa and Sb) show that at least

some of the beads were of glass. The dark coloring
may have been due to the copper corrosion products
which had permeated the concretion in which the beads
were found, staining at least the concretion a very
. dark green.
Neither the shapes nor the decorative features of the
beads are very distinctive. All three types are found
in the Near East," and at least Types 1 and 3 are to be
seen in the Cyprus Museum ; indeed, beads of Type 1 I
have been retrieved by divers from Swiss lakes, and are FIG. 145. Olive pit.
undoubtedly found even farther from the Mediterranean.
Another use for a single astragal on board a ship
IV. BONES must be considered. Astragalomancy, in which the fall of
1. ASTRAGAL the bone indicated divine will, V·/aS a later form of divi-
nation 12 particularly associated with Hermes, patron of
(Figs. 139 and 143. L. .048.) clerornancy;" a god with Mycenaean origins in Greece,
A single knucklebone, from area G, was the only and, at the same time, god of merchants. Knucklebones
large animal bone found all the ship and was, therefore, seem also to have been used for divination in the Near
probably 110t fr0111 a sheep slaughtered for food. The East.':' What more suitable talisman was there than
game of knucklebones, similar to dice," is mentioned a knucklebone for a merchant captain in a time when
by Horner," and has been played at least from the time the sea "vas feared? VVe know that divine guidance was
of the eighteenth Egyptian dynasty 0 until the present; 10 sought by the captain before he chose the route which
the occurrence of such bones in Neolithic Crete suggests he would sail."
that the game may go back much further.v'
2. FTSH BONES
G Tufnell, Lacliisli II, pl. XXXI V shows beads of all three
types. (Fig. 144)
7 Oxford CI. Dict., 110.
s mad, 22.88. Fish vertebrae and other small bones appeared in
9 Schaeffer, Uqoritica IV, 103; P. Fox, Tutanklwlllen's area G, which also yielded olive pits, and may have
Treasure, pl. 66.
10 Knucklebones is still played in Europe and the Near East, 1~ PT/V, suppl. vol. IV (1924) 51 ff., astragalomanteia.
according to friends who have played it or seen it played 111 13 Dar.-Sag. 11, 301, Divination.
France, Turkey, and Syria. See also W. Deanna, Le Mobitier 1'1Tufnell, Loctnsli II, 94.
oeu,« 332-333. 150d. 3.173. Seymour, Ufe in tlic Homeric Age, 525, lists
11 Montelius, Grece Prect., 7. instances of cleromancy in Homer.
134 CAPE GELJDONYA: A BRONZE AGE ST-lIPWRECK [TRANS. AMER. PHIL. SOC.

been further remains of a meal; one diver reported a Olives were cultivated in Palestine, Syria, Asia
small, hollow bone which he assumed to have been Miner, the Aegean Islands, and the Greek mainland
from a bird, but this was never further identified. Fish at the time the Gelidonya ship sank, and the existence
vertebrae were often saved in antiquity and sometimes of the pits, therefore, offers no clue concerning the
pierced, and have been found in great numbers on route of the ship."
land, where they may have served as gaming pieces."
VI. CLAY
The possibility exists that these bones were encased in
the concretion which held the olive pits only by chance Two round balls, about .08 in diameter, were found,
and had nothing to do with the ship. one in area E and the other concreted to the underside
of oxhide ingot 21. These were at first believed to
V. OLIVE PITS have been ballast stones, but they consisted of fine, buff
powder when dry. It seems likely that they were balls
A number of olive pits were raised from area G, six of clay which had been luckily protected against the
in the air lift (fig. ]45). Olives playa large part in action of the sea water, which may have disintegrated
the cliet of modern Turkish sailors .. and it was believed more exposed lumps of clay elsewhere on the ship.
that some of these pits were simply dropped over the
side of 111andalin(i while it was anchored over the
Whether or not this represents clay used in making •
molds for metal objects is impossible to say, but one
wreck; the presence of pits deeply imbedded in the might expect molds to be found in a cargo like that
concretion from the "cabin area," however, proves that uncovered at Cape Gelidonya; the clay could equally
at least those were ancient. well, however, have been used for capping jars.
Hi Decnna, op, cit. (sttpra, n. 10) 331. 17 Tu lnell, Lacliisl, IV, 313; Zenner, Hcst. of Tech. I, 359.
 X. THE WEIGHTS

GEORGE F. BASS

Every merchant ship undoubtedly carried sets of number of pieces of concretion, discarded after having
weights in its cabin, for these would have been needed been broken into small pieces, were pulverized in a
almost immediately upon arrival at each port (fig. 49).1 last search for small finds. The discovery of five of
Weights were equally important in any metal-working the unit weights (the smallest and, therefore, most
establishment and are normally shown in scenes of easily missed) of the seven sets of weights used on
Egyptian smiths at work," including an illustration board, however, indicates that few weights were en-
which shows copper and tin ingots being melted clown tirely overlooked (fig. 44).
and cast." Not only did prices of finished artifacts The weights were in a variety of shapes, including
depend on their weights, as they often do today/ but forms commonly found in Aegean and Near Eastern
proportions of metals to be mixed must have been sites: "sphendonoid" (after its superficial resemblance
carefully set ill advance. The importance of metal to sling bullets)' or spindle-shaped (fig. 146); domed
objects being carefully weighed is indicated in Hittite o (fig. 147); sugar-loaf shaped, or blunted conical (fig.
• and Ugaritic texts, the latter perhaps mentioning the 147); spherical, with flattened base; cylindrical, with
weighing of ingots ;? copper ingots are almost invaria- length greater than diameter; discus-shaped; and ir-
bly depicted with balances on Linear B tablets, again regular pieces of stone and metal. Figure 148 gives
showing the necessity of weighing each one. K 0 com- typical examples of each type.
plete balance was found on the shipwreck, although The weights seem, to the untrained eye, to be mostly
one fragment of metal may be the remains of a balance of the same material, except for those made of bronze.
pan (B211). Chips from t\VO broken weights have been identified
Sixty small objects of stone and metal at Gelidonya as haematite."
were catalogued as weights or possible weights. These, In order that the reader may judge the validity of
with two exceptions (VV 52 and VV 60), were almost the conclusions reached in the following study, the
certainly from the cabin of the ship; most of them were catalogue lists all possible weights, although some of
found either in area G (the cabin area) or in area lVI, these have proved to be probably only chips of stone
where many small objects from G seem to have been
carried by the current. Only Vif 60-, which was firmly
fixed with a number of bronze tools in area P VI, can
be considered apart from the others-a consideration
of some importance in the study of the objects.
further word of explanation is necessary concerning
the proveniences of the weights. They were among the
smallest objects on the site and were the most likely to
A

'"
,--,
.... ,
have been discovered out of context; some were found
a year after the completion of the excavation when a

1 A scene of Syrian ships arriving in an Egyptian port, with

merchants weighing merchandise with balances and weights,
is found on the 'wall of Tomb 162 at Thebes. Norman de G.
Davies and R. O. Faulkner, JEA 33 (1947) 40-46, with pl. 8;
J. B. Pritchard, ANEP, 33, fig. Ill, and pp. 262-263.
2 Norman de G. Davies, The Tomb of Rekh-1J/.i-ri}C I, 52,
with pl. LV, row 2. P. E. Newberry, Bene Hasan: 1,31, pl. 11;
Pritchard, ANEP, 37. fig. 122, and p. 264. Pritchard. ANEP.
==~' .2__
FIG. 146. Sphendonoid weights.
40, fig. 133, and p. 265 (Tomb of Mereru-ka at Sakkara, Sixth
Dynasty). , Arthur J. Evans, Corolla. Num .. 348; PM IV, part II, 655.
S N. de G. Davies, The Tomb of the Two Scutptors, pl. XL. The theory that this shape is Syrian in origin is given by
George Steindorff and Keith C. Seele, When Egypt Riaed the Flinders Petrie, Ancien! Weighls and Measures, 6.
East, 185, with fig. 6L For the importance of copper and tin sA. Eric Parkinson has reported on one sample: "Roughly
ingots at Gelidonya, infra, p. 163. spherical. Color dark red. Wt. 146.4 gms. Specific gravity
± The prices of almost all items made by the Bodrum black- (assuming object a sphere with diam. 3.8 cm.) : 5.1; this is
smith for the expedition, including objects of pipe, angle-iron, within the haematite range. Hardness 5.5-6.0 (correct for
chain, and sheet metal, were based 011 weight, almost without massive haernatite ). Spectroscopic analysis of tiny chips of
regard to varying amounts of labor which had gone into them. the small amount of incrustation still attached to the surface
5 J. B. Pritchard, AlVET, 208. showed calcium, iron, copper and possible traces of magnesium
6 O. R. Sellers, lnterp. Diet. of the Bible I, 343; C. H. Gor- and strontium. The incrustation was, therefore, calcareous
don, Ugaritic Literature, 64; J. A. Montgomery, JAOS 54 contaminated with copper salts and containing iron, probably
(I 934) 60. oxide from the haematire.'

135
136 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRANS. lHIER. PHIL. SOC.

W 6. H .. 018, D .. 018. Spherical stone weight. 12.30

grams.
V\T 7. L. .034, 'N .. 021. Possible weight. Irregular piece

•• •
of metal. 16.00 grams (M).
W 8. H .. 025, D .. 015. Possible weight. Conical piece
of metal. 17.80 grams (]\,1).
W 9. D .. 025. Discus-shaped stone weight. 20.60 grams
(G).
VV 10. L. .026, 'vV.. 017, Th .. 012. Rectangular stone
weight with rounded corner. 20.80 grams (Iv1).
FrG. 147. Domed and blunted conical weights. W 11. H. 025, D. 025. Conical metal weight. 26.10 grams
(G)
VV 12. H .. 022, D .. 020. Domed stone weight; flattened
top. 26.20 grams.
'IN 13. L. .04, 'vV.. 02. Sphendonoid weight; highly pol-
ished with the appearance of tarnished brass. 28.00
grams (M).
VV 14. L. .026, D .. 018. Cylindrical stone weight. 29.20
W14 grams (G).
W 13 \!l/ 15. H, .022, D .. 024, Spherical weight of metal. 29,80 •
grams (G).
'vV 16, D .. 03, Th .. 016, DIscus-shaped weight of haem a-
tite. Broken, but all pieces preserved. 35.00 glams
(G)
'vV 17, H .. 018, D .. 028. Domed stone weight; flattened
top. 36.00 g,"ams (G).
VV 18. H, .017, D .. 025. Thick disk of stone, rounded on
sides. 36.50 grams (G).
W42 \V 19. L. .035, W .. 025. Irregular chip of stone, probably
haematite. 42.70 grams (G).
\\T 20. H .. 024, D .. 026. Domed stone weight. 43,80 grams
(G).
\V 21. L. .04, D, .025. Sphendonoid stone weight with
blunt ends. 45.50 grams (G).
VV 22. L .042, W .. 021. Sphendonoid weight of metallic
stone with crystals showing in break. Tip broken
away. Actual weight 44.80 grams. Estimated
original weight ca. 46.00 grams.
\V 23. H, .02, D. ,027. Domed stone weight; flattened
top. 47.70 grams (G).
W53 'vV 24. H. ,025, W .. 025. Roughly cubical stone weight.
48.20 grams (G).
\'V 25, H .. 025, D, .03, Sugar-loaf stone weight. 49.40
grams.
VV 26. H .. 024, D .. 027. Domed stone weight; flattened
top. 51.50 grams.

OQ-· Vl 27. D .. 025. Possible weight. Metal ball, flattened on

one side. 54.20 grams (M).
'0/ 28. H .. 032, D .. 04. Domed stone weight; flattened top,
W 16 W4 55 ..\0 grams (M).
VI/ 29. D .. 03. Irregular sugar-loaf of stone. 56.00 grams
(G)
5 W 30. L. .034, D .. 025. Cylindrical stone weight 57,50
FIG. 148. Typical weights from shipwreck. grams.
W 31. H .. 025, D .. 028. Sugar-loaf of stone. .\9.00 grams
(G).
or metal. Area location on the wreck follows each V\T 32, L. .053, W .. 025. Rough sphendonoid weight of
entry. metallic stone. Humped back. Unfinished? 63.90
grams (G).
WI. H .. 009, D .. 013. Irregular but roughly spherical
V\T 33. L. .045) W .. 028. Sphendonoid stone weight; iron
stone weight. 3.50 grams (M).
L. .038, VV', .0] 2. Sphendonoid stone weight. content of stone has corroded on side and one end.
W 2. Ca. 2 grams missing from end. 63.50 grams, with
Straight scratch across top at right angle to long
estimated original weight ca. 65.50 grams (G).
axis. 9.30 grams (M).
W 3. H .. 014, D .. 016. Spherical stone weight. 9.50 VI 34. L. .043, \Y', .02. Sphendonoid stone weight with
grams (M). very blunt ends. 65,50 grams (G).
W 4. H .. 014, D .. 015. Spherical stone weight. 10.30 W 35. L. .04, D .. 028. Sphendonoid stone weight. 66.50
grams. grams (M).
W 5. L .024, D .. OIG. Cylindrical metal weight with \Y' 36. L. .053, W .. 03. Possible weight. Broken stone
pitted surface. 10.50 grams (M). in rough sphendonoid shape. 67.50 grams (M),
 VOL. 57, PT. 8, (967] BASS: THE WE1GHTS 137
VV 37.. 034, .024, .019. Possible weight. Shapeless lump Even after the use of these stone objects had been
of metal. 69.00 grams. recognized, some excavators neglected to report their
W 38. L. .042, VV .. 025. Roughly cylindrical stone weight.
69.80 grams (G). weights;" or gave weights of only a few of a larger
W 39. H .. 02, D .. 032. Sugar-loaf of stone. 76.00 grams number,» or recorded them inaccurately." Others tell
(G). us that certain weights were inscribed with signs, but
W 40. L. .035,. \IV .. 027. Roughly cylindrical stone do not mention which sign coincides with which weight
weight. 76.70 grams.
VV 41. H .. 033, D .. 037. Possible weight of stone with in a given list.ls Finally, even when weights were pre-
rough sugar-loaf shape. Badly pitted surface made sented admirably in catalogues, little study has usually
all of lime concretion impossible to remove. 79.30 accompanied mere lists.
grams (5). General articles seem almost unrelated to one an-
W 42. H .. 028, D .. 033. Sugar-loaf shaped stone weight. other, but it is not the authors who are at fault. A
86.00 grams (G).
W 43. H .. 025, D .. 035. Domed stone weight; flattened natural confusion arises from the vast number of
top. 92.00 grams. varieties within standards used in different places and
'vV44. H .. 03, D .. 034. Domed stone weight. 93.20 grams at different times in antiquity. Sellers lists, for exam-
(11). ple, fourteen weigbts inscribed with identical notation,
W 45. H .. 028, D .. 04. Domed stone weight; flattened top.
but no two of them weigh the same.'' It does not
• 99.60 grams (G) .
seem to be overly bold to suggest that only closed de-
W 46. L. .06, 'vV.. 023. Roughly sphendonoid weight of
copper or bronze (?). 109.50 grams. posits, containing sets of weights used by one man at
W 47. H .. 037, D .. 035. Domed haematite weight. 146.40 one time, will clear this picture, and that ancient wrecks
grams.
will offer the best possibilities for such deposits.
W 48. L. .065, W .. 033. Rough stone oblong, concave
beneath. 177.00 grams (G). An attempt to study the Gelidonya weights with
VV 49. H .. 032, D .. 04. Domed stone weight. 185.50 external evidence failed because even such a common
grams (M). standard as the qedet has dozens of variations; dividing
VV 50. H .. 022, D .. 042. Domed stone weight; flat top. sixty weights by variations of only hundredths of grams
188.00 grams (11).
W 51. H .. 035, D .. 04. Domed stone weight. 194.00 grams. was as unrewarding as it was laborious. Without
W 52. L. .075, W .. 033. Stone oblong with rounded ends. regard to known standards, therefore, the weights were
204.00 grams (P). placed along the edge of a grid and each weight was
W 53. H .. 04, D .. 047. Domed weight of highly polished then divided by appropriate numbers (2 to 15 for the
stone ; flattened top. 233.00 grams (M).
W 54. H .. 041 D .. 04. Spherical (?) stone weight with lighter weights, but up to 60 for the heavier pieces).
pitted surface and concretion adhering. 244.00 Table 1 is an example of the results.
grams. Using this grid, a frequency count was made of num-
W 55. H .. 036, D .. 047. Domed weight of polished stone. bers which repeated exactly, and a secondary count was
279.50 grams (11). made of numbers which were coincidental to within one-
W 56. L. .072, W .. 038. Sphendonoirl stone weight with
flattened ends. 284.50 grams (G). tenth of a gram. The recurrence of 9.32 grams (four
W 57. H .. 06 D .. 06. Domed stone weight. 457.00 grams
1
times) and 9.33 grams (three times) was immediately
(11). apparent (indicated by tJ; the only marked weight
W 58. L .09, \IV .. 043. Roughly sphendonoid stone weight, C'iV 2) was taken to be the unit of this standard, al-
with little tapering at ends. Irregular surface con-
tains metallic crystals. 468.00 grams (G). though it is two hundredths of a gram off at 9.30 grams.
W 59. L. .07, H .. 03, W .. 04. Stone oblong with rounded 7.30 grams also appeared frequently (indicated by t),
ends. 470.00 grams (G). as did 3.50 grams (indicated by *).
W 60. L .095, W .. 056. Sphendonoid weight of haerna- N ext it was noteel when two possible standards could
tite. Strongly tapering ends flattened at tips. be used for one weight. VV 22, for example, is di visible
501.00 grams (P).
The determination of the standards of the weights I had catalogued the first of the Gelidonya weights (\;Y 60)
as a sling bullet before being corrected by Miss Taylor; it is
seemed important not only for the study of ancient remarkable that Schliemann, without realizing their significance,
metrology, but also for any light which might be shed recorded the exact weights of two of his "sling bullets."
on the route of the ship before it sank, and its planned 10 C. \111. Blegen, et ai., Tro» I, part 1, 359, and throughout
ports of call; it was assumed that the merchant carried Troy publication; precise dimensions are, however, given. Axe!
\;y. Persson, Dendra, 65 ff.
standards used in each of the localities he wished to 11 C. Schaeffer, Uoaritica I, 44-45. Professor Schaeffer has
visit. very kindly supplied me 'with information concerning the uri-
This task was made more difficult by past excavation published weights, which will presumably be published at a
reports, for little attention has usually been given to later date.
12 Infra, ns. 36 and 37.
the study of weights by field archaeologists. In early 1.~C. C. Mcf.own, Tell en-Ncsbeli, 259. A weight inscribed
archaeological reports weights were often not recog- with the name phn is mentioned, but not its actual weight.
nized as such and some were described as sling bullets." 14- Sellers, IDE 4, 832. See also F. G. Skinner, Hist. of
Tech. I, 779 H., and VV. M. F. Petrie, Ellcy. Brit. 15 (1954 ed.},
e Heinrich Schliemann, Troia, 118; Ilios. The City and 144-145, for examples of varying conclusions reached by three
Country of the Trojans, 436-438. This is a natural error and authorities.
138 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

TABLE 1
EXAl\,PLE OF GRTD USED IN DETER,\IlNJNG COMMON STANDARDS

12 13 15
Weights 2
--
3
---
4
---
5 6
--
7
--- ---
8
---
9 10
--- --- " --- ---
14
--- ---

\VI 3.50* 1. 75 1.17 .87 .70

\V2 9.30\ 4.65 3.10 2.31 1.86 LS5 1.33 1.16 1.03 .93
\V3 9.50 4.75 3.17 2.37 1.90 LS8 1.36 1.19 1.05 .95
\V4- 10.30 5.15 3.43 2.57 2.06 1.72 1.47 1.29 1.14 1.03 .94
\,\IS 10.50 5.25 3.50* 2.62 2.10 1. is LSO 1.31 1.17 LOS .95
\1\16 12.30 6.15 4.10 3.07 2.46 2.05 1.76 LS4 1.37 1.23 1.12 1.02 .95
W7 16.00 8.00 5.34 4.00 3.20 2.67 2.29 2.00 1.78 1.60 1.45 1.33 1.23
\V8 17.80 8.90 5.94 4.45 3.56 2.97 2.54 2.22 I.98 1.78 1.62 1.48 1.37
\\"9 20.60 10.30 6.86 5.15 4.12 3.43 2.94 2.58 2.29 2.06 1.87 1.72 LS5
W10 20.80 10.40 6.93 5.20 4.16 3.46 2.97 2.60 2.31 2.08 1.89 1.73 1.60
\V11 26.10 13.05 8.70 6.52 5.22 4.35 3.73 3.26 2.90 2.61 2.37 2.17 2.01
\V 12 26.20 13.10 8.73 6.55 5.24- 4.37 3.74 3.28 2.91 2.62 2.38 2.18 2.02
\V13 28.00 14.00 9.33\ 7.00 5.60 4.67 4.00 3.50* 3.11 2.80 2.54 2.33 2.15
\>VI4- 29.20 14.60 9.73 7.301 5.84 4.86 4.17 3.65 3.24- 2.92 2.65 2.43 2.25
W15 29.80 14.90 9.93 7.45 5.96 4.97 4.25 3.72 3.31 2.98 2.71 2.48 2.29 •
WI6 35.00 17.50 11.66 8.75 7.00 5.83 5.00 4.37 3.89 3.50* 3.18 2.91 2.69 2.50 2.33
W17 36.00 18.00 12.00 9.00 7.20 6.00 5.14 4.50 4.00 3.60 3.27 3.00 2.71 2.57 2.40
W18 36.50 18.25 12.16 9.12 7.301 6.08 5.21 4.57 4.05 3.65 3.32 3.04 2.80 2.61 2.43
W19 42.70 21.35 14.23 10.68 8.54- 7.12 6.10 5.34 4.75 4.27 3.89 3.56 3.39 3.05 2.85
W20 43.80 21.90 1460 10.95 8.71 7.301 6.25 5.47 4.86 4.38 3.98 3.65 3.37 3.13 2.92
wz: 45.50 22.75 15.16 11.37 9.10 7.58 6.50 5.68 5.05 4.55 4.13 3.79 3.50* 3.25 3.03
\\122 46.00 23.00 15.33 lLSO 9.20 7.67 6.57 5.75 5.11 4.60 4.18 3.83 3.53 3.29 3.07
W23 47.70 23.85 15.90 11.92 9.54 7.95 6.81 5.96 5.30 4.71 4.34 3.98 3.67 3.41 3.18
\\'24 48.20 24.10 16.07 12.05 9.64 8.03 6.88 6.03 5.36 4.82 4.38 4.02 3.71 3.44 3.21
\V25 49.40 24.70 16.47 12.35 9.88 8.23 7.05 6.17 5.49 4.94 4.49 4.12 3.80 3.53 3.29
1'126 51.50 25.75 17.17 12.87 10.30 8.58 7.35 6.44 5.72 5.15 4.68 4.29 3.96 3.68 3.43
\N27 54.20 27.10 18.07 13.55 10.84 9.03 7.74 6.77 6.02 5.42 4.93 4.51 4.17 3.87 3.61
\V28 55.50 27.75 18.50 13.87 11.10 9.25 7.93 6.93 6.17 5.55 5.05 4.63 4.27 3.96 3.70
W29 56.00 28.00 18.67 14.00 11.20 9.33\ 8.00 7.00 6.22 5.60 5.09 4.67 4.30 4.00 3.73
1'130 57.50 28.75 19.16 14.37 lLSO 9.58 8.21 7.18 6.38 5.75 5.23 4.78 4.42 4.11 3.83
W31 59.00 29.50 19.66 14.75 11.80 9.83 8.43 7.38 6.55 5.90 5.36 4.92 4.53 4.21 3.93
1'132 63.90 31.95 21.30 15.97 12.78 10.65 9.13 7.99 7.10 6.39 5.81 5.32 4.91 4.56 4.26
W33 65.50 32.75 21.83 16.37 13.10 10.91 9.35t 8.19 7.27 6.55 5.95 5.45 5.04 4.68 4.37
'v\134 65.50 32.75 21.83 16.37 13.10 10.91 9.35 8.19 7.271 6.55 5.95 5.45 5.04 4.68 4.37

by 11.50, 7.67, and 5.75; W 30 is divisible by 11.50 to one or another standard, lies in the fact that most of
and 5.75, but not by 7.67; and VV 40 is divisible by those weights which were listed as merely "possible
7.67, but not by 11.50 or 5.75. There is no contradic- weights," including most of those of metal, do not fit
tion here, however, for 7.67 proves to be two-thirds of any pattern. It is possible that a few of our pieces may
11.50, and 5.75 is one-half of [1.50. vVe have, then, have worked out only by chance} but certainly the ma-
a shekel of 11.50 grams, a pim (2/3 shekel) of 7.67 jority may be accepted with confidence. It should also
grams, and a belea (1/2 shekel) re of 5.75 grams. Only be noted that changes in the weights from their long
one standard is involved. submersion in sea water need not be considered, for the
The units and their multiples which occur most fre- polished surfaces which some of them still retain show
quently are 7.30 grams, 9.32 (or 9.33) grams, 9.50 this to have been negligible.
grams, 10.30 grams, 10.50 grams, and 11.50 grams; It may now be shown that all of these units were of
12.30 grams fits into no pattern and is taken to be a unit common standards. To arrive at these conclusions it
weight for a seventh standard which was not otherwise was necessary to restudy the weights from the various
represented on board; SOLO grams, previously noted to land sites (using grids as for the Gelidonya weights),
have been found separate from the others, also proves to for a weight of 92.00 grams might have been described
conform to a separate standard. The proof of these sys- as 10 qedets of 9.33 grams, when in fact it is exactly
tems, for those who might suspect that the "...eight of 8 shekels of 11.50 grams. Below are listed the stand-
any piece of stone might be juggled until it conformed ards from Gelidonya, based entire1y on internal evi-
dence, with the various localities in the Mediterranean
15 Sellers, supra, n. 14. R. B. Y. Scott, Bib/Arch 22 (May where they were used, followed by the evidence for
1959) 35: " ... the beka is the only weight, the name of which each locality.
appears both in the Old Testament and on recovered weights,
and relationship of which to the shekel is given (cl. Exod. (1) 7.30 grams (found at Ge1idonya in multiples of 4,5,6,
38,26) ." 8( i), 9, 15, 20, 28, 32, 64).
 VOL. 57, PT. 8, 1967J BASS TH E \NEIGI-lTS 139
This standard agrees most closely with the so-called significant that actual qedets and debens do not occur here,
Phoenician standard of 7.32 grams. It occurs in Egypt/G but the relationship with these mainland weights is at least
Syria, Palestine, and Crete. At Ras Sharm-a the system of worth noting. Weights from Crete, as noted in the dis-
weights has been reported to be based on a mina of 50 cussion of the shekel of 7.30 grams, have been recorded as
shekels, rather than the 60 shekels of the Babylonian sys- possible qedets, but the deviation from the normal seems too
tem: 17 if the mina of 439 grams found there, however, was great to allow full acceptance.
composed of the normal 60 shekels, we would find a shekel (3) 9.50 grams (found at Gelidonya in multiples of I, 5,
of 7.31 grams. From Gezer we see weights of 7.25 and 7.35 7,8, 9( 1), 30, ancl9 X 1/2( 1»,
grams, but much closer are the weights of 43.75 grams The qedet of 9.5 grams occurs in Syria, Cyprus, Palestine,
(6 X 7.29) and 182.75 grams (25 X 7,31).18 A lead weight and Crete. At Ras Shamra is a weight of 9.5 grams.33
from the Palace at Knossos equals 22.05 grams, or 3 shekels At Enkorni on Cyprus are weights of 18.996 grams, or a
of 7.35 grams.!" and also from Crete is a lead-filled bronze double qedet of 9.498 grams; and 47.756 grams, or 5 qedets
head of an ox which weighs 73.62 grants,"? which could of 9.551 grams.:" A weight of 95.26 grams, found at Gezer,
be either 8 qedets of 9.20 grams or, more likely, 10 shekels is marked on its back with a straight scratch (1 deben},
of 7,36 grams A limestone cylinder, from Zakro on Crete, and another weight from the same site is 95.22 grams.v"
is marked with 6 dots and weighs 220 grams; the dots The famous gypsum weight with octopus reliefs found in
indicate that the weight is six times a unit of 36.66. and this the Palace of Knossos has been thrice recorded by Evans
is described by Evans and Glatz 21 as being the equivalent as being a talent weighing ;'a11110stexactly" 29,000 grams.v''
of 4 qedets of 9.165 grams. It seems more likely that this but its weight has also been recorded as 28,600 grams.et
• was in fact a unit of 5 shekels of 7.33 grams. If these This latter weight would yield a unit of 9.53 grams if the
calculations al-e now accepted it is evident that none of the talent is 60 minas of 50 units each." On the other hand,
variations of the standard deviate more than six one- if the mina is composed of 60 units on Crete, we may judge
hundredths of a gram, while earlier conclusions were based the stone weight from Soteia (1,140 grams) SD to be two
on variations of at least sixteen one-hundredths of a gram minas based on a unit of exactly 9.50 grams; this would
for the qedet iust mentioned. indicate the possibility that the "octopus relief" weight was
(2) 9.32 (or "9.33) grams (found at Gelidonya in multi- possibly not a weight at all,40 and it has been suggested that
ples of 1, 3, 6, 7( 1), 10, 19, 20( 1),25,30,49, 50( 1), it was in fact a stone auchcr,"" although the material of
and 25 X 1/12), which it was made makes this last theory unlikely. At any
This is the most common weight for the qedet, the na- rate, the suggestion of 9.50 grams as the unit on which the
tional Egyptian standard." (1/12 qedet = 1 seal or S(J}I, Soteia weight was based requires less "juggling" than the
10 qedets = 1 deben, 10 debens r- 1 sep).:!.3 It occurs in suggestion that it was twelve dozen shekels of 7.916
most countries of the eastern Mediterranean, including grams." The confusion arises from the use of minas of 25,
Syria, Cyprus, Palestine, the Hittite empire, TroYI and pos- 50, and 60 shekels in antiquity, and all must be tried on
sibly Crete and Greece. At Ras Sbamra is a weight of unmarked weights.'l-3
18.7 grams, marked as a double unit with two strokes.w (4) 10.30 grams (found at Gelidonya in multiples of 1, 2,
Multiples of 9.33 grams have been found at Enkomi on 5,18, and 7 X 1/2 anel7 X 2/3),
Cyprus."! At Gezer are a qedet of 9.36 grams and A nesef of 10.37 grams existed in Syria."-
c1ebens of 93.69 and 93.45 grams,211 at Megiddo a weight of (5) 10,50 grams (found at Gelidonya in multiples of 1,
28.00 grams 2, (3 qedets of 9.33 grams), and at Tell en- I X 1/3, 13 X 1/3, 5 X 1/2, IS X 1/2, 4 X 2/3, 5 X
Nasbeh a 'weight of 9.324 grams (marked as a nesef, but 2/3,7 X 2/3, and 10 X 2/3),
indicating that a unit of 9.32 grams was used).28 A This is perhaps another form of the Palestinian nesef
weight from Bcgazkoy, of 140 grams.w may be considered mentioned above, which has been found ranging between
as 15 qedets of 9.33 grams, and from Troy are two weights 10.0 and 10.50 grams." It occurs at Tell en-Nasbeh.w as
of half qedets of 9.33 grams."? For Greece the evidence is a beka of 5.25 grams in Cezer," and as a third of a shekel
less solid: a weight from Corinth 31 is lacking its bronze at Ras Shamra.O In Greece, at Malthi, a weight of 632.00
nng (similar to rings on weights from Ras Shanu-a ) and
weighs 18 grams (a double qedet with the addition of the 33 Schaeffer, Syr-ia 18, 150-151.
rillg?). At Malthi in Messenia 32 are Late Helladic weights 34 Evans, Corolla Nwm., 350-351.
of co. 7.77 grams (10 seals) and its multiples (7.80 grams. as Macalister, Ceser II, 282-283.
31 or 4 X 7.77 grams. 308.80 or 40 X 7.72 grams); it is 36 Evans, PM IV, part II, 651; BSA 7, 42; Corolla N1un ..
342,
Hi G. Glatz, Acgean Civil .. 192. 37 Xanthoudides, ATchEph (1906) 152; Glatz, Aegean Cwit.,
1. Schaeffer, Uqaritico I. 44-45. 191,
18R. A. S. Macalister, Ge:;l'r II, 280-283. 38 Evans, Coral/a N-ll1l1.,. 343.
10 Arthur Evans, Corolla NII-m., 348. so Xanthoudides, ArchEph (1906) 151 ff., pI. XI, 14.
20Evans, FNI IV, part II, 655, and Coral/a Nul/l." 353. 40 EvallS was overly fascinated by the weight of 29,000 grams,
21 Evans, Corol1a NurJI.., 346; Glatz, Aegean Civil.> 193, which he supposed to be a unit used in making copper ingots
22Petrie, Enc)l. BTit. 15 (1954 ed.), 144. This figure IS (CMol/a, Hu'/ll. 357-359). The demonstration that ingots were
almost universally accepted as the weight of this standard. not made in accordance with a precise weight (supra, pp.
~3 Alan Gardiner, Egypt. GmJn1llar, 200; Petrie, Anc. Weights 7l f.) removes some of the basis for the theory that this
alld iV/ eas .. 13. was actually a weight.
24 Schaeffer, Syria 18, 150. 41 H. Frost, Under thc Mediterranean, 46.
25 Evans, Corolla- NWJII.. 350-351. 42 Glatz, Aegean Civil., 192.
26 Macalister. Geze-r n, 282-283. 43 Skinner, H·ist. of Tech. I, 780; Sellers, Inferp. Dict. of
Z7 Lamon and Shipton, Megiddo I, pI. 104. the Bible 4, 830-831.
28 McCown, Tell ell-Na,~be/r 1, 259. 44 Skinner, [-rist. of Tech. 1, 781.
29 K. Bittel, Bofja.zkoy III, 32. 45 Petrrc, Ency. Br·it. 15 (1954 ed.), 144.
30 Petrie, Anc. We-ights and Mea ..., 43, and p1. 49, 46 McCown, Tell en-No,ybeh I, 276.
31 G. R. Davidson, Cm'iJlth XII, 204. 41 Macalister, Ge:::er TI, 282-283.
32 N. Valmin, Paid .. prehistoy-iqllcs ... , 30-34. 48 Schaeffer, Syria 18, 150.
140 CAPE GELIDONYA, A BRONZE AGE SHIPWRECK [TRANS. A~lER. PHlL. SOC.

... e? Bogazkoy

•
Syrlo

0-
Babylonia

hUe" N.,beh
..0•
.o.400e
Gezer

~ • X Egypt

DISTRIBUTION OF WEIGHTS: 1:. 7.30grams o 10.50

... 9.32
• 11.50
o 9.5 X 12.30
.10.30 <0501.00
FIG. 149.

grams equals 60 units of 1.0.53 grams, or perhaps a slightly grams (one beka) and 115.7 grams (10 shekels of 11.57
inaccurate mina." grams).54 This weight may have continued as the Persian
(6) 11.50 grams (found at Gelidonya in multiples of 4, silver standard of 11.53 grams.55
5,6, 8, and 15 X lI2, 10 X 2/3, and 13 X 2/3). (7) 12.30 grams (found at Gelidonya in multiples of 1, and
The Hebrew shekel has been calculated as being around possibly 4, 7, 9 X 1/2, and l I X 1/2).
11.30 to 11.47 grams by some authorities.v'' but Scott reckons Were it not for the well-preserved weight of 12.30 grams,
the common shekel to have been 11.7, depreciating to 11.4 this would 110t be considered a standard at Gelidonya.
grams.51 Vile find examples of this at Gezer in shekels of Evans gives this (12.25 grams) as the gold unit of Crete
11.51, 11.54, and 11.56 grams, and bekas of 5.78 ancl 5.77 and Egypt,56 but Glatz's statement that it occurs in Egypt,
grams, as well as a double shekel of 23.02 grams.5~ A Crete, Cyprus, and Palestine, between 12.30 and 13.98
weight of 7.65 grams from Tell en-Nasbeh may be COI1- grams, is perhaps too loose for consideration here."
sidered as one pim.t" At Bogazkoy are weights of 5.76 (8) 501.0 grams.
4\) Valmin, Poids prehisto'riques ... , 34.
50 Sellers, IDB, 831. 5'1 Bittel, Bogazkoy III, 32.
51 Scott, BiblArch 22, 39. ;'5 Skinner, Hist. of Tech. I, 781.
52 Macalister, Ceser II, 280~281. ,,6 Evans, PM IV, part II, 665.
5:1 McCown, Tell en-Nesbeli I, 276. :i1 Glatz, Aegean Cioit., 193.
 VOL. 57, PT. 8, 1967] BASS: THE WEIGHTS 141
This is the Babylonian mina of 502 gt-ams.t" There were \1I,T 27. 54.20 grams. Conforms to no standard. Probably
no other examples of this weight and its parts on the not a 'weight CV1).
Gelidonya ship, but as has been noted, it does not seem to W 28. 55.50 grams. 9 bekas of shekel of 12.30 grams (6.17
have been in use in the cabin. We can offer no explanation for 615) (M).
for its having been on board unless, perhaps, it was left W 29. 56.00 grams. 6 qedets of 9.32 grams (9.33 for 9.32)
over from an earlier trip, or some infrequent commodity (G)
was sold only by the mina of S02 grams. W 30. 57.50 grams. 5 shekels of 11.50 grams.
W 31. 59.00 grams. Possibly 8 shekels of 7.30 grams (7.38
We may now present a list of the Gelidonya weights for 7.30) (G).
with their assigned values. In this we have borne in W 32. 63.90 grams. Conforms to no standard. Un-
mind the statement by Petrie that "in general we should finished? (e).
VV 33. 65.50 grams. Approximately 7 qedets of 9.32 grams,
not accept any multiple which is unlikely, such as 11,
using estimated original weight (G).
13, 23, 28, 33, 46, which all appear as supposed mul- W 34. 65.50 grams. 9 shekels of 7.30 grams (7.28 for
tiples in a recent paper on weights." 69 A close look 7.30) (G).
at the evidence leaves us with 110 choice but to accept '.IV 35. 66.50 grams. 7 qeclets of 9.50 grams (M).
some such multiples. It is certain that most of the VV 36. 67.S0 grams. Possible stone weight, broken. Maybe
11 bekas at shekel of 12.30 grams (6.14 for 6.15)
weights belonging to the merchant were recovered, and (M).
• it is clear that no set was "complete," as having 1, 2, W 37. 69.00 grams. 6 shekels of 1150 grants
3, 4, 5, 10, 20, 30, etc., units. Is it not possible that the W 38. 69.80 grams. 10 puns of shekel of 10.50 grams
merchant knew what combinations of various un- (6.98 for 7.00) (G).
W 39. 76.00 grams. 8 qedets of 9.50 grams (G).
matched units would equal the desired and missing fig- VV 40. 76.70 grams. 10 pims of shekel of 11.50 grams.
ures? In this case the multiples may not be as arbitrary W 41. 79.?? grams. Possibly 15 bekas of shekel of 10.50
as they seem, but were based on usage over a period grams (S).
of years. W 42. 86.00 grams. ] S bekas of shekel of 11.50 grams
(11.46 for 11.50), or 9 qedets of 9.50 grams (9.55
WI. 3.50 grams. lj3 shekel of 10.50 grams (M). for 9.50), or 7 shekels of 12.30 grams (12.28 for
W 2. 9.30 grams. 1 qedet of 9.32 grams (M). 12.30) (G).
W 3. 9.50 grams. 1 qedet of 9.50 grams (1\11). W 43. 92.00 grams. 8 shekels of 11.50 grams.
W 4. 10.30 grams. 1 shekel of 10.30 grams. VV 44. 93.20 grams. 1 deben (10 qedets of 9.32 grams)
W 5. 10.50 grams. 1 shekel of 10.50 grams (M). (M).
W 6. 12.30 grams. 1 shekel of 12.30 grams. W 45. 99.60 grams. 13 pints of shekel of 11.50 grams (7.65
W 7. 16.00 grams. Conforms to no standard. Probably for 7.67) (e).
not a weight (M). W 46. 109.50 grams. 15 shekels of 7.30 grams (M).
\V 8. 17.80 grams. Conforms to no standard. Probably W 47. 146.40 grams. 20 shekels of 7.30 grams (7.32 fat"
not a weight (1\'1). 7.30) (G).
W 9. 20.60 grams. 2 shekels of 10.30 grams (G). W 48. 177.00 grams. 19 qedets of 9.32 grams (e).
W 10. 20.80 grams. Standard not apparent (M). W 49. 185.50 grams. 18 shekels of 10.30 grams (10.31 for
W 11. 26.10 grams. Conforms to no standard. Probably 10.30) or two debens (20 qedets of 9.28 for 9.32
not a weight (G). grams) (M).
W 12. 26.20 grams. S bekas of shekel of 10.50 grams (5.24 W 50. 188.00 grams. 2 deb ens (20 qedets of 9.40 grams)
for 5.25). (M).
VV 13. 28.00 grams. 3 qedets of 9.32 grams (9.33 for W 51. 194.00 grams. 25 seals (25 twelfth-qedets of 9.33
9.32), or 4 pints of shekel of 10.50 grams (M). grams) .
W 14.29.20 grams. 4 shekels of 7.30 grams (G). W 52. 204.00 grams. 28 shekels of 7.30 grams (7.28 for
W 15. 29.80 grams. Conforms to no standard. Probably 7.30) (P).
not a weight (G). W 53. 233.00 grams. 21 debens (25 qedets of 9.32 grams)
W 16. 3S.00 grams. 5 pims or 10 third-shekels of 10.50 or 32 shekels of 7.30 grams (7.28 for 7.30) (M).
grams (G). W 54. 244.00 grams. Original weight not known exactly.
'vV 17. 36.00 grams. 7 bekas of shekel of 10.30 grams Perhaps 26 qedets of 9.32 grams.
(514 for 5.15) (e). W 55. 279.50 grams. 3 clebens (30 qeclets of 9.32 grams)
\\1 18. 36.50 grams. 5 shekels of 7.30 grams (G). (M).
\V 19. 42.70 grams. 9 half-qedets of 9.50 grams, or pos- W 56. 284.50 grams. 3 deb ens (30 qedets of 9.50 grams)
sibly not a weight (G). (9.48 for 9.50) (G).
\V 20. 43.80 grams. 6 shekels of 7.30 grams (G). W 57. 457.00 grams. 49 qedets of 9.32 grams (M).
W 21. 45.50 grams. 13 third-shekels of 10.50 grams (G). W 58. 468.00 grams. 5 debens (50 qedets of 9.32 grams)
'vV 22. 46.00 grams. 4 shekels of 11.50 grams. (9.36 for 9.32) or 64 shekels of 7.30 grams (7.31
VV 23. 47.70 grams. 5 qedets of 9.S0 grams (9.54 for for 7.30). Corroded weight (G).
950) (G). W 59. 470.00 grams. 5 debeus (50 qedets of 9.40 grams)
W 24. 48.20 grams. 7 pims of shekel of 10.30 grams (6.88 (e).
for 6.87) (G).
vv60. 501.00 grams. ] Babylonian mina (8.35 X 60, for
W 25. 49.40 grams. 7 pints of shekel of 10.50 grams (7.0S
for 7.00), or 4 shekels of 12.30 grams (12.35 for
=
8.37 X 60 502.20 grams) (P).
12.30). As has been seen, most of the doubtful weights were
W 26. 51.50 grams. 5 shekels of 10.30 grams.
discarded as not being weights at all, but a few irregular
58 A. Scgre, JAOS 64, 73. pieces remain. It is possible that these reflect standards
59 Petrie, Aile. We·ights and Meos., 7. only by chance, but their inclusion is based on modern
142 CAPE CELIDONYA, A BRONZE ACE SHIPWRECK [TRANS. A~1ER. PHIL. SOC.

usage: today in Bodrum, Turkey, merchants may be standards would have been used for weighing the
seen using irregular, but carefully weighed, field stones ingots; certainly one large weight was needed for this
to replace lost iron weights in their sets. purpose.
One problem remains to be clarified. Small weights
CONCLUSIONS
were suitable for weighing tools, weapons, and small
scraps of metal." but were far too light to be of use in It is immediately apparent that weights of the Late
weighing ingots and large ingot fragments. A large Bronze Age were much more accurately made than has
stone," possibly an anchor or anvil;" was found lying been supposed;" The frustration of finding weights
on planks of the ship in area S. In weighing all of the bearing identical identification marks, but of varying
objects from the ship, with the sole intent of estimating weights, seems to have been due largely to the fortunes
its tonnage, this was recorded as weighing just under 74 of land excavation, where mixtures of weights of dif-
kilograms. The white ballast stones were weighed as ferent localities and dates are sometimes difficult to sepa-
a unit, for individual weights seemed unnecessary, but a rate. We see that of fifty-two weights belonging to one
squarish diorite stone (.022 X .018 X .016) from area man and being used in the same year, twenty-six are ac-
P IV was catalogued separately because of its size and curate to within one-hundredth of a gram or less per unit.
material; this was also weighed, with no particular pur- Six others are inaccurate by only one-hundredth of a
pose other than to add to the estimate of the 'weight of gram per unit. Only the presence of two quite accurate
the total cargo, and proved to he 10.5 kilograms. weights, W 50 and \V 59, based on a qec1et of 9.40
The possibility arises that these were also weights, grams, halfway between the qedets of 9.32 and 9.50
and they are now tentatively added to our list (see also grams, is difficult to explain. Balances were obviously
chap. 7) , more sensitive than previously supposed;" and weights
could be made with great exactness.
\'tl,r 61. (also St 5) Ca.. 73,900 grams. Possibly two talents The Gelidonya merchant had an intimate knowledge
of 60 minas of 60 shekels of 12.30 grams (73,800
grams). This standard has been called the "gold of the weights, for none are marked and none of the
standard," 63 which brings to mind the statement shapes seem to correspond particularly to any standard.
made by Evans that "it can hardly be doubted that Only the standard of 9.32 stands out. for most of the
some of the tablets from the Palace of Knossos qedets are sphendonoid and most of the debens are
relate to equations between the value of these bronze
'talents' and a gold unit." 64 domed; even here there are exceptions. It is possible
VV 62. (also St 7) 10.500 grams. This would be one that each set 'vas kept in a separate bag, but the scat-
thousand nesefs of 10.50 grams, which is a standard tered proveniences of the weights on the site offer no
found at Gelidonya, and a standard which was evidence for this theory.
decimal. "multiplying up to 1,000." 0;>
The weights show that the ship was capable of trading
Possibly only one of the two stones should be con- with merchants from Egypt, Syria, Palestine, Cyprus,
sidered as a weight, for it seems unlikely that two Troy, the Hittite Empire, Crete, and probably the Greek
mainland. This tells us virtually nothing of the route
;;0 Similar weights have also been found with bronze scraps
on land; Macalistcr, Cezcr II, 283.
being followed by the ship, but this very freedom of
t11 Supra> p. 36, St 7. choice is significant in the study of ancient trade. (See
U2 Supra> p. 128. fig. 149.)
63 Evans, (.'II/pm> 11.56).
loco cit.
6,j. Evans, Corolla, Nwm., 361. See, however, supra, p- 71. 66 Skinner, Hist. of Tech. I, 781 ; Sellers, IDB 4, 832.
65 Petrie, Ene)l. Brit. 15, (1954 ed.), 144-145. ur A. E. Berriman, lEA 41 (1955) SO.
 XIV. CONCLUSIONS'

GEORGE F. BASS

At the close of the Bronze Age in the Eastern Medi- All of the elements necessary for bronze-making
terranean, a nine- or ten-meter merchantman sailed were on board-pure copper, pure tin, and bronze
toward Cape Gelidonya, the southernmost point of the scraps and ingots for being recast-and it is likely that
Anatolian Peninsula. Her last major port of call had the merchant not only traded with such raw materials,
been in Cyprus, where the crew had loaded on a ton but .....
vas a tinker who worked with them himself. Three
of metal cargo. Four-handled copper ingots, weighing large, hard, close-grained stones .....
vould have served as
about twenty kilograms apiece, were carefully wrapped anvils. A small bronze anvil was broken in antiquity
in matting and stacked in neat piles fore and aft, and and was only part of the cargo of scrap, which might
bronze bun ingots were stacked among them. Wicker suggest that a bronze swage block near the bow of the
baskets filled with bits and pieces of broken ingots and ship was also only part of the cargo were it not for
broken bronze tools were placed wherever they would a number of other items in the "cabin" area. There
fit on the layer of brushwood which protected the thin rested two stone "maceheads," which might very well
planks of the hull. Tin ingots, scraps of lead, and have been examples of the "highly polished and deli-
• pieces of uuworked crystal may already have been on cate" 6 perforated stone hammers used for forging in the
board. A jar full of colored beads and a bracelet of Bronze Age.' Nearby were more than half a dozen
adjustable size were items of trade, near the bow. The stone rubbers and polishers, with a whetstone which
contents of other jars were perishable, but they may hung, perhaps from the merchant's neck, by a string.
have included spices.' In the "cabin" area also were a bronze cold chisel and
At least the more important members of the crew a punch. If only the bronze tools had been on board,
lived near the stern, in an area lit at night by a single there would be the possibility that all had been but
oil lamp. Their meals, as those of Mediterranean sail- part of the cargo, for such tools, broken and complete,
ors today, were supplemented with olives and, perhaps, often appeared in contemporary founders' hoards. The
fish which they had caught with lead-weighted lines. existence of the stone implements, however, makes this
Storage jars probably held other food, and wine and unlikely. All that was missing from a traveling smith's
water; skins and leather bags may also have contained complement of material" were molds, and these may
food, but this we have no way of knowing." have existed in the form of soft clay 'which was washed
On board was a merchant, prepared to trade in al- away by the currents after the ship sank. Only ham-
most any Eastern Mediterranean port. His balance-pan mering and sharpening and polishing were done on
weights, necessary for judging the price of metal in
shore of Crete, and thus come up to Peloponnesus from the
either scrap or finished form, allowed him to cleal with south.c-.a route more than twice as long for Diomed, but keel)-
other merchants in Egypt, Syria, Palestine, Cyprus, ing close to land all the way. Indeed they did not dare 'cleave
Troy, the Hittite Empire, Crete, and probably the the open sea' until they received a guiding omen from the
Greek mainland. For official transactions he carried gods." Seymour, Life 1·n the Homeric Age, 305.
G Coghlan, Preh; Met. of Copper cuui Bronze, 77 (see quote,
his personal seal, a cylinder which seems to have been infra, n. 8).
an heirloom handed down from generation to genera- 7 Coghlan in Hist. of Tech. I, 609. Guido, Sardinia, 153,
tion. He, or perhaps members of the crew, also carried mentions the frequent association of maceheads with mining.
five scarabs as sea1s, souvenirs, or religious talismans, The Gelidonya maceheads have been considered too finely made
for. rough work, such as forging, but only a highly polished
Religious charms were then, as now, of comfort to hammer could produce a smooth, undented surface on metal.
travelers/ anel an astragal in the cabin, if it was not 8 Coghlan, op, cit. (supra, 11. 6), 76-77: "The basic tools of
used to while away the hours in the game of knuckle- a modern smith are the forge, anvil, hammers, tongs, chisels,
bones, offered divine guidance to the captain when he metal saws, fullers, flatters, swages, and a variety of punches.
was in doubt." The fullers, swages, and flatters which we illustrate are those
of the iron-smith and are generally made and used for far
1 Statements 110tannotated in this chapter have been discussed heavier work than was ever dealt with in prehistoric days.
more fully elsewhere in the book. However, there is 110 doubt that the prehistoric smith would
2 Ventris and Chadwick, Documents, 135-136. have had miniature tools of somewhat similar form with which
a Odyssey, 2.354, 5.266. to carry out his more advanced operations. .. The forge,
<I Wen-Amen carried an idol, a "traveling Amon," on his anvil, and hammers were very simple . an ordinary fire,
eleventh-century voyage from Egypt to Phoenicia (Pritchard, suitably walled in, would serve for the forge, especially as
ANET, 26, n. 12, trans. by John A. Wilson), just as modern some measure of artificial draught must have been applied.
travelers carry St. Christopher's medals. For the date of the A heavy block of bard, flat, and close-grained stone would
trip see Albright, Al A 54 (1950) 174, with n. 44. serve as the main anvil; later, for the execution of fine work,
5 "On their return from the siege of Troy, Menelaus, Nestor, the well-known tanged anvil of the Bronze Age came into
and Diomed at Lesbos (-y 169) 'pondered over the long voyage,' use ... excellent hammers appeared as soon as bronze came
-c-whether to sail across the open Aegean Sea to the southern in. There is ample archaeological evidence to show that stone
end of the island Euboea, about 110 miles, with the island of hammers were also in C0111010n use . Very frequently the
Psyria for their encouragement, about mid-way.c-cr to follow perforated stone hammer was no rough tool but a highly
the coast of Asia Minor, to the south, and then to skirt the polished and delicate instrument."

163
164 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

board; furnaces were quickly made of stone and clay been cargo only. The lamp from the cabin area, hO\"1-
at various stops along the route." ever, was surely the ship's lamp, which presumably was
The ship, taking the C01111110n route from Syria and treated with care, and it is strictly Syro-Palestinian in
Cyprus to the Aegean," followed the coast and took form.
advantage of the westward current which rushes "with Although the \...'eights would have allowed the ship to
augmented violence towards Cape Khelidonia, where, trade in almost any port of the Near East or the Aegean,
diffusing itself in the open sea, it again becomes equal- we have seen that the majority of the standards are
ized," 11 In only a few hours the crew could be taking N ear Eastern, and that the shapes are best paralleled
on fresh water from the abundant supply at Phoenikous outside of Mycenaean Greece.
(modern Finike}, but first the cape and its string of The stone maceheads and mortars were used by the
tiny islands, later described by Pliny as "extremely crev v, and their origins, according to Miss Taylor's
dangerous to mariners," 1:0 had to be passed. The ship study, are also to be sought along the Syro-Palestinian
never made it. Attempting to sail between the two coast 01- on Cyprus.
large islands nearest the mainland, she ran onto jagged Finally, Miss Western has pointed out that the con-
rocks and sank in nearly thirty meters of water. dition of the wood does not permit definite conclusions
about the original port of the ship, but the fibers used
This much of the story 01 the ill-fated voyage may in the basketry and matting came mainly from the East.
easily be deduced from a study of the finds, but a more The beads may not be considered personal posses-
detailed study reveals that the ship was a Phoenician 13
sions, for there is nothing to indicate that they were
merchantman which sank around 1200 H.C. while sailing
other than cargo for trade. Not only were they stored
toward the Aegean. in a jar with a bracelet, but they were the only possible
The main cargo of bronze and copper, as we have "personal possessions" which did not come from the
shown, was picked up in Cyprus, but such a cargo could "cabin" area, with the exception of the stone mortars.
have been carried in a Mycenaean, Cypriot, Phoenician, Thus we see that the individual bits of evidence,
Sea-Peoples, or Egyptian bottom; all of these people
studied simultaneously but independently by a number
are known to have been seafaring to a lesser or greater
of scholars, all lead to the conclusion that the ship at
extent during different periods of antiquity. It is not an Gelidonya sailed with a Syrian merchant from a Syro-
examination of the cargo, however, but of the personal
Palestinian port.
possessions of the crew and merchant on board which
will lead us to discover the nationality of the ship. The approximate date of the calamity which befell
The cylinder seal, according to Buchholz, is Syrian. the ship may be reached also by a number of means:
He concludes that, as it was the only one on board, it Miss Ralph states that the most acceptable dates ob-
was not an item of trade but the personal seal of the tained from a carbon-14 measurement of the brushwood
merchant. Indeed, it would have been strange had on the ship is 1200 B.C. ± SO years.
there not been one on the ship, for every Eastern Hennessy and Miss Taylor arrived independently at
merchant had his o\V11 seal.!" the same date in their study of the pottery; they prefer
Schulman, in his study of the scarabs, has shown that a clate after 1200 B.C., but point out the difficulty of
they are almost certainly Syro-Palestinian imitations of offering more than a suggestion.
Egyptian scarabs. He points out that they might have The cylinder seal seems to be much older than the
been trinkets picked up by the sailors. The importance shipwreck and provides no evidence for dating, but
of divine protection for men of the sea is known, how- Schulman feels that the scarabs suggest "the thirteenth
ever, from the eleventh-century statue of Amen-of-the- century B.C. as the most probable date of the wreck,
Road carried by Wen-Amen to the well-known libations although a slightly earlier or slightly Jater date is
of the Greeks, and it seems most likely that the scarabs possible."
were more than mere souvenirs of a voyage to the East. We have seen further that the latest evidence for
The pottery, as Hennessy and Miss Taylor have oxhide ingots is their representation at Medinet Habu
stated, could have been obtained in a number of Syro- in the first half of the twelfth century, but at least one
Palestinian or Cypriot ports, and some of it may have and probably all of these scenes showing ingots were
copied directly from the thirteenth-century Ramesseum,
I} Catting, Cypriot Bronzeuiork , 302: "The Mathiati hoard
suggests an itinerant craftsman who set up a temporary foundry There is, therefore, no positive evidence for such ingots
and smithy at the settlements he visited, rather than a local later than the thirteenth century, although ingots of the
industry." same type appear at least two centuries earlier.!"
10 Lorimer, Homer and the Motunnents, 60-61, described the The bronzes have posed a greater problem. Catling
route, even mentioning the fresh water supply near Gelidonya,
dates them to the twelfth century, but I believe that the
before the wreck was found at Gelidonya.
11 Beaufort, Karamasuc, 42. 15 The discovery by C. Schaeffer at Enkorni, after this
12 Pliny, V.35. chapter was written, of a god standing on an oxhide ingot
1& I prefer to use here the word "Phoenician," in its broad (Antiqttity, 39, p. 56) has not yet been fully published; but
Greek sense, rather than "Canaanite;" the two words are the date of that ingot seems to fall in the opening years of
"historically, geographically, and culturally synonymous" (Al- the twelfth century B.C. (its date is compared to that of the
bright, The Bible Q,Hdthe Anc. N.E., 328). statue of another god from Enkomi, found and published by
14 E.g., G. R. Driver, Sem·itic Writing, 15, 63. P. Dikaios, AA 1962). Addendum: see AOT 21 (1966) 59 ff.
 VOL. 57, PT. 8, 1967] BASS: CONCLUSIONS 165

very early twelfth century is the latest that they may be Gelidonya ship, there would be but one or two seals
dated, and feel that a thirteenth-century dating is more with each cargo of hundreds of items, whether these be
probable. pottery or metal. Further, the seal would remain in the
In summation, several independent methods of dating foreign port most probably only if it were lost or if the
have led to a date for the shipwreck of 1200 B.C. ± 50 merchant died while abroacl. The distribution of Syrian
years, with more eviclence pointing to an earlier rather cylinder seals, therefore, may be stronger evidence for
than a later dating in that range. Near Eastern trading ventures than the distribution of
Mycenaean pottery for a monopoly of Mycenaean
Advocates of a 1\'1ycenaean monopoly on sea traffic at
shipping.
this early date 'will be able to explain away each of
Phoenician merchant ships, including that at Geli-
the personal items, saying that the cylinder seal and
donya, would not have returned with empty holds to
scarabs 'were trinkets picked up with the jar of glass
Cyprus and the Near East, and it is reasonable to as-
beads en route, that the lamp was a replacement for an
sume that their cargoes consisted largely of Mycenaean
original broken or traded during the voyage, that the
pottery which often contained perishable goods. But
weights were necessary for an~ymerchant who might be
what had the ships originally carried westward on their
trading in Near Eastern ports, and that the stone ob-
• jects and basketry were also purchased in foreign ports.
outbound voyages? No longer must we "resort to
guessing the nature of the Syrian merchandise received
Thus the ship could be Mycenaean or Cypriot as well as
in exchange" for Mycenaean goods." No longer can
Syrian. I must emphasize, therefore, that there is not
we say that "few actual oriental objects have as yet been
a single element to suggest that the ship was other than
found in Mycenaean Greece." 18 It was metal, above
Syrian, and our present conclusions were reached with
all, that arrived in Greece, on ships such as that which
no preconceived ideas of the nature or date of the wreck,
sank at Gelidonya while carrying its cargo towarcl the
or of Bronze Age shipping in general.
Aegean.
The conclusion that the ship at Gelidonya was a
I have shown that copper oxhide ingots were dealt
thirteenth- or very early twelfth-century Phoenician
with by Semites, and not by Aegean merchants as com-
merchantman, however, applies to but one ship. This
monly supposed. Hi Buchholz has suggested that bun
alone has little historical significance, for a single ship,
ingots also were associated with Syrians;" but that
found by chance, cloes not necessarily represent a
primitive Iorm of bronze seems too widespread to be
merchant fleet. More significant, the excavation at
assigned to anyone people.
Gelidonya has led to a careful restudy of the types of
Hoards of bronzes on the Greek mainland, often con-
objects carried on board, and this restudy of paralJel
taining fragments of oxhide ingots, may now be accepted
material from other sites, even without the finds at
as Phoenician merchandise. The ingots and broken
Geliclonya, would have Jed to the conclusion that a great
tools were, to be sure, usually picked up in Cyprus, but
deal of commerce was in the hands of Phoenician sea-
the bronzes were most often non-Aegean types which
men and merchants during the Late Bronze Age.
had originated in the Near East, and we have further
The distribution of Mycenaean pottery along coasts
shown good cause to believe that the oxhic1e ingots were
and navigable rivers had fostered the prevailing view
made to Phoenician specifications.
that, as S. Immerwahr has written, "Mycenaean com-
Ivory and gold and cloth and spices also came from
merce would thus seem to have been almost entirely in
the East, as indicated by Semitic words for these items
the hands of Mycenaean traders and to have travelled
found on Linear B tablets;" and at least the latter may
very little farther than they could sail in their seagoing
have formed part of the contents of the Canaanite jars
ships." 16 That the pottery was transported by sea can-
which arrived in Greece during the fourteenth and thir-
not be doubted, but a cargo does not indicate the na-
teenth centuries."
tionality of the ship which carries it; the cargo at Geli-
The absence of greater amounts of Near Eastern pot-
donya, for example, was certainly Cypriot, but there is
tery in Greece is explained by the nature of this mer-
nothing to suggest that the ship also was Cypriot.
chandise. for metals and cloth and ivory do not require
Ignored in the theory of a Mycenaean leadership in
pottery containers. Indeed, we should expect to find
maritime trade are the equally important "vest bound
few traces of any of these goods in the Aegean, for the
cargoes, which we will discuss below.
cloth and spices have long since disappeared, and the
Far more indicative of the nationality of a ship than metals and ivory would be found in altered form if pre-
its cargo are, as we have suggested, the personal or served at all. This is borne out by excavation; Near
official possessions on board. Buchholz's study of the Eastern ingots and implements appear mostly in found-
distribution of cylinder seals has shown that seals in the
17 Albright, o p, cit. (supra, 11.13) 337.
Aegean during the second millennium B.C. are 1110st 18 Barnett, The Aegean and the N.E., 214.
closely associated with Syria, and that they are found III Dussaud had earlier recognized a Phoenician collaboration
along the sea route from the Levant to Cyprus to with Cyprus in the ingot trade. Schaeffer, Enkomi-Atasia, 2-4.
Southern Anatolia to the Aegean. Naturally the seals ec Buchhola, PZ 37 (1959) 15.
21 Barnett, loco cit. (supra" n. 18); Ventris and Chadwick,
are not as numerous as pottery finds for, as on the
Documents, 135-136.
16 S. Immerwahr, Archaeology 13 (1960) 6. See also foot- 22Grace, "The Canaanite Jar," The Aegcan and the N.E.;
note 41, infra. Wace, BSA 50 (1955) 179.
166 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRAi\'S. AMER. l'HIL. soc.

ers' hoards which had, for one reason or another, been temporary Gelidonya ship, for it surely marks the period
lost before being hammered or recast into typically when copper and bronze production was at its height.
Aegean forms. We may conclude that the extent of There 1S nothing to suggest that the pottery accompanied
Mycenaean shipping has been highly overrated simply Mycenaean colonization; non-ceramic remains, in fact,
because her chief export commodities, pottery and goods show that Cyprus was almost totally outside the sphere
shipped in pottery, left such durable remains. Although of Mycenaean influence." Interestingly, the cult places,
the goods received in exchange by the Mycenaeans are as on Sardinia, seem to have a Near Eastern flavot.:"
no longer so immediately apparent, they were certainly At the close of the thirteenth century, destruction on
of equal value to the people who traded them. Cyprus and the introduction of new elements, including
I cia not suggest that the Phoenicians heJel a monopoly LH IIIC pottery and ashlar masonry, have suggested
on maritime trade during the Late Bronze Age, but that the arrival of some Mycenaean immigrants for the first
they played a major part in it. Albright has sum- time. Cypriot culture was not crushed by the invasion,
marized the history of "the Phoenician problem" during however, and local pottery continued in use, except at
the last century, and has discussed the varying roles Kition;" alongside the newly introduced LH IIlC
allotted to the Phoenicians by past historians i " the wares. What did disappear at this time, H we may
implications of our present conclusions, that Phoenician judge from our completely independent conclusions con-
commerce was widespread at a date much earlier than cerning the production of copper ingots, were the
that now generally accepted, are many, and we may but Phoenician refineries. The reason for this is clear,
touch briefly here on some of these implications which for the invaders, whom we may suppose to have been
deserve further study. at least part of the Peoples of the Sea,32 continued on
The allusions to Phoenicians in the Odyssey have to the Phoenician homeland. Ugarit was destroyed and
been considered by Miss Lorimer as the "most im- did not revive; LH lIIC pottery is not found on the
portant" pieces of evidence for the dating of that poem, site, so far as is known, because none of the invaders
She has written that the middle third of the eighth cen- remained and settled there, as they did 011 Cyprus."
tury B.C. "is the period reflected in the picture of This makes it quite certain that the bronze implements
Phoenician activities given in the Odyssey/J 24 which is in the Geliclonya wreck and, therefore, those in Greek
in agreement with Nilsson's statement that "in our and Cypriot hoards, were made prior to that time, for
present state of knowledge the Homeric passages re- .Ugarit has offered some of the closest parallels for
ferring to the Phoenicians fit in best with the eighth bronzes in the hoards."
century, and this is in agreement with the fact that all The period before 1400 B.C. is less clear. As we have
these passages are found either in the Odyssey or in such seen in the discussion of ingots, there seems to have
parts of the Iliad as are recognized to be late." 25 Al- been in the fifteenth century some sort of joint Minoan-
bright has seen earlier, tenth-century Phoenician trade Phoenician control over copper production and trade.
and exploration as influencing the composition;" These This is too vague to be properly understood and, until
are but three of the many opinions which must now be the ingot-bearing ship of that date is located and ex-
reconsidered in the light of still earlier Phoenician cavated in the Bay of Antalya, I must leave further
voyages in the Mediterranean. conclusions to the linguists. I should point out, how-
Such voyages took Phoenicians to Cyprus and Sar- ever, that the Cypriot linear script (Cypro-Minoan},
dinia, where we have seen that Semitic commercial ac- anCat ling, C'Jlftr. Br aneeusorle, Chap. II, and "Minoan and
tivities began in the Bronze Age rather than in the Iron Mycenaean Pottery: Composition and Provenance," Archae-
Age. I do 110t suggest Phoenician colonies at this early oJnetry'6 (1963) 5; Desborcugh, Last Mycenaeans, 196.
80 Catling, Cvpr. Broneetoork, 40; Guido, Sa,rd1·nia, 188, but
date, but "temporary trading 'factories,''' to use Al- also 136.
bright's expression for posts of a later period." In 81 Desborough, Last Mvcenoeans, 279, after Karageorghis,
these factories copper was smelted and cast into four- ILlY, 22 Dec. 1962, 1012 ff.
handled ingots for export abroad. se Ibtd., 238: "It is reasonable to suppose that the disasters
in Cyprus were the work of the Land and Sea Raiders ....
In my discussion of the ingots, I pointed out that And if this was so, then it is a probable conclusion that at this
their manufacture by Syrians on Cyprus began at least point of their journey the Land and Sea Raiders were accom-
by 1400 B.C. and continued until 1200 n.c. It is not panied by a fairly powerful group of Mycenaeans .... "
33 Ibid. As may be seen, I have drawn heavily on Des-
coincidental that in those two centuries a tremendous
borough's picture of what happened in Cyprus and Syria at the
amount of Mycenaean pottery arrived on Cyprus in re- time of the transition from LH IIIB to LH IIIC pottery,
turn for some Cypriot export, which was presumably which he places around 1200 B.C.
copper." The great quantity of the imported pottery 34 In Chap. VI I pointed out a number of reasons for
adds further strength to a thirteenth-century dating of believing the Cypriot hoards and the Gelidonya cargo to be
the Cypriot and Greek hoards, as well as of the con- from the thirteenth century, although single parallel bronzes
had often been dated to either the thirteenth or twelfth century.
23Albright, op. cit. (sHpra, n. 13) 343-349. The historical evidence, which was not mentioned at that time,
24 Lorimer, Homer alld the Monuments, 505~506. emphasizes at least the contemporaneity of the Cypr-iot bronzes,
25 Nilsson, Homer and ilI:vc., 135-136. for hoards were too vital to the economy to have been pre-
"Albright, AlA 54 (1950) 173. served as such unless lost in a disaster. In the case of the
27 Albright, op. cit. (supra, n. 13) 348. Cypriot hoards, this disaster was the destruction at the end of
28 Catling, C':Ilpr. Bronceiaork, 36, 49. the LH IlIB (LC IIC) period.
 VOL. 57, 1'1'. 8, l,)67] BASS: CONCLUSlONS 167
which bears closer superficial resemblances to Linear A ingots, nuraghi, and "the introduction of certain reli-
than to Linear B, appeared on Cyprus in the fifteenth gious practices such as the worship of water at sacred
century and lasted until about 1200 B.C., ..."hen its last wells. "39 The introduction of ingots and nuragt-i
use was on copper ingots and clay tablets; in other have, of course, often been credited to Mycenaean
"words. it appeared with the beginning of Semitic copper influence.
production, continued in greater use during the height Miss Lorimer, writing of a later date, traced the
of that production, and disappeared with the Aegean route along which the Phoenicians traveled to Sardinia,
or Achaean colonization of Cyprus around 1200 B.C. and mentioned the possible connection between the
and afterward. This seems more than coincidental, and name Phoenix or Phoenikous and water stations along
the generally discredited view that Northwest Semitic the way.?'' One Phcenikous she neglected to mention
influenced Cypriot writing should be re-examined. was that within sight of Cape Gelidonya, the modern
Further, the conservatism of some authorities in ac- Finike with its abundant supply of fresh water in an
cepting Cyrus Gordon's discovery that Minoan (Linear otherwise waterless area. Is it not more than extraor-
A) was Phoenician (Northwest Semitic) has been dinary coincidence that the only Bronze Age ship-
based largely on their inability to see Semites in the wreck known was a Phoenician ship which sank while
Aegean during the Bronze Age. Our present study apparently heading toward that port?
• offers independent archaeological evidence which dis- In summation, the excavation of a Bronze Age ship-
proves the conservative views. It would further con- wreck at Cape Gelidonya has shed new light on early
firm As-our's picture of Aegean-Semitic relations, based metallurgy, metrology, trade, and ship construction, and
on an examination of second-millennium onomastica, has shown that proper excavation under water is as
which I am linguistically incompetent to judge." possible and necessary as that on land. Above all,
Astour's study shows a strong North Syrian foothold however, it has led to a restudy of known artifacts which
on Crete and Cyprus by 1500 B.C.; I should mention, reveal extensive Phoenician activity in the Mediter-
however, that his views on Cyprus are based on the ranean at a time prior to that usually supposed. Are
common assumption that Alasia and Cyprus are one, these the Phoenicians that Homer saw as traders, sea-
an assumption for which, as I have stated, the proof is farers, and craftsmen? Was Horner's essentially nega-
still lacking. tive attitude toward them a sign of the times, when
Turning to the \N est, we can now state with some Sea Peoples from the Aegean would sweep them fr0111
confidence that Phoenician copper refineries in the Cyprus and attack their homeland? Further study is
Bronze Age preceded the later and better-known necessary, but our findings support the recent statement
Phoenician colonies. It was not until after deciding by Stubbings that "there is no anachronism in Homer's
that oxhicle ingots probably were manufactured under Phoenicians; his picture of the heroic age would indeed
Phoenician supervision 011 Sardinia as well as on Cyprus be less true without them." 41
(see chap. V) that I read Albright's suggestion that
39 Ibid.
Nora once bore the "Phoenician name "Tar-shish,' mean-
40 Lorimer, Homer (llJd the Monsnnents, 75.
ing 'Smelting Plant, Refinery.''' 36 Deriving the word H Wace and Stubbings, Companion to Horner, 543. j ohn
"tat-shish" from an Akkadian word meaning "to melt, Chadwick (quoted by Grace in "The Canaanite Jar"), basing
be smelted," 3. he further suggested that Biblical Tar- his opinion on linguistic evidence, wrote : "The generally ac-
cepted view that Phoenician influence on Greece was greatest in
shish ships were not ships fr0111Tarsus or Tartessos, as
the ninth-eighth centuries B.c., when the alphabet was bor-
had been speculated earlier, but were the Phoenician rowed, may have to be revised in the light of this evidence
ships which carried smelted ores. Although Albright from the thirteenth century." How far such a statement and
writes of a later period, how well his ideas fit the Celi- this study vary from conservative views is only suggested by
donya ship and our conclusions on Sardinian mining. representative quotations: Nilsson, Homer alld Myc., 133:
"There was no place for commercial activities of the Phoeni-
Mr s. Guido correctly suspected the early Phoenician cians bef ore the end of the Mycenaean Age." Contenau, La
influence on Sardinia, attributing to a few leaders who Civ. pheniciennc, 277: "Iusqu'au XfI" siecle la mer appartient
had arrivecl "only a few centuries before the Phoenician aussi et surtout aux Egeens." Lorimer, Henner Gild the j110lLII.-
trading-posts were established, several features of Sar- snents, 505-506: " ... the earliest documented appearance of
the Phoenicians in the west is in Sardinia, in the first half of
dinian prehistory:" 38 oriental types of armor, copper the eighth century. There is no evidence that their ships had
as Michael C. Astour, "Second Millennium B.c. Cypriot and come even as far as the Aegaean at any earlier date . . ."
Cretan Onomastica Reconsidered," JAOS 84 (1964). I wish Albright, op, cit. (supra, n. 13) 342: "Vie can state with co~-
to thank David 1. Owen for bringing this to my attention. siderable confidence that Phoenician commercial expansion In

au Albright, op, cit. (supra. n. 13) 347. the Mediterranean did not seriously begin until after . about
37 Albright, BASOR 83 (Oct. 1941) 21-22. In a letter dated 1060 B.c." Deshayes, Utiles de Bronze, 387: "Avant tout,
26 April, 1963, Dr. Albright also kindly informed me that he nous devons insister sur l'importance du commerce mycenien et
"had recognized in the Egyptian name of a ship, kura, a sur son influence durant la second moitie du Ire millenaire."
Semitic loan word meaning 'refinery' (Hebrew ktr1·). The Kantor, Al A 51 (1947) 103: "After the close of the :MM II
word also appears, as I hope to show before long, as a word period, and throughout the later part of the second millennium,
for a special kind of ship, sr», in Ugaritic. The Egyptian only the sailors, merchants, and craftsmen of Mycenaean Greece
word is written with the determinative for 'ship.''' can justifiedly lay claim to the honor of forming the links
38 Sardinia, 187-188. connecting the Aegean with the Orient."
 APPENDIX

1. CARBON-14 DATES FOR WOOD tioning by razor for examination, but it was found in
E. K. RALPH
practice that all the specimens were to a varying extent
impregnated with copper salts and metallic copper,
A sample of twigs, weighing 42 grams, from the which made the process of sectioning very difficult;
brushwood which lined the hull, was dated by the Radio- specimen L, in fact, has proved impossible to section by
carbon Laboratory of the University of Pennsylvania. cutting and has had to be impregnated with synthetic
The results are as follows: resin and ground.
Determinations were as follows:
1020 ± 50 B.C. Calculated with 5568 half-life.
1110 ± 50 B.C. Calculated with 5730 half-life. Worked Timbers
1200 ± 50 B.C. Calculated with effective half-life, A Conifer Cwpressus sp.
based on measurements of samples of known age. B Conifer Cnpressus sp.
The hall-life tolerance (± 30 (or 5568, ± 40 for 5730, C Conifer Cnpressus sp,
and an uncertain amount for the "effective" half-life) E Hardwood Quercus sp.
have been omitted. The main reason for this is that the H Hardwood probably Quercus sp.
change from 5568 to 5730, for example, is 162 years. L Hardwood Q uercus sp.
Therefore, the ± 30 (or ± 40), which is a measure Brushwood
only of the consistency of three half-life determinations,
D Hardwood Quercus sp., possibly an evergreen
is not realistic.
oak.
Since the editors of Radiocarbon continue to publish
F Hardwood probably QueTClls sp., but immature
dates calculated with the 5568 half-life, 1020 B.C. is the
branch.
value to be used for comparison with other dates in this
journal. However, since 5730 is now the accepted value
e Hardwood Quercus sp., but immature.
I Hardwood Querclfs sp.
for the half-life of C-14, the 1100 B.C. date is closer to
] HarcIwood probably Querens sp.
the true age of the sample.
K Hardwood possibly Quercus sp., but atypical.
In addition to this change, however, the measurements
of tree-ring dated samples in this age range indicate The poor condition of the specimens renders it diffi-
that a date of 1200 B.C. is a better fit in relationship to cult to make more than a "probable" identification in
true ages (in this time span). This date of 1200 B.C. some cases. The wood has been attacked by boring
has been obtained by multiplying the B.P. date of 2986 organisms of at least two kinds or sizes, which destroyed
(5568 half-life) by 6 per cent. parts of the structure and greatly weakened the spec-
In other words. the chances are that the sample is imens, which were considerably deformed. The im-
not younger than 1060 B.C. (1110 minus 50) and may pregnation with metallic copper and copper salts has
be as old as 1250 B.C. (1200 plus 50). When more masked the structure of most specimens, making many
samples of 1,no\"111age from this period of time have anatomical features invisible, although specimens A and
been measured, it will be possible to relate the age of B were little more than stained green.
this sample to its true age more precisely. The brushwood specimens were all immature, and
there is therefore the possibility of confusion in deciding
2. IDENTIFICATION OF WOOD whether they are Quercus or Castanea. The latter
grows in southern Europe, North Africa, and Asia
A. C. WESTERN
Miner. It has apparently been recorded once from
Two specimens of charcoal were raised from an Lebanon and Anti-Lebanon/ but Holmboe > does not
unidentified portion of the wreck by the air lift. One report it from Cyprus at all. In immature wood the
was identified as a species of Quercus) while the other chief distinguishing feature between these two genera,
was in too poor condition for determination. the broad rays, is absent, but in general these samples
Twelve specimens of damp wood were submitted for correspond more nearly with immature Qu,ercus spec-
identification. Six of these were given the distinguish- imens, It is not normally possible to make specific
ing letters A, B, C E, Hand L, and these appeared to identifications from material of this sort, or indeed to
be timber of some size, thought to be fragments of the distinguish one species fr0111 another within many
structure of the ship. The other six specimens were genera, even in good modern material.
pieces of brushwood which had apparently been used as 1 Reported in George E. Post, Flora of S.vria, Palestine mid
a bed upon which the cargo of copper ingots was stowed, Shw/: (2 vols., 2nd cd., American Press, Beirut 1932-33) 2,
p.524.
These were given the distinguishing letters D, P, G, I, J 2 jens Holmboe, Studies on the Vegetation of Cvprus, Ber-
and K, Samples of all the specimens were impregnated gens Museums Skrifter. Ny. Raekke. (Bergen, 1914) 1:
with polyethylene glycol (Carbowax 4000) prior to sec- no. 2.

168
 VOL. 57, PT. 8, 1967] APPENDIX 169
From archaeological evidence it appears that the ship JIllethod: Samples were excited by DC arc method
was traveling westwards, perhaps to the Aegean and with graphite electrodes. The resulting spectral lines
probably from Cyprus, with a main cargo of copper were identified and compared visually.
ingots, and on this hypothesis one might suggest that Results:
the brushwood also was loaded in Cyprus for the cargo
to rest on. The six specimens of brushwood are all 1. Magnesium (Mg), Lead (Pb), Tin (Sn), Nickel
identified as Quercus species, or probably so, and if one (Ni), Cobalt (Co), Barium (Ba), Calcium (Ca),
can assume that modern vegetation gives us some guid- Silicon (Si), and Arsenic (As) were found in all
ance as to ancient vegetation, the possible species of samples.
Quercus at present found in Cyprus are as follows: Q. 2. Iron (Fe) was absent in samples: In 20, In 29, and
infeetoria, Oliv., Q. aln:ifolia Poech., and Q. cocciiera L., In 36.
with which is included Q. polaestina Ky. If it is con- 3. Zinc (Z11) was absent in samples: In 11, In 34, and
sidered that the brushwood was loaded at a Syrian or In 36.
Cilician port all these species of oak except Q. alnifolia 4. Bismuth (Hi) was absent in samples: In 8, In 34,
are also possibilities, together with a number of others:
, Q. petraea Liebl., Q. ehrenberg':': Ky., Q. cerris L., Q.
and BI frag.
• syriaca Ky., Q. libani Oliv., and Q. ilex L.
5. Titanium (Tj) was present in samples: In 12, In 13,
In 17, In 31, In 34, and In 39.
The question of where the ship was built is a more
6. Aluminum (AI) was present in samples: In 4, In 5,
difficult one than that of where the brushwood came
In 9, In 10, In 12, In 13, In 17, In 20, In 21, In 28,
from. The archaeological evidence seems to suggest In 31, In 34. and In 39.
that the original port was not in the Aegean, but further
7. Manganese (Mn) was absent in samples: In 1, In 6,
east. The specimens which Illay have been part of the
In 10, In 11, In 18, In 23, In 24, In 25, In 27, In 29,
structure of the ship comprise three Quercus species
In 33, In 35, In 36, and In 37.
and three Cupr essns species. The same species of
Cu pressus are recorded recently for Cyprus as for Syria The relative amount of each element present in a sample
and Lebanon, and extending northwards to the Amanus is indicated by the following order: 5, 4, 3, 2, 1, P, and
Range and beyond, namely C. semperuirens L., C. a (absent), as shown in the accompanying table:
liorizonialis Mill., and C. p)!'Ya/midalis Nyman., though
the spontaneity of the last species in Cyprus is doubtful.
Even if the specimens of Cnpressus could be determined TABLE 2

specifically they would give no clue to the port of origin, RELATIVE A"WUNTS OF ELEMENTS PRESENT

since the same species grow at present on the island and IN GELlDONYA INGOTS

on the mainland, and we have no reason to suppose that

Mg Pb So Ni Co Ea Ca Si Fe Zn Bi Ti Al Mn As
this was not so in the past. The pieces thought to be --- - - - - - - - - - - - - - - -
In I 2 3 I 2 2 1 2 II P 1 0 0 0 1
worked timber and identified as oak might possibly In 2 4 1 P I I 23 P I P 1 0 0 p P
10 3 5 4 1 2 2 I 4 2 2
yield more information, because one species, Q uerC1/S In 4 3 3 I 3 3 I 2 3 2
P
1
2
I
0
0
0
1
I
I
2
1
In 5 5 3 I 2 2 I 2 3 2 1 I 0 2 2 2
alnifolia, only occurs in Cyprus, and the Lebanon, In 6 I 3 I 2 2 I P P 2 P I 0 lJ 0 2
Amanus and the Taurus mountains produce at present 10 7 4 3 I 2 2 I 3 2 I P I 0 0 2 I
In 8 2 2 3 2 3 P P 3 3 I 0 II 0 I P
a number of species not found on the island. The diffi- In 0 4 2 2 3 2 2 2 4 2 1 1 0 P I 2
In 10 2 I I 2 2 2 3 2 I P 1 0 P 0 2
culty is that these specimens are all deformed and In 11 2 3 1 2 2 I 2 1 2 0 I 0 0 0 1
In 12 3 3 1 2 2 1 2 2 1 1 1 1 I 1 1
masked by copper deposits and it is unlikely that suffi- In 13 4 2 I 3 3 2 2 3 2 P I I I P 3
In 14 3 3 1 2 2 2 I P I P I 0 0 I 1
ciently good sections could be prepared from them to In 15 No sample was submitted.
In 16 1 P I 2 2 2 1 I I 2 I 0 0 P 1
reveal adequate detail for a specific determination. In In 17 .S 3 2 2 2 2 3 2 3 I 1 1 2 3 2
In 18 2 P P 2 2 2 I P 1 I' I II 0 P
these circumstances the question of port of origin of the In 19 2 3 2 2 2 I 3 P 3 2 1 0 0 I
1
2
ship cannot be answered conclusively from the evidence In 20 4 3 1 2 2 I 3 2 0 I I 0 I 2 1
In 21 4 1 I 3 3 2 2 3 2 P 1 0 P P 2
of the timber. In
In
22
23
4 I P 2 2 2 2 2 I I I 0 0 I' 2
4 P P 2 2 2 2 P I P I 0 0 0 1
In 24 4 3 1 2 2 I 2 2 I P 1 0 0 0 p
In 25 3 3 I 2 2 1 2 P 2 I I 0 0 0 1
3. REPORT ON COPPER INGOTS BY In
In
26
27
5
2
I
3
I' 2
2
2
2
I
I
P
2
3
P
I P 1 0 0 P
p
I
1 1 I' 1 0 0 0
SPECTROGRAPHIC ANALYSIS In
In
28
29
3 P
2
I 3
2
2
2
I 2 2 1 P I 0 P I 2
I I I 1 P 0 P I 0 0 0 p
In 3U 2 3 2 2 2 2 2 P 4 I I 0 0 P I
MARK C. HAN In 31 4 I I 2 3 2 2 3 I P 2 I P P 2
In 32 3 2 I 3 3 3 2 P 2 P I 0 0 P 2
In 33 1 2 I I I I 3 P 2 I P 0 0 0 p
Object: To determine the trace elements that are In 34 3 2 1 1 I 1 3 4 3 0 0 1 3 I 1
fn 35 2 3 1 2 2 1 P P I P 1 0 0 0 1
present in the ingots, in order to compare any difference In 36 2 3 I 1 1 1 2 P 0 0 I 0 0 0 p
In 37 I 3 1 2 2 1 2 P 2 I 1 II 0 0 p
among samples bearing different signs or shapes. In 38 4 I 1 3 3 2 2 2 2 P 1 c 0 P 2
In 39" 4 2 I 3 3 2 2 4 2
Specimen: Copper or copper alloy found 90 feet un- 51 14 3 P 5 2 2 2 3 2
I
1 P
I
I 0
I 2
lJ
2 3
2
P
derwater in the Mediterranean, just off Cape Gelidonya, HI
[rag. 2 I 5 2 1 P 1 2 1 P 0 0 0 P 2
Turkey.
170 CAPE CELIDONYA: A BRONZE ACE SH IPWRECK [TRANS. ""ER. PHIL. SOC.

4. CHEMICAL ANALYSIS AND ISOTOPE efficiently as to leave less than 0.01 per cent of silver
RATIO EXAMINATION OF LEAD (L 22) remaining, and Hellenistic metallurgists extracted down
ROBERT H. BRlLL to at least 0.02 per cent and possibly better. The high-
est silver content we have found among the few galena
A sample from this fragment was included in a study ores analyzed was 0.10 per cent silver in a sample taken
of lead isotope distribution ratios in archaeological ob- from a modern working at Kamareza nearby the ancient
jects made of lead and those of galena ores from mining workings at Laurion.
areas known to have been worked in antiquity.' This
study has shown that objects made of lead from three Sa. ANALYSIS OF BEAD
broad geographical areas are clearly distinguishable on ROBERT H. BRILL
the basis of isotope measurements. The first area is
Roman Br-itain, the second is Greece, the third includes The sample as received was part of a small bead
Spain, Wales, and Sardinia. which had been very badly deteriorated during its long
The isotope ratios of the lead in sample L 22 are: submersion. For the 1110Stpart it had crumbled into
flakes of a fragile buff-colored weathering product.
206/204 = 18.99 Microscopic examination showed that this was the re-
206/207 = 1.1996 mains of a glass bead, and was not, for example, Egyp-
206/208 = 0.4831 tian Blue or faience. Near what had originally been the
These values place the sample within a group of about core of the bead was found a very small quantity of the
ten samples of varying dates from Athens, Mycenae, "original" glass, associated with a porous green material
Corinth, Ceos, Cyprus, Sardis, and Antioch. These which appears to be a weathering product heavily
have been designated as Group L leads. stained with copper salts. By careful dissection under
The isotope composition is consistent with this lead the microscope, it was possible to pick out a few grains
having originated from the mines near Laurion, but it of the clear yellowish-green glass. Even these few
could also have come fro III geologically similar deposits grains had been spalled into rounded fragments which
elsewhere. For example, there are some mines in suggested that they, too, had been partially hydrolyzed
Turkey and Iran which yield this type of lead. and could not be considered representative of unaltered
A chemical analysis was also made of the lead sample glass. Nevertheless, it seemed worthwhile to attempt
by a semi-quantitative spectrographic technique. A a spectrographic analysis. The sample was so small
separate quantitative analysis was made of the silver that it was not possible to obtain reliable quantitative
content. The results are shown below. results, but the estimates made are given in the table
The first thing that is apparent about this composition below.
is that the lead is very pure, but this is not unusual for It was borne out that the sample taken was not repre-
ancient samples. The lead was probably melted from sentative of the original glass. Silica was the major
large well-formed crystals of galena are and thus high ingredient and copper had been used as a colorant. The
purity lead could reasonably be expected to result. The alkali, usually Na 2 0, bas apparently been leached out
other significant point is that the silver content of 0.04 by the long submersion. The considerable magnesium
per cent is somewhat higher than we have found in 1110st content is presumed to have been caused largely by ex-
of the hundred or so samples analyzed. This probably change from sea water, since the calcium is so low.
means that no attempt was made to desilvetize this This effect, the ion exchange of magnesium for calcium,
lead. In other words, the miners were probably seeking has been observed before in badly weathered glasses. It
lead per se and the lead was not just a by-product of would not be safe to speculate further on the composi-
an operation designed to obtain silver. The Roman tion of the glass.
smelters were often able to extract silver from lead so RESULTS OF ANALYSIS

Lead Major
Si02 Major
Tin O.OOX CliO 5-10 per cent
Copper O.OX J'vIgO 5-10 per cent
Bismuth O.OOX (low)
O.OOX
*A1203 o.x
Antimony Fe20~ o.x
Zinc Not found PbO o.ox
Magnesium Not found Sb205 O.OX
Not found MnO O.OX
Silicon
CoO 0.2
Aluminum Not found
Silver 0.04 per cent * The estimates of the second group of elements is unreliable.
---
1. The
sample is sample no. 39 in the study described in The low values could be the result of deterioration and/or un-
R. H. Brill and ]. M. Wampler, "Isotope Studies oi Ancient avoidable contamination with weathering products which would
Lead," AiA 71 (1967): pp. 63-77. have consisted chiefly of Si02•
 VOL. 57, PT. 8, 1967J APPENDIX 171
Sb. ANALYSIS OF BEAD 6. ANALYSIS OF TIN SAMPLE
EDWARD V. SAYRE F. R. DYKSTRA

Unfortunately for our purposes we came to consider As described by Dr. Bass (p. 82), three shapeless
that the bead was not composed of glass per sa. A micro- white masses were discovered beneath a number of ox-
scopic examination revealed its core to be a heteroge- hide ingots. At one location the material appeared to
neous mass of small particles, of about the size of a fine have flowed or have been washed from a shallow,
sand, with a bluish binding material between them. The rectangular void in the concretionary bottom cover. The
structure was that of faience, or perhaps a very poorly void, about 6 ern. square, gave the appearance of a mold,
melted glass. The binding material appeared to have possibly representing the original dimensions of an
been decomposed throughout through the hydrolytic ingot. It was also reported that the copper showed
action of water. It lacked cohesive strength, crumbling markedly intensified corrosion or deterioration when-
easily under light pressure and had the appearance of ever in direct contact with the white material.
hydrolyzed vitreous material. When raised to the surface the suspected tin had a
Although we were convinced that the hydrolytic de- consistency of tooth paste. Approximately 8 kgs. were
• composition would have altered the initial composition reclaimed and from this a sample of about 12-15 gms.
of the bead greatly, we undertook an analysis of the was brought back to Philadelphia sealed in an aluminum
overall core body, the data of which are presented in the film cartridge. By the spring of 1964 this sample had
accompanying table. Surprisingly, the analysis is con- "set-up" into a dull white, sugary-textured body-a
siderably closer to that one would expect from a dec0111- sandy agglomerate about the size of a finger tip. The
posed glass than of a faience. The only unusual con- mass could be broken by hand only with some difficulty.
centration for a decayed glass was the very high J\T 0 variance in color or banding was evident. The mass
concentration of magnesium. Magnesium was deter- appeared homogeneous, indicating that alteration or re-
mined in this instance by spectrographic analysis in two placement was essentially complete. The film cartridge
different spectra ranges, and the average of these two showed considerable corrosion inside, to the point of
analyses, which were both high but not in close agree- perforation. This attack appeared to be of galvanic
ment, is reported. \IVe concluded that the magnesium orrgm.
was not uniformly distributed in the bead, and very The sample was broken into tVI'O approximately equal
probably in part had been deposited by the sea water pieces. One was forwarded to Mr. 'fer Braake of
rather than initially having been at this high concentra- Galveston, Texas, and the other submitted for analysis
tion level. The glass of the period of the bead would to the laboratories of E. J. Lavina. Unfortunately the
have been characterized by a high magnesium oxide former was largely lost in the mails.
concentration, about five per cent lVIgO, but not as high A semi-quantitative spectrographic analysis of the
a concentration as we have found here. The high cop- remainder showed the following elements to be present:
per concentration would indicate the bead coating and
binding medium had been colored blue by means of Ca (Calcium) +++
Sn (Tin) ++
this element and the high antimony concentration would Ni (Nickel) 1'1'. (This was probably residual
indicate it had been deliberately rendered opaque. in the instrument)
Of course, in light of the decayed condition and in- Cr (Chrome) Tr. (This was probably residual
in the instrument)
homogeneity of the specimen my conclusions are quite Ti (Titanium) Tr. (This was probably residual
tentative.' in the instrument)
Al (Aluminum) Tr. (Result of container deterio-
PERCENTAGE OF VARIOUS OXIDES TN BEAD ration?)
Cu (Copper)
Lithium Li20 0.00045 Chromium Cr203 0.0027 Fe (Iron)
Sodium Na20 3.8 Manganese .MnO 0.13 K (Potassium)
Potassium K,O 0.115 Iron Fe203 4.85 CI (Chlorine)
Rubidium Rb,O <0.00040 Cobalt CoO 0.015 Ag (Silver)
Magnesium MgO 12 Nickel NiO 0.017 S (Sulfur)
Calcium CaO 6.6 Copper CuO 9.9
Strontium SrO <0.014 Zinc ZnO <0.018 These data were confirmed by incomplete wet analysis
Barium BaO <0.00085 Zirconium Zr02 0.0041 as follows:
Boron H203 0.066 Silver Ag20 <0.00026
Aluminum Ah03 2.19 Tin 5n02 0.0096 CaC03 (Calcium Carbonate) 71.00%
Phosphorous P205 0.092 Antimony Sb205 0.69 Sn02 (Stannous Oxide) 13.84
Titanium Ti02 0.12 Lead PbO 0.0044 Ab03+ Fe203 (Alumina + Ferric Oxide) 1.40
Vanadium V205 0.016 Bismuth Bi203 <0.0010 SiQJl (Silica) 1.66
c,. (Copper) .01

1 This material and the table were extracted from a letter to The inadequate size of the sample precluded more ex-
the author from Dr. Sayre, dated December 26, 1963. tensi ve analyses.
172 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK IT.RANS. A'\IER. PHIL. SOC.

These results seem to confirm that the material was Many students maintain that the bulk of the tin of
the remnant of what had been at one time metallic tin, this era was obtained from Cornwall. The analyses
largely replaced by calcium carbonate. The evidence of above seem to indicate some source other than the
the rectangular void points to a manufactured product. British Isles for this particular sample. Traces of
It may be suggested, notwithstanding the vague evidence germanium and cobalt are reported to be so C0111l11on
of the mold, that the tin might have originally been pres- in the Cornish ores of tin that their absence here may
ent as unrefined ore. This wou.d have been Cassiterite be regarded as evidence of some other source. (It
(S110,)-a compound so stable that it would probably seems unreasonable to think that, if originally present,
have resisted any alteration. all traces of germanium and cobalt would have been
The chemistry of this alteration is not completely completely extracted.)
clear. The relative positions of calcium and tin in the
electro-motive series make electrolysis difficult to pic- 7. SPECTROGRAPHIC ANALYSES
ture. It may be conjectured, however, that some condi- OF POTTERY
tion, possibly related to the presence of copper, allowed A. MILLET
for some difference in potential inducing the partial dis- The Iollowing results of spectrographic analysis were
solution of the tin and its replacement by what was conducted by the Archaeological Research Laboratories
probably calcium hydroxide-thus the original pasty at Oxford with the permission of Dr. E. Hall. The
nature of the sample. Oxidation of the metallic tin chart shows comparison of two of the Mycenean sherds
probably preceded the calcification. Subsequent ex- from the Gelidonya wreck with groups from Arpera
posure to air would account for the conversion of the Chiffik and Enkorni previously published by Dr. Catting
calcium hydroxide to the carbonate. and others (see chap. VII, 111.3).

Oxides of:

%of Magnesium Calcium Aluminium Iron Sodium Titanium Chromium Manganese Nickel
Total i\IgO CaO AbO. FeO NU20 TiO. CnO. MnO NO
Gelidollya (W9 l ) 6.9 24.4 10.3 9.7 1.68 0.77 0.055 0.077 .05S
Arpera Chifl.ik 1 100 5.8 ± 1.7 18.3±3.7 11.3±2.5 1O.7±2.1 1.48±0.31 0.<J2±0.21 0.070 ±0.021 0.095 ±0.025 O.OIU±O.002
GeJidonya (282) 5.1 10.1 14.5 10.2 1.58 1.06 0.060 0.049 0.021
100 5.4±2.2 14.3±3.1 13.5±3.1 9.9±2.0 1.42 ±0.37 O.96±O.28 O.12±0.O9 U.084±O.023 0.014 ±0.O09
Enkomi II

"Mycenaean Sherds

Geliclonya 9.9 22.2 10.6 12.1 1.36 1.26 0.076 O.()48 0.076
11.2 6.1 12.8 13.8 0.46 1.42 0.115 0.034 O.OM
Oeldonva (233) 0.037
Oetidonva (232) 6.3 15.6 16.2 16.6 0.60 1.27 0.124 0.058
Gelidonya (467) 4.3 16.6 8.9 11.6 1.47 1.23 0.094 0.038 0.040

Water Jar Sher d

Oeltdonva \V.9(12) 14.0 15.9 1l.2 26.5 1.89 1,89 > 1.0 .117 .{)41

Plain White Ware Jug

7.1 19.5 12.7 3.6 0.71 1.42 0,075 0.055 0.01 \

Celtdonva (144)

8. MODEL INGOTS IN EGYPTIAN ancient Egyptians customarily deposited in pits groups

FOUNDATION DEPOSITS' of materials and objects of a symbolic and magical
nature. These foundation deposits were often, but not
DAVID O'CONNOR
always, placed at points which were to be covered by
During the foundation ceremonies 2 which preceded corners ancl doorways and are usually associated with
the erection of an important religious building, the temples dedicated to the royal funerary cult or to the
1 There appears to be no detailed study of Egyptian founda- gods. In rare cases foundation deposits are directly
tion deposits. An excellent general account, with references, associated with royal tombs, either pyramids or rock-
is given in H. Bonnet, Rea1le.1:ikon, 263-264, and another dis- cut chamber tombs." In only one instance have I found
cussion is found in P. Montet, Les ewonnes de Tanis, 144-148.
Very incomplete lists of known foundation deposits are a Pyramid of Amenemhe I, Lisht; H. Winlock, EMMA, pt. 2
found in: G. A. Reisner, "The Barkal Temples in 1916," lEA (Nov. 1921) 16-17.
4 (1917) 221; K. Bittel and A. Hermann, "Grabungsbericht Pyramid of Senwosre II, Illahun; VV. M. F. Petrie, Iltaln.n,
Hermopolis 1933," MDIK 5 (1934) 14 n. 3; A. Rowe, "Dis- Kalnni and Gurob, 5.
covery of the famous temple and enclosure of Serapis at Tomb of Hashepsowe, Thebes; T. M. Davis, H. Carter, The
Alexandria," ASAE Cahier no. 2 (1946) 13-19. Tomb of Hatshopsitii,> 104-107.
2 For a description of these ceremonies see H. Bonnet, op. Tomb of Tuthmosis IV, Thebes; H. Carter and P. New-
cit. (su..ora, n. 1) 264-266. berry, The Tomb of Tho11tluosis l V, pp. xxix, 1-5.
 VOL. 57, PT. 8, 1967] APPENDIX 173
foundation deposits associated with a construction made
for the lise of a non-royal person." The custom of
foundation deposits may have existed in the Old King-
dcm," and is certainly attested from Dynasty XI G on-
wards and continued well into Ptolemaic times. Foun-
dation deposits can usually be closely dated, quite apart
from the often badly destroyed or denuded buildings
with which they are associated, by means of the names
of the royal donor which are frequently found on some
of the objects in the deposit.
The kinds of materials and objects deposited varied
at different periods, but frequently symbolic representa-
tions of some of the raw materials (stones, semi-
precious stones, faience, metal, wood, etc.) used in
building and decorating the temple or tomb were in-
I cluded in the deposit. Amongst these materials were
• copper and bronze 7 and it is with four specific repre-
sentations of these materials that this note is concerned.
In 1896 W. M. Flinders Petrie excavated at Thebes eMS

two funerary temples, one of King Akhenre'<setpenre' FIG. 161. Model ingot from Twosre's deposits
Merenptah-siptah and one of Queen Sitre'<meryamun at Thebes (after Petrie).
Twosre-seteptenmut." During the course of his work
Petrie found seven foundation deposits of Siptah and
eight of Twosre. In each of two of Siptah's deposits
and two of Twosre's deposits was one probable "sheet
copper" 9 model ingot, a total of four model ingots in
all. None of the ingots is described as being inscribed,
but the inscriptions on the associated objects leave no
doubt of the elates of the ingots." Of the four ingots
Petrie illustrates only one (cf. fig. 161), from Tv vosre's
deposits, but from his text one infers that the four
ingots were all of the same type; and, as shall be
shown below, one of Siptah's ingots was almost cer-
tainly of this type.
Petrie had been forwarding material to the Univer-
sity Museum of the University of Pennsylvania since
1890, when the Museum had begun to make financial
contributions to his work. Although he cloes not spe- FrG. ]62. Model ingot in University Museum.
cifically say so in his text, it is clear the Petrie gave
the University Museum material from the founclation
deposits of Siptah, since the Museum's collections con- by 1\1r. A. E. Parkinson, University Museum Chemist,
tain material identical with material illustrated by in the winter of 1965-1966. The cleaning revealed two
Petrie as coming from these deposits." incised cartouches on one side of the ingot and the
A model copper or bronze ingot {Museum Catalogue remaining traces of the hieroglyphs leave no doubt that
Number E 2029) was cleanecl at the writer's request they originally read Akhen (re'<setpenre") Merenptah-
siptah (see fig. 162), thereby securely dating the ingot.
.. The tomb of Senenmut, Thebes (Dynasty XVIII); H. The provenance of the ingot E 2029 is almost cer-
Winlock, EMMA, pt. 2 (Feb. 1928) 38. tainly the foundation deposits of Siptah's funerary
e G. A. Reisner, loco cit. (supra, n. 1). temple at Thebes. The University Museum catalogue
G Nebhepetre' Menthotpe II: H. Winlock, BMMA, pt. 2
(Dec. 1922) 28-29. carel for E 2029 describes it as a model ingot of un-
"Model tools and representations of raw metal are described known period coming from the Ramesseum at Thebes,
in the publications as being at "copper" and "bronze." Since collected by Sir Flinders Petrie in 1895-1896 and ap-
analysis seems hardly ever to have been carried out it is clear
parently forwarded by the Egyptian Research Account
that the objects so described may be of either metal.
8 Vol. M. F. Petrie, Six' Temples ru Thebes, 13-17. in 1896. In fact, the University 1\1useu111did receive
'Ibid" 15. in 1896 material from the foundation deposits (of
10 ibid., 14,17 and pls. XVI to XIX.
1.1 E.g., University Museum No. E 2104 = Petrie, op, cit.
Ramesses II) of the Ramesseum but Quibell, in de-
(sup'ra, n. 8) pls. XVII.23, XIX.10. scribing these deposits nowhere refers to the occurrence
 174 CAPE GELlDONYA: A BRONZE AGE SHIPWRECK [TRANS. AMER. PHIL. SOC.

of a model ingot." Since Petrie and Quibell were lowing appears to be a plausible reconstruction of the
excavating at Thebes in ] 896 in collaboration 13 and last reigns of Dynasty XIX.
material from the different temples presumably arrived Siptah was the immediate successor of Sethos II.
simultaneously at the University Museum it seems In the first year of his reign Siptah's prenomen and
that the attribution of a Ramesseurn provenance to nomen were Sekha' -enre'<setpenr e' Ra'messe-siptah,
E 2029 was a mistake on the Museum's part. The and these were changed from his second year onwards
most likely provenance for a model ingot of Siptah to Akhcnre-serpenre' Merenptah-siptah, Siptah was
collected in 1896 by Petrie is, of course, the funerary probably only a boy at his accession and during the six
temple of Siptah, where such ingots were found. years of his reign the real powers in Egypt seem to
The four model ingots of Siptah and Twosre appear have been Twosre, chief wife of Seth6s II, and the
to be the only such examples from Egyptian foundation Chancellor Bay. At some point, presumably after Sip-
deposits." To my knowledge symbolic representations tah's death, Twosre assumed the titles and powers of
of copper or bronze as a raw material in foundation a king and she probably died in her eighth regnal year.
deposits occur first in Dynasties XI and XII, when It is uncertain whether Twosre dated her regnal years
thin rectangular plaques of copper or bronze, as well as from the death of Seth6s II,21 thereby giving a maxi-
other materials, were inserted into sun-dried mud bricks mU!11 total of eight years for the combined reigns of
which were then placed in the foundation deposit pit." Siptah and herself, or from the death of Siptah," giving
The next examples I can find are four rectangular a maximum total of fourteen years for the two reigns.
"bronze" plaques from a foundation deposit of Hashep- It was at some time during these eight or fourteen
sowe at Der el-Bahri." 'while a "bronze model" 17 from years that the foundation deposits were made; unfor-
a foundation deposit of Amenophis If at El Kab looks tunately there is no clear evidence as to the dates of
in the illustration as if it might be a fragment of a the building of the two funerary temples, but it has
model ingot similar to those of Siptah and Twosre. been suggested that Twosre's temple must have been
The object is, however, more likely to be a crudely begun after Siptah's death."
shaped model tool; only examination of the actual Egyptologists have not yet reached agreement on the
object might provide a positive identification. Next absolute dates to be assigned to these eight or fourteen
chronologically come the four ingots of Siptah and years and it would be out of place in this brief note
Twosre, followed by eight small rectangular "bronze" to enter into the arguments concerning these dates.
plaques from a foundation deposit of Ramesses IV.18 Since the general period to which these model ingots
From Dynasty XXTI onwards it appears to have belong is more important than the absolute dates of
become a standard practice to include small plaques of any specific example, I will give here the dates assigned
bronze or copper in foundation deposits, often in Dy- to Siptah and Twosre by the 1110st recent authorities
nasty XXVI and Ptolemaic times accompanied by so as to indicate the likely chronological range of these
smaller but thicker "bricks" of copper or bronze." four ingots in absolute terms,
Studies published by several scholars over the last The foundation deposits 'which contained the four
twelve years ac have clarified the originally obscure model ingots discussed in this note must, therefore,
chronological relationship between Siptah and Twosre. be dated between 1209 at the earliest and 1185 at the
Although interpretations stilt differ III details, the fol- latest.

A. Gard i ner'" W. Hayes" W. Helck2S E. Hor-nung"

1961 1962 1962 1964
Siptah : 1208-1202 Siptab ) Siptah ) Siptah )
1209-1200 1192-1185 1200/1193-1192/85
Twosre : 1202-1194? Twosre Twosre Twosre

12 J. E. Quibell, The Remessesnn, 5, 6 and pl. XV. 18 R. Anthes in U. Holscher, The Bxccuation of A/ed·inet
13 tu«, 2. Halm II, 116-117, pl. 58.
a My search for parallels has certainly 110t been exhaustive, 19 For. examples, see: P. Montet, op, cit. (supra, n. 1) 133-
but I have examined the published records of approximately one 143; VV. M. F. Petrie, Nebeshelv and Dejenneb (Tanis II),
hundred and thirty-five foundation deposits of the period cover- 15, 40-41, 55, with pls. V, XXII-XXIII; Petrie, Naucrtuis I,
ing the Old Kingdom down to the reign of Ramesscs IV, and 28-30, pl. XXV; Petrie, Abydos I, 32, pl. LXX.6-9, 11; Dows
of a considerable number of foundation deposits of Dynasties Dunham, Royal Cemeteries of Kush Il, N1Wi, Chart III.
XXII to XXVI, the Ptolemaic period and the Napatan-Meroltic 20 For most recent studies, with references to preceding stud-
period. ies, see: J. Von Beckerath, "Queen Twosre as Guardian of
15 Dyn. XI: Nebhepetre" Menthotpe II: H. Winlock, BMMA, Siptah," lEA 48 (1962) 70-74; C. Aldred, "The Parentage of
pt. 2 (Dec. 1922) 28-29, fig. 18. King Siptah,' lEA 49 (1963) 41-48.
Dyn. XII: Amenemhe I: H. Winlock, BMMA, pt. 2 (Nov. 21 J, Von Beckerath, op, cu. (snpra, n, 20) 72.
1921) 16-17, figs. 9-11. 22 A. Gardiner, Egypt of the Pharaohs, 445.
Dyn, XII: Senwosre I: W. M. F. Petrie, Abydos II, 20, pL 23 J. Von Beckerath, op. cit. (supra, n. 20) 71.
LXII.86, 96, 120, 121. 24 A. Gardiner, lac. cit. (supra, 11. 22),
16 E, Naville, The Temple of Deir el Bahari VI, 9, pI. 25 W, Hayes, CAR fasc. 4, Chronology, 20,
CLXVllI. 26 Vv. Heick, Ag}lp/ologische AbhalLdlungel~ 5, 101.
"J. E. Quibell, El Kab, 16-17, pI. XXI.34. 27 E. Hornung, Agyptologische Abhandlnngen 11, 96-97.
 INDEX

Adzes, see tools Bones: astragal, 45, 82, 133, 163; fish, 117; pruning hooks, 95, 117; punch,
Aegean influence in Cyprus, 117-119, 166- 133-134 117; razor, 117; scale pan, 111;
167 Bracelets and rings, 34, 109-111, 117, 132, shovel, 94, 117; sickle, 95, 117;
Agricola, 70, 73 163, 164 spatula, 117; spearpoints, 18, 105,
Aigina : cylinder seals, 152; ingots, 62 Brea, L. B., 81 117; spit, 117; swage, 117; tripods,
Ain Shems : pottery, 122-123; scarabs, Bronzework : dating, 117-120, 164-166; 108; un worked castings, 113; vessels,
143, 145 place of manufacture, 121, 165; see also 107, 117
Air lift, see excavation methods tools, ingots, household objects, spear- cylinder seals, 68, 151
Alaca: stone maceheads, 126, 128 points, tripods, offering stands, brace- dating of bronze production, 117-121
Alasia, Asy, 77-78, 167 lets, casting waste lack of tin, 83
Albenga, 19 Buchholz, H.-G., 52-53, 57, 69, 76 lead, 131, 170
Albright, William F., 83, 88, 166-167 Bun ingots: 32, 33, 35, 52, 73, 78ff.; net weights, 131
Amen-em-opet, Tomb of, 65 analyses of, 78; distribution of, 80-81, oxhide ingots from, 18, 57, 61, 71, 76-
Amenhotep II, 65 165; lading, 44, 78; method of manu- 77, 165
Amorgos : cylinder seal, 152-153,158 facture, 78, 80f.; purpose, 81 pottery, 123-125, 164
Analyses: of beads, 170-171; of clay, 125, Burroughs, Sir Bernard, 18 representations of oxhide ingots, 68-
172; of copper and bronze, 41, 78, 82, Byblos : bronze chisel, 99; scarab, 145 69, 74
84, 169; of lead, 170; of metal foil, scarabs, 143, 145, 147
131; of tin, 171-172 Cape Gelidonya, geographical position of, Semites on, 77, 120-121, 166-167
Anchor, see stone objects 15-16 stone objects: maceheads, 126, 128, 164;
Antalya, Bay of: oxhide ingots, 61, 70, Carbon 14, 164, 168 mortars, 128, 164
76-77, 166 Casson, S., 69 weights, 139-140, 142, 163
Authedon : bronze hoes, 88-89, 117 Casting, 64-65, 69-70, 81, 113-116 wood, 168-169
Antikythera, 18, 22 Casting waste, 34, 114-116, 131 see also Apliki, Ay Ir-ini, Enkomi,
Anvil, see tools and stone objects Catling, H., 57, 61, 76, 77, 84, 88, 89, 93, Evreti tombs, Kouklia, Kourion,
Apliki : basketry, 160; pottery, 124; ham- 102, 103, 105, 107, 108-109, 111, 117- Lapithos, Laxia tou Riou, Mathiati,
mer stones, 130 118,120,164,172 Myrtou-Pigadhes, Soli
At-as, Kemal, 14-15, 17,21,27 Chisels, see tools
Areas of site, 27, 32ff., 44 Clark, J. G. D., 71 Daggers, 18, 103
Argive Heraion : cylinder seal, 156 Clay, 125, 134, 163-164, 172; see analysis Dating of ship, 164-165
Arkolochori : bun ingots, 81 of Decompression, see di ving
Artemision, 19,22 Cochlan, H. H., 163 11. 7-8 Delos: cylinder seal, 153
Asine : pottery, 122 Cochran, Drayton, 15, 17-18 Dendra: bronze mirrors, 113; foil cov-
Astragal, see bones Cochran, John, 17 ered vases, 131
Astrakous : cylinder seal, 154 Ccmparette, T. L., 71-72 Desborough, V. R. d'A., 3, 89, 118
Athens; oxhidc ingot, 62 Concretion: 30, 32-34, 83; composition Deshayes, J., 84, 88-89, 93, 95, 99, 167 n.
Athens, Acropolis Hoard: bronze ob- of, 40; removal of, 28-29, 33, 39, 41 41
jects: double axes, 95; hoes, 88-89, Copper: sources of, 76; management of, Dikaios, P., 3, 61, 120
117; mirrors, 113 76, 120, 166-167; see also analysis of, Divanli, Rasim, 14, 17-18
Aeropagus: foil covered vessels, 131~ casting, smelting, ingots Divination, 133, 163
132 Cousteau, jv., 19-20 Diving: bends, 18, 22, 23 i decompression,
Pnyx : bronze tripod, 108 Crystal, see stone objects 22, 32; embolism, 23; methods and
Awls, sec tools Cult-places, Near-Eastern inAuences on, equipment, 14, 24-25; nitrogen narcosis,
Axe-adzes, see tools 166 23, 26; photography, 17, 18, 22-26, 32;
Axes, see tools Currency, 69, 71-72, 82 sharks, 23; visibility, 23-24
Ay Irini : stone maceheads, 128 Current at Cape Celidonya, 16-18, 24-25, Dumas, Frederic, 19-20, 23, 28, 40
30-31, 40, 45, 164 Dunnage, 49
Ball, Terry, 21 Cylinder seals: at Cape Gelidonya, 45, 82, Dutbuit, Claude, 33
Ballast, see stone objects 117, 163-165; on Crete, 153-156; on
Barnes, Robert, 70 Greek mainland, 156-158; on Greek Egypt: bronze hoe, 90; bronze chisel,
Basketry, 33-34, 45, 117, 160-164 islands (excepting Crete), 152-153 100; mirrors, 113-114; weights, 139;
Beads, 34, 42, 132-133, 163-164, 170-171 Cyme: oxhide ingot, 61; near source of for representations of oxhide ingots,
Bean, George, 14 copper, 76 see el Amama, Karnak, Medinet Habu,
Beaufort, Francis, 16 Cypro-Minoan script, 72-73, 76, 91, 166 Thebes
Bends, see diving Cyprus: 15,18,73, 163-167 el A marna: representations of cxhide
Benoit, F., 19 as Alasia, 77-78, 167 ingots: Tomb of Meryra I, 66; Tomb
Benson, ]. L., 84,108 as source of copper, 76-77 of Meryra II, 66, 71; Tomb of Huya,
Beth Pelet : scarabs, 143, 145 beads, 133 66, 69, 71; representations of bun in-
Beth Shan: bronze tripod, 108; offering bronze objects: anvil, 117; awls, 102, gots: Tomb of Meryra I, 81
stand, 108 117; axe-adzes, 99, 117; bracelets, Enkomi : bronze objects: horned statu-
Beth Shemesh : bronze hoe, 88 109, 117; casting waste, 116; chisels, ette, 116; hoe, 88; chisel, 100; dating
Billhooks, see tools, pruning hooks 100, 117; daggers, 18; double axes, of copper industry, 120; oxhide ingots,
Bodrum, 14, 15,18,21,41,84,142 95, 117; hammer, 102; hoes, 88-89, 57, 61; possibly 011 seals, 68; as statuette
Bogazkoy : supposed oxhide ingot, 57; 91, 93, 117; knives, 117; mirrors, base, 69, 120, 164n. 15; pottery, 123-
tablet KBo XII 38, 78; weights, 139- 113; molds, 95, 117; needle, 117; 125; stone swage block, 102; slag, 76;
140 offering stand, 108; picks, 84, 86, 88, weights, 139; see also Founders Hoard,
175
176 CAPE GELIDONYA: A BRONZE AGE SHIPWRECK lTR,'\NS. AMER. PHIL. soc.

Gunnis Hoard, Ingot Hoard, Stylianou Ialysos : cylinder seals, 153; pottery,124; Megiddo : bronze objects: bracelets, 109;
Hoard, Tresor de Bronzes, Weapons foil covered pottery, 131 chisel, 99~100; hoe, 89; offering
Hoard Iuunerwahr, Sara, 3, 165 stand, 107-108; bronze stands, 109
Erbin, Ahmet, 14 Ingegnoli, Franco, 14 pottery, 122-125
Evans, Arthur ]., 67 Ingot Hoard (Enkomi ) : dating, 120 scarabs, 145
Evreti tombs (Cyprus) : stone ruaceheads, Ingots, see bun ingots, oxhide ingots, slab stone objects: maceheads, 126-128
126 ingots, tin, silver weights, 139
Excavation methods: air lift, 30-31, 34- Ivory, 76, 132, 165 il1eftclI1,17
35; balloon, 29-30, 39; drawing under Mersin : bronze chisel, 100
water, 27; metal detector, 32, 34, 39; Ierablus : stone mortar, 128 Meryra I, Tomb of, 65
photography, 25-27; sand removal, 30- Meryra IT, Tomb of, 65, 71
31; surveying, 25-27, 32 Kakovatos : cylinder seal, 156 Metal detector, see excavation methods
Kameiros : cylinder seal, 153 Minet el Beida: pottery, 124
Farnsworth, Marie, 131-132 Kamilari : cylinder seal, 154 Minoans, 74-75, 77, 166
Finike, 15,21, 164, 167 Kantor, Helene J., 167 n. 41 Mirrors, see household objects
Fishing, 128, 131, 163 Kapkin, Mustafa, 14, 17-18 Mochlos : cylinder seals, 155; oxhide
Foil, metal, 39, 41, 131-132 Karnak, representations of oxhide ingots: ingot, 61
Food, 45, 82,134,163 Relief of Tuthmosis III, 63; Relief of Mohenjo-Daro : bun ingots, 80
Forbes, R]., 81n. 153, 83 A menhotep J r, 65 Molds, 70, 95, 113, 117, 134, 163
Founders Hoard (Enkomi ) : bronze ob- Karphi: pottery, 124-125 Mortars, see stone objects
jects: axe-adzes, 99; hammer, 102; Katsamba : cylinder seal, 154 Myceuae : bronze double axe, 95; bronze
hoes, 88-89, 91; tripods, 108; dating, Keltiu, 62-64, 74--77 knives, 102; cylinder seals, 156-157;
88, 108, 119; ingots, 108, 119 Kenna, V. E. G., 150 n. 21, 158 oxhicle ingots, 57, 61, 71
Founders' hoards: 163, 165-166; dating, Knives, see tools Mycenaean pottery as evidence for trade,
88-89, 166 Knossos: 76; copper management, 76-77; 75, 165-166
Frankfort, Henri, 158-159 cylinder seals, 154-155; foil covered Myrtou-Pigadhes : bronze tripods, 108;
Frost, Honor, 21, 32, 83 vases, 131; linear B tablets, 68, 71, 142; pottery, 122, 124-125
Furumark, A., 89, 124-125 oxhide ingot, 61; weights, 139
Kouklia : pottery, 124; stone rnacehcad, K axes : bronze double axes, 95
Gagnan, Emile, 19 126-128; stone mortar, 128 Nebamuu. Tomb of, 50, 65
Gaza: bronze socketed tool, 93; scarabs, Kourion : pottery, 124; bronze tripod, Nebamun and Ipuky, Tomb of, 65, 83
143, 145; stone disk, 128 107; ox-hide ingot on bronze stand, Needles, see tools
Cerar: iron mattock, 94; pottery, 122 68, 74 Nilsson, Martin P., 16711.41
Gezer: scarabs, 146; weights, 139-140 Nitrogen narcosis, see diving
Cjerstad, E., 89 Lachish : astragals, 133; beads, 133; pot- N uzi : stone mace heads, 126, 128
Glass, see beads tery, 122, 125; scarabs, 143, 145
Gold,81 Lamboglia, Nino, 19 Odyssey, 49, 166
Gordon, Cyrus, 3, 77, 167 Lamps, 39, 45, 82,117,125,163-165 Offering stands, 107-109, 117
Goymen, Nazif 21 Lapithos : stone macehcads, 128 Olympia: cylinder seal, 157
Grace, Virginia, 15 Larkum, A. W. D., 40 Oxhide ingots: 17-18, 27, 33, 35, 39-42,
Grand Congtoue wreck, 19-20,22 Laxia tau Riou: stone mace heads, 128 52ff.; analysis of copper, 62, 169;
Gruben, G., 49 Lead, 41, 63-iJ7, 70, 73, 131, 163, 170 bronze, 62, 70; dating, 69, 119-120,
Guido, M., 167 Lindos : cylinder seal, 153 164: distribution in Mediterranean, 57,
Gultekin, Hakki, 14, 17-18 Linear .A script, 73, 76, 77, 167 74-75; electrum, 64, 70; in Asia Minor,
Cunnis Hoard (Enkomi): bronze axe- Linear B: script, 73, 167; tablets, 68, 71, 61, 77; in Crete, 57, 61, 71, 77; in
adzes, 99; bronze hoe mold, 113 74, 135, 165 Cyprus, 57, 61; in Greece, 57, 61-62,
Ladas, 21 71, 77; west of Greece, 61-62; in Syria-
Hagia Pelagia: cylinder seal, 154 Lorimer, H. L., ]66-167 Palestine, 57, 76; in Egypt, 62, 77;
Hagia Tr-iada : oxhide ingots, 61; cylinder lead, 63-64, 67, 70; manufacturers, 74,
seal, 154; bronze chisel, 100 Maceheads, see stone objects 76-77, 166; methods of casting, 70;
Hama : stone mortars, 128 Mahdia wreck, 19-20 miniature, 57, 61-62, 76, 172ff.; pur-
Hammers, see tools and stone objects Makarska (Dalmatian Coast): oxhide pose, 69-72; representations on linear
Harden, Donald, 68 ingot, 61 B tablets, 68, 71, 74, 135; representa-
Harrington, G. L., 49 Malatya : maceheads, 128 tions on seals, 67-68, 74; representa-
Hazor : pottery, 122, 125 Mallia: cylinder seal, 155 tions, see also Thebes, Karnak, el
Hazaidakis, J" 71 Malthi : weights, 139 Amarua, Medinet Habu; signs on, 52,
Hepu, Tomb of, 65 Marden, Luis, 32 72-74, 76-77, 167; silver, 63-64, 67, 70:
Herakleion: cylinder seal, 154 Masson, 0., 68 stacking on ship, 44, 52, 73, 163; tin,
Herodotus, 75 Mathiati : bronze objects: castings, 113; 63-64, 70; types, 52-53; weights, 52,
Hoards, see founders' hoards sickle, 95; unidentified, 111; hoe, 91; 57, 68, 71, 73
Hodges, H. W., 43 oxhide ingots, 61; mold, 113
Hoes, see tools Matting, 34, 44, 52, 160, 164 Palaikastro : cylinder seals, 155; oxhide
Household objects: mirrors, 111-113, 118; Mattock, see tools ingot, 61
razor, 105, 117; spatula, lOS, 117; spit, Mavrospileio : cylinder seal, 155 Parkinson, A. E., 131, 135 n. 8, 173
109, 117; vessels, 105-107, 117 Maxwell-Hyslop, R., 84, 97,117 Paras: cylinder seal, 153
Hull: 44; construction, 45, 48-50; esti- Medinet Habu, representations of oxide Pendlebmy, J. D. S., 71
mated size, 45, 163 ingots: Relief of Ramesses III, 67, 69, Penl:,et, Tomb of, 65, 69
Huston, John, 18 70, 164 Peoples of the Sea, see Sea Peoples
Huy, Tomb of, 67, 69 Mega Monasterion: cylinder seal, 156 Perati: cylinder seal, 157
Huya, Tomb of, 65, 69, 71 }.,t(egaw,Andrew, 18 Petrie, VV. 1L F., 90, 173
 VOL. 57, PT. 8, 1967J INDEX 177
Phipps, Susan, 17 Silver, 63-64, 67, 70, 73,81,131 Tiryns: bronze cauldron handle, 107;
Phoenicians, 75, 77, 120, 164-167 Sitias : oxhide ingot, 61 cylinder seal, 158
Photography, see diving and excavation Sjoovist, E., 18, 120 Titan, 20
methods Slab ingots, 35, 39, 44, 52, 81-82, 131 Tools (bronze): 33, 41; adzes, 84, 95,
Picks, see tools Smelting, 80-81, 166 97-99, 117; anvils, 102, 117, 163; awls,
Plalanos: cylinder seal, 155 Soli, Bay of, 61 102, 117; axe-adzes, 84, 99, 117; axes,
Platon, N., 3, 61 n. 23 Spatula, see household objects double, 34, 95, 117; chisels, 117, cold,
Pliny, 16, 164 Spearpoints, 15, 18, 103-105,117 99-100, deep bar, 100, socketed, 100;
Plowshares, see tools Spit, see household objects hammers, 102, 117; hoes, 88-93, 113,
Poi de bard, A., 19 Sponge diving, 14, 18 117; knives, 84, 102, 117; mattock, 94,
Polishers, see stone objects Stewart, r. R., 83 117; needles, 102, 117; picks, 84-88,
Potter y : 34-35, 41-42, 45-46, 122fl., 164; Stone objects: anchor, 26, 45, 142; anvils, 117; plowshares, 88, see picks and
see also lamp 35; ballast, 35, 39, 48; crystal, 130; hoes; pruning hooks, 95, 113, 117;
Prosymna: cylinder seal, 157 hammer, 163; maceheads, 35, 39, 45, punches, 102, 117; shovels, 94, 117;
Pruning hooks, see tools 82, 126-128, 163-164; mortars, 17, 128, sickle, 95, 117; swage, 102, 117; un-
Punches, see tools 164; polishers, 128-130, 163; swage, finished, 95, 113-114
Puyemr-e, Tomb of, 63, 114, 116 102; whetstones, 39, 45, 82, 117, 163; Tresor de Bronzes (Enkomi): bronze
Pyla (Kokkinokremmos ) : bronze axe- see also weights objects: axe-adzes, 99; bracelets and
~ adze, 99, 121 Strabo, 16, 76 rings, 109; dating, 119, 121
Stylianou Hoard (Enkomi): bronze hoc, Tripods, 107-109, 117-118
Ramesses 111, 67, 69-70 88, 91; bronze sickle, 95; dating, 108; Troy: bronze bracelet, 109; cylinder seal,
Ras Sharma (Ugar-it) : 76, 166 cauldron handle, 107 158; weights, 139
astragals, 133 Surveying, see excavation methods Tuthmosis III, 63
bronze objects: bracelets, 109; chisel, Swage, see tools and stone objects Tylecote, R. F., 80, 11411.157
99; hoes, 93; offering stand, 108; Syrians, representations of, 49-50, 62-67, Tvlissos : cylinder seal, 156; oxhide
tripod, 108-109 74-76 ingots, 61
oxhide ingot, 57
pottery, 122-125 Useramon, Tomb of, 63
Tailliez, P., 19-20, 22
stone mace heads, 128
Tarsus: pottery, 122-124; stone objects:
weights, 139 Vaphio: bronze axe-adze, 99
crystal, 130; mortar, 128; sinker, 128;
Razor, see household objects Vari : cylinder seal, 158
whetstone, 128, 130
Rekh-mi-re", Tomb of, 81, 160; see also Taylor, Sir George, 161 Vercoutter, J., 63--65, 77 11.135
Thebes Vessels, bronze, see household objects
Tell Abu Hawarn : bronze cauldron
Retnu, 63-65, 67 handles, 107; pottery, 122-123
Rings, see bracelets Wace, A. J. B., 71
Tell Ajj ul : scarabs, 143; stone mace-
Routes, sea, 49, 76, 164, 167 Wainwright, G. A. 74 108, 77, 81
heads, 128 11. n.
Rutsi : cylinder seal, 157 Tell Beit Mir sim : bronze hoe, 88; ox- 155, 83
Ryan, Eric, 61 Waterman, Stanton, 15, 17
hide ingot, 57; pottery, 123; scarabs,
145; stone macehead, 126 Weapons, see spearpoints
Samos : cylinder seal, 153
Tell el Fara : bronze pick, 88; pottery, 122 Weapons Hoard (Enkomi): bronze dou-
Sanders, N. K., 84, 102 ble axe, 95 ; dating, 119
Tell-el-Hesy : bronze hoe, 91-93
Sardinia: 76-77, 83, 167; oxhide ingot,
Weights : 34, 37, 45, 47, 82; accuracy of,
61--62,166 Tell en-Nasbeh : pottery, 122-123; sca-
rabs, 145; weights, 139-140 142; determination of standards, 137-
Scarabs, 35, 45, 82, 117, 143-146, 163-165
Tell Er Ratbeh : scarabs, 145 138; material, 135; shapes, 135-136;
Schaeffer, C. F. A., 3, 57, 61, 69, 84, 99, value ill determining mute of ship, 142,
109, 120 Tello: mace handle, 128
Thebes (Egyptian) : model oxhide ingots:
163-164
Scrap metal, 33, 81, 87, 114-117, 163
173-174; representations of oxhide in- Whetstones, see stone objects
Scylax, 16
gots: Tomb 119, 62-63; Tomb of Wood, 29-30, 48-51, 35, 39, 42, 45, 164,
Sea Peoples, 120, 164, 166-167
Seals, oxhide ingots on, 67-68, 74; see Puyemre, 63; Tomb of Useramon, 63; 168-169
also cylinder Tomb of Rekh-mi-re-, 62-65, 70, 74, Wrecks, 14, 15; sec also Albenga, Anti-
Seltman, Charles, 71 76-77, 81; Tomb of Penhet, 65, 69; kythera, Artcmision, Grand Congloue,
Semitic, 77, 165, 167 Tomb of Nebamun, 50, 65; Tomb of Mahdia, Titan
Ships: at Cape Gelidonya, see hull; from Amen-cm-opet, 65; Tomb of Hepu, 65; Wreszinski, VV., 63
Alasia, 78; Homeric, 45, 49; Kef tin, Tomb of Nebamun and :r pukv, 65, 81;
74; Syrian, 49-50; Tarshish, 167 Tomb of Huy, 67, 69 Yassi Ada, 14, 19,22,32
Shovels, see tools Thebes (Greek): cylinder seals, 157-138 Young, Rodney, 18
Sicily: oxhide ingot, 61, 81 Throckmorton, P., 61
Sickles, see tools Tin: 34-35, 41, 52, 63-65, 73, 81-83, 131- Zafer Papoura : cylinder seal, 156
Signs: on bronzes, 91, 97-98; on ingots, 132, 17I-I72; ingots, 66, 70, 82-83, 163; Zakro : oxhide ingots, 57, 61, 76; weights,
52, 72-74, 76, 166 sources of, 83, 172 139