Professional Documents
Culture Documents
PROCEEDINGS OF THE
INTERNATIONALCONFERENCE
ON
GREEKARCHITECTURALTERRACOTTASOF THE
CLASSICALAND HELLENISTICPERIODS
DECEMBER 12-15, 1991
EDITEDBY
NANCY A. WINTER
PROCEEDINGS OF THE
INTERNATIONALCONFERENCE
ON
GREEKARCHITECTURALTERRACOTTASOF THE
CLASSICALAND HELLENISTICPERIODS
DECEMBER 12-15, 1991
EDITEDBY
NANCY A. WINTER
CIP
General
Terrescuites architecturales,peintureset mosaiquesaux Ve
MARm-FRAN9oIsEBILLOT:
et IVWsiecles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
MainlandGreece:ThePeloponnese
MARY C. ROEBUCK:ArchitecturalTerracottasfrom Classicaland Hellenistic Corinth 39
CHARLEsK. Wn.ums, II: Roof Tiles from Two CircularBuildingsat Corinth . . . 53
FREDERICK P.HEMANs: GreekArchitecturalTerracottasfromthe Sanctuaryof Poseidon
at Isthmia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
STEPmNG. MILLER: Sosildes and the Fourth-CenturyBuilding Program in the
Sanctuaryof Zeus at Nemea . . . . . . . . . . . . . . . . . . . . . . . . 85
KLAusHoF -uLwR: Die Dachterrakottendes Artemistempelsvom Apollon-Heiligtum
in Agina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
CALUOPI KRYSTAIVOTSI: ApXvTexTovLxi Teppaxox6e aco6 -v ApXada ELXuJav 113
DoRis GNEIsz:Die Dachterrakottenvon Aigeira . . . . . . . . . . . . . . . . . 125
JOACHIMHiDmEN:KlassischeDacher aus Olympia . . . . . . . . . . . . . . . . . 135
PETRosTmmius: HellenisticArchitecturalTerracottasfrom Messene . . . . . . . 141
Greece
Northern andAlbania
GERnu HibBNER:Die klassischenTempeldachervon Kalapodi (Phokis) . . . . . . 171
AmALA VLCHOPOULOU-OuONOMou: Ta Epayqtotaa Kep.t(xcv ct6 -o Ikp6 -rrS
A caadVjq . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 181
DmIwosTEN BUDINA: ArchitecturalTerracottasfrom the Towns of Kaonia: Antigonea
and Buthrot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
TheBlackSea,AegeanIslands,andAsiaMinor
KoNRAD ZIMMERMANN: Traufziegelmit Reliefinaanderaus dem Schwarzmeergebiet 221
VOLKERKAsTNER: Kleinasien und Griechenland. Dachterrakottennacharchaischer
Zeit aus Pergamon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
-STELLAG. MnzR: ArchitecturalTerracottasfrom lion . . . . . . . . . . . . . . 269
iV TABLE
OFCONTENTS
SouthItalyandSicily
JoHNF. KENFmiEL: High Classicaland High Baroque in the ArchitecturalTerracottas
of Morgantina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
GIOVANNA GiuEcOE MARIA JOSE'SRAZZULA:Le terrecottearchitettonichedi
Elea-Veliadall'eti arcaicaall'eta ellenistica . . . . . . . . . . . . . . . . . .283
PETERDANmmw: Ein architektonischesTerrakottafragmenthellenistischer Zeit aus
Reggio Calabria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
BER1 BRANDES-DRUBA: Einige tarentinerArchitekturterrakotten . . . . . . . . . 309
LUISAVIoLA:A Late ClassicalSima from Heraclea in Lucania . . . . . . . . . . 327
JOURNALABBREVIATIONS
AA = ArchdiologischfrAnzeiger
AAA = ApXatoToyLXAviXxtca eF AjO~vv
AbhLei = Abhandlwgen derSichsischen Akademie derWusenschafin Kiasse
zu Lez4g, Philologisch-historische
AdI= Annal dell'Isitutodi corrispondeza archeologica
AION= Annal delseminario di studidelmondoclassic. Archeologia e storiaantica
AJA = American journalofArchaeology
AM = MitteiugendesDeutschen Archaologischen Instidtsh,AthnischeAbtilng
AnnPisa= Annal dellascuolanormalsuperiordi Pisa
AntK= AntikeKunst
= Archaeology.
Archaeology An OfficialPublication of theArchaeological ofAmerica
Institute
ArchCl= Archeologia classica
ApXEp = 'ApXatoXoyLxh 'Eyqniep(
= Architectura.
Architectura ZeitschifftfrGeschichte derBaukanst
ASAtene = Annuario dellaScuolaarcheologica diAtenee deleMissioniitalianein Oriente
Atheaeum= Athnae . Studiperiodici di letteratura Universit4
e storiadell'antichith, di Pavia
AttiMGrecia = Ati e memoriedellaSocietaMagnaGrecia
AuA= AntikeundAbendland
BCH = Bulletindecorrespondence helinique
BdA= Bolettinod'arte
BICS= Bulletinof theInstituteof ClassicalStudiesof theUnivershy ofLondon
=
Boreas Boreas.Muinstersche BeitrdgezurArchaologie
BSA= Annualof theBritishSchoolatAthuns
BSR = Papersof theBritishSchoolat Rome
BWPr= Wnckebmannsprogramm derArchdologischen Gesellschaftzu Berli
CR= ClassicalReview
CronCatania= Cronache e di stoia dell'arte,
di archeologia Universit4 di Catania
Dacia= Dacia.Revued'archlologie etd'histoireancienne
Ackc = ApXaLo)oyLx6vAckrdov
DiaUrch= Dialoghidi archeologia
'Epyov = T6 'Epyo V ri 'ApXato)oTyX4q'EEratpdaq
FA= Fastiarchaeological
FuB = Forschungen undBerichte. SaaticheMuseenzu Berlin
GazArch= Gazettearchiologique
GetyMusj= TheJ. PaulGetytMuseum Journal
Gnomon = Gnomon. KritischeZeitschriiftfrdiegesamte klassischeAltrm wsenschaft
Gymnasium = Gymnasium KulturderAntike
Zeitschnrftftr undhwnanistische Bilwg
Hesperia= Hesperia. Journalof theAmerican Schoolof ClassicalStudiesatAthens
IstMitt= Mitiugen desDeutschen Archdologischen Insttuts,Abting Istanbul
JbBerlfus = Jiahrbuch derBerlinerMuseen
jdI = Jahrbuch desDeutschen Archdologischen Instituts
JFA = JournalofFieldArchaeology
JHS = journalofHelknicStudies
jRS = journalofRomanStudies
Alearchos= I~earchos.Bolletinodell'AssociazioneamnidelMuseonazionale di ReggioCalabria
vi JOURNALABBREVIATIONS
profile above a fascia, over which is generally with a palmettoon both faces, attachedto the ridge
painted an anthemion. tile.
RIDGE TILE (Fr.tuile faitidre, Germ. Firstziegel,Gk.
palmetta drittaorritta = uprightpalmette. xepcxj4qxopu(ata). The uppermost pan and/or
palmettapendula = pendant palmette. cover tile that overlapsthe ridge of a sloped roof.
PAumErm=. A stylizedpalm leaf with fanned petals. rive. SeeEaves.
PALMETTE HART (Germ. Blattkern, Palmettenkern, rivestimento. SeeRevetment plaque.
Gk. avOe?co-6 nupivaq). The roundedbase from roll. SeeTorus.
which the palmette petals fan out. ROOFTILE (Fr. tuile, It. tegola, Germ. Dachziegel,
Palmettenkern.SeePalmetteheart. Ziegel, Gk. xepaVzq tvq miryvq). A piece of
palmettescirconscrites= circumscribedpalmettes. firedclay,generallylaid on the roof to overlap and
PANTam (Fr.tuile courante, It. tegola, Germ. Flach- be overlappedby another, to protect the building
ziegel, Gk. a-p x5ipaq). The tile designedto carry below.
rain water off a sloped roof. Rundstab. SeeTorus.
napae-c( atErV.SeeLateralsima.
perle e astragali. SeeBead and reel. sagoma. SeeMolding.
Perlstab.SeeBead and reel. scaglie. SeeScale pattern.
PLAINTE (Germ.gewbhnlicherZiegel, Gk. ayeXaa ScALEPATTERN(Fr.6cailles,It. scaglie, Germ. Schup-
xepaVis). The ordinarytiles of the roof slope. penmuster,Gk. 9p6Xx ypoLc8x&).A painted de-
niXxa. SeePlaque. sign resemblingoverlappingfish scales.
PLAqtuE (Fr.lastre,It. lastra,Germ. Platte,Gk. nitxa). Schachbrett= checkerboard.
A verticalelement that can be separateor attached Schragsima. SeeRaking sima.
to one end of a cover tile (formingan antefix)or Schuppenmuster.SeeScale pattern.
a pan tile (forminga sima). scrolls. SeeVolute.
Platte. SeePlaque. sigla. SeeTile stamp.
Plattenborte.SeeFascia. SIMA(Gk. aoErV).A pan tile with a vertical plaque
nXoXV6q. SeeGuilloche. attached along one edge to divert rain water off
Pressform.SeeMold. the roof.
Profilleiste.SeeMolding. sima d'egout. SeeLateralsima.
sima de rampant. SeeRaking sima.
pa836yX'upo.SeeTorus. sima de rive. SeeLateralsima.
RAKINGsIM (Fr.sima de rampant,sima rampante,It. sima di gronda. SeeLateralsima.
sima di rampante, sima rampante, Germ. Giebel- sima di rampante. SeeRaking sima.
sima, Schragsima, Gk. enae-cs aEVI, xepaVzq sima en cavet. SeeCavetto sima.
enaeitct). The sima that runs along the edge of sima laterale. SeeLateralsima.
the pediment (the rakes). sima "megarienne".SeeOvolo sima.
Ranken = scrolls,tendrils. sima rampante. SeeRaking sima.
Rankensima= scroll sima. aoErI.SeeSima.
Rautenziegel = diamond-shapedtile. SLI (Fr.engobe, Germ. tberzug, Gk. etEXptoa~). A
recouvrement. SeeFlange. thin layer of purifiedclay spread in liquifiedform
Reliefmaander= relief meander. acrossthe surfaceof the piece before firing.
revetement. SeeRevetment plaque. sockel. SeeFascia.
REvEmENT PLAQUE (Fr.revetement,It. rivestimento, SoFFrr(Germ.Unterseite). The undersideof a piece.
Germ. Verkleidungsplatte, Gk. etv~uan). A Sonnenblume = sunflower.
plaque that is generallynailedto a wooden element spacer. Se Tile spacer.
to protect it. Oapp6yLaO1 xCpaJACOv. SeeTile stamp.
RIDGE(Fr.fate, It. colmo, Germ. First,Gk. xopuyp5). spiral. SeeVolute.
The apex of a sloped roof, at the ridge beam. spirale. SeeVolute.
RIDGEPALMErTE (Germ. Firstpalmette,Gk. axpoxi- atzyiy -p(pLXqT.See Hip roof.
papqo xopuypaca).An upright plaque decorated Stempel. SeeTile stamp.
GLOSSARY xi
and Ioanna Driva, our Comptroller, has handled all the financial arrangements. Our
principalreceptionist,ChristinaTraitoraki,has done the translationsof the abstractsfound
in your conferencepackets. The School'sAdministrativeAssistant,Maria Pilali,and Cashier,
Niamh Michalopoulou, helped with the planning. And two of our students, Sheila Dillon
and Blanche Menadier,have also assisted. To all these and to the other staff of the School,
whose names it would be impossiblefor me to mention individually,we are all very grateful.
The staffof the Schooljoins me in wishing this conferenceevery success.
D. E. COULSON
WILLIAM
Director
AMERICAN STUDIES
SCHOOLOF CLASSICAL
54 Souidias Street
GR-106 76 Athens
Greece
December 15, 1991
TERRES CUITES ARCHITECTURALES,PEINTURES
ET MOSAIQUES AUX VEET IVESIECLES
PLATS 14)
I. SITUATION DU PROBLEME
. 1. Quelques
exemples a l'archaisme
empruntes tard introduiront
plusefficacement (P1.1 et 2).
cepropos
L.1.1. Aux margesdu groupe des simasdites ?megariennes>>(parex. P1. 1:b),fabriquees
nombreusesa Corintheet imitees ailleurs,caracteriseespar leur profilet par leur anthemion
Plusieursconversationsavec A.-M. Guimier-Sorbetssont Al'origine de cette 6tude. J'ai tire grand profit
I
I. 1.2. Le decor des simasdu Temple des Alcmeonidesde Delphes, malaise a dechiffrer,a
W plusieursfois dessin6,notammentpour la publication.6Reste m~connu, un relev6 in~dit
effectu6par l'architecteH. Lacoste quelquesann~esplus tard,7parailtplus fiable (P1.1:d). II
donne l'impressionque le peintre a compose entre l'origineathenienne des commanditaires
et la provenance parienne du marbre : de meme que le profil, les palmettes triesagrees
relevent du repertoiredes ceramiqueset des marbres atheniens, notamment des simas du
Telesterion d'Eleusiset du Temple d'Athena des Pisistratides;8les lotus sont empruntes au
repertoirecycladiquealorsbien represented au Tresorde Siphnos9et peut-etredeja au Tresor
Heiden 1987, pp. 61-102, pl. 7-11, planches de profils 3-6; Roebuck 1990, pp. 55-59, fig. 2, pl. 6, 7. Par
exemple ici Musee d'ArgosC. 27517, P1. 1:b.
3 Vocabulairearchitecturalet d~coratifemprunt6AGinouves et Martin 1985, pp. 178-179, pl. 61:7.
' Musee de Polygyrosinv. 78.703: Cambitoglou 1978a, p. 28, fig. 31; Cambitoglou 1978b, p. 86, pl. 73:b;
d'ouiTouchais 1979, pp. 587 et 589, fig. 149; ApXacxtc Maxebovta-Ancient Macedonia,no 167, p. 224 avec
fig. (fin du VI' siecle).
5 Euphronios, fig. pp. 62, 68-69, 78-79, 90, 97, 110, 116-117, 119, 121, 129, 138. Bien que ce schema
de liaison se perpetuejusqu'aux environs de 470 (par ex. col de l'amphore du P. d'Oreithyie, Munich 2345,
Arias et Hirmer 1960, fig. 158, 159), la sima etant traitee en figure noire, il n'y a pas de raison de la dater
beaucoup plus tard que 500 av.J.-C. On remarquerasur les vases cites et sur bien d'autresque les palmettes
s'opposent aux palmettes, les lotus aux lotus (<anthemiona double file et a correspondence>>, cf. Ginouves
et Martin 1985, pp. 178-179), comme tres frequemmentsur la c~ramiqueattique depuis 570 env. et plus tard
encore, ainsi que sur les simas attiques de la fin du VI' et du Ve siecle: Buschor 1929, simas noS X/XI et
XVI a XXI, fig. 20-23, 38-47 et 51, pl. 1 et 7-10; Vlassopoulou 1989, n0s 41-43 et 66; Vlassopoulou 1990,
nOS41-43, 63 et 65. La sima de Toron6 est un intermediairesCrementattique entre, d'une part, les simas
omegariennes>corinthiennesou corinthianisantesdont elle reprend le profil et l'anthemion a double file et
A alternance, et d'autre part, la ceramique attique pour la forme des lotus et l'articulationdes deux files par
des tiges entrelacees. Elle est aussi un intermediairecontribuantA susciter,a Athenes meme, d'autres essais
authentiquementatheniens(simasX/XI a double file de palmettescorrespondantes,Buschor 1929, fig. 20-23,
pl. 1;Vlassopoulou 1989 et Vlassopoulou 1990, no' 41-43) et des imitationscorinthianisantes(simasXIII AXV
et XXVII, Buschor 1929, fig. 31-33 et 37, pl. 3-6 et, 11, avec anthemion a double file et a alternance; cf.
Vlassopoulou 1990, no 56).
6 Courby 1927, pp. 102-103, pl. XIII. Ohnesorg 1993, pp. 38-40, pl. 83.
7 Athenes, Ecole Francaised' Archeologie,Planotheque,dessinno 400; De La Coste-Messeliere1942-1943,
p. 67, fig. 13 au centre et n. 2.
8 Noack 1927, pp. 67-68, fig. 31; Shoe 1936, pl. XVIII, 10;Wiegand 1904, pp. 119-122 (fig. 118), et 124-
126, pl. X, 1:a et b; De La Coste-Messelidre1942-1943, p. 67, fig. 13 au centre et n. 2. Ohnesorg 1993,
pp. 41-44, frontispice,pl. 84, 85.
9 Daux et Hansen 1987, encadrementde la porte: pp. 121-137, pl. 54-60; soffitesde larmiershorizontaux:
pp. 190-197, pl. 87-95; simas: pp. 215-221, pl. 101-103. On peut aussi rapprocherla fleur de lotus de
celles de la frise de calcairemarneux: De La Coste-Messeliereet De Mire 1957, p. 310, fig. 16.
TERRESCUITESARCHITECTURALES,
PEINTURESET MOSAIQUES 3
1.1.3. A l'oeuvre dans le meme temps ou quelques annees plus tard, le peintre des
&picranitisdu Tr~sor des Ath~niens1Is'est form&dans les officines atheniennes de vases; il
y a pr~par&ses cartons,deux anth~mionsdifferent de palmettescirconscritespour la cella12
(P1.2:b), un anthemion de palmettes circonscriteset de fleurs de lotus pour le pronaosI3
(P1.2:a). Ce transfertde la ceramique a l'architecture-que les formes vegetales rendent
possibleddsla derniered~cenniedu Vje siecle14 -n'est pas simplecopie, mais s'accompagne
d'une creation original: on chercheraiten vain sur les vases un anth~mionqui alterne des
palmettescirconscriteset cette forme de lotus que l'on ne rencontre,a la fin de l'archaYsmeet
au debut de l'epoque severe, qu'isoleea l'extremited'une tige (parex. P1.2:c).15Tenu a une
lisibilite et a une symetrie parfaites,le peintre a soigneusementdessine le calice de sepales
A base doublement incurv~e comme si elle &pousaitle trac&de deux volutes divergentes
echappees de la tige. Ce type de calice a deux larges sepales concaves restera longtemps
10 Larmieret simas du Tresor vde Marsellew: Daux 1923, pp. 68 et 78, fig. 69 et pl. XXIII. Les anthemions
delphiquesont ete utilement rapprochespar E. Langlotz dans le cadre d' une interpretationgenerate qui fait
la part belle a Phocee: Langlotz 1975, pp. 47-48, pl. 7. Mais G. Gruben et son cole revendiquentpour
cycladiquesl'architectureet le travail du marbre parien du Tresor de Marseille: Schuller 1991, pp. 88-112
passim. Et une fleur de lotus comparableest encore peinte dans le 2e quart du Ve sur la sima, en marbre de
Paros,du <<Thesmophorion>> de Delos: Vallois 1978, pp. 356-358; Bruneauet Ducat 1983, no 48, pp. 159-160;
Mertens-Horn 1988, pp. 58-59, pl. 14:c;Ohnesorg 1991, p. 176, fig. 9; Ohnesorg 1993, pp. 32-33, frontispice,
pl. 32.
11 Le debat sur la date de 1'edificeoppose desormaistroistheses: 1)celle d'un edifice commemorant,d'apres
Pausanias(10, 11,5), la victoire de Marathon,donc edifie peu apres;2) celle d'un edifice entierementconstruit
dansles dernieresannees du VIe siecle,vraisemblablementpour celebrerla reform de Clistheneet l'avenement
de la democratie (etat du debat et bibliographicprincipal de ces deux theses par Kleine 1973, pp. 94-97,
et T6l1e-Kastenbein1983, pp. 580-58 1; 3) celle d'un edifice construiten marbrede Paros a la meme epoque,
peut-etre donc pour les memes circonstances,jusqu'a la frise comprise, et acheve dans un autre materiau
quelques annees plus tard, a) n~cessairementavant 490, et meme avant l'achevement du temple d'Aphaia
a Egine disent les specialistesplus sensibles aux caracteresanciens non seulement de sa sculpturemais aussi
de l'architecturedes parties hautes de l'edifice (par ex. Bankel 1990, p. 410-412; Bankel 1993, pp. 157-172;
Schuller 1991, pp. 107-111, b) apres 490 disent ceux qui en retiennent au contraire les caracteresles plus
regents (par ex. Busing 1979; Busing 1988, pp. 29-32). Mon propos n'est pas d'alimenterpar la date de ces
decors peints le debat dont Dinsmoor 1946, pp. 86-121, et De La Coste-Messeliere1953, pp. 179-182 ont,
chacun a leur maniere, montre la vanity: il est vraisemblablementanterieura 490 dans les hypotheses2 et 3 a,
postbrieurA490 dans les hypotheses 1 et 3 b. Maisje suis convaincue que ceux des epicranitisprecedent celui
du linteau de la porte (infra,p. 25).
12 Audiat 1933, pp. 45-46, pl. XXIII; effectuepar A. Tournaire,le releve inedit de l'anthemion du mur de
refend(Planothequede l'Ecole Francaised' Arch(ologie, ici P1.2:b)paralitplus fiable que celuit de la pl. XXIII.
13 Audiat 1933, p. 45, pl. XXII.
14 Euphronios, pp. 54-55, 155 et 159 pour les anthemions de palmettes, sans compter les compositions de
surface,par ex. pp. 54-55, 110-111, 127, 133 et 239 (Smikros);pour les lotus a sepales renfles avec decoupe
inferieureen arc de cercle et a petales lateraux individualistsen lame de faux, voir pp. 114, 169, 176 et 227
(Oltos),ou sur l'amphorenicosthenienned'Oltos, Paris,Musee du Louvre G. 2, Arias et Hirmer 1960, fig. 99.
15 Par ex. encore chez le P. de Pistoxenos:Boardman 1989, fig. 68.
4 MARIE-FRAN9OISE
BILLOT
pp. 221-226, ill. 12-15, p1.42. Bankel 1993, pp. 103-107, 112 et 158, fig. 61, 62 et 91:b, p1.32:1-3 et p1.53:2
(dessinsde C. Haller von Hallerstein).
23 Par ex. Corinthe FS 28: Roebuck 1990, p1.7.
6 MARIE-FRANQOISEBILLOT
38 Buschor 1929, pp. 27-30 et 44-45, fig. 31-33, pl. 3, 4 et 11;Vlassopoulou 1990, no 56.
39 Buschor 1929, pp. 40-41, fig. 45, pl. 8.
40 Beazley 1974, pl. 7-9. PMlikE du P. de Syleus (coll. Niarchos, Paris):MunzenundMedaillen A.G.,Kunsiwerke
derAntke,AukionXVIII,29. Nov. 1958, no 114, pl. 36, pp. 39-40; Boardman 1975, fig. 197.
41 Fleur archaYsante, car elle remonte au moins Al'epoque des simas de l'H&catompedon,avec ses petales
laterauxen lame de faux incurv&epour epouserle tracedes tiges, son petale ou pistilcentralen goutte ou larme,
et ses deux groupes de trois 6tamines: Schuchhardt 1935/36, pp. 1-98, pl. 1-21. L'interpretationpeinte
en petit format qu'en donne par exemple l'atelier d'Euphronios,avec le petale central losange ou fusiforme
(Euphronios,pp. 62, 68, 78, 90, 97, etc.), est reprise dans les annees 480-470 et persiste sur les vases et sur
les simas, avec une variante exceptionnelle sur la sima XV dont le lotus remonte aux sources plus anciennes
(ci-dessousn. 46).
42 Buschor 1929, pp. 41-44, fig. 46, 47, 50, 51, pl. 9, 10.
43 De La Coste-Messelidre1942-1943, p. 67; Ohnesorg 1993, pl. 32, 83 et 84.
44 Buschor 1929, fig. 35, 36.
PEINTURESET MOSAIQUES
TERRESCUITESARCHITECTURALES, 9
notamment du groupe traits en <<clair sur sombre>>45 (premiertiers du Ve siecle). En meme
se la
temps, elle situe dans post~rit6directe de la sima 1'H~catomp~donpar les detailsde la
de
facture du lotus au p~tale central en forme de goutte et aux etamines distinctes, arrondies
et groupies par trois46(P1.2:f). Elle assuredonc pleinement la continuityavec le VIe siecle.
De maniere plus triviale, les peintres de simas n'hesitent pas a exploiter toutes les
possibilitiesdes memes cartons, ainsi pour la sima trouvee au carrefourdes rues Eupolidos,
Apellou et Lycourgou(P1.3:f;supran. 33): les seriesX/XI (P1.3:a)Aanth6mionsdoubles de
palmettescorrespondantesen sont le modulepremier;mais la liaison par arceauxverticaux
est abandonn6e au profit de lyres comme sur la serie XIX (P1.3:e); la petite fleur de lis,
A qui lotus et palmettes des simas XIX A XXI (P1.3:e et h) font g~n6reusementplace, se
glisse ici Agrand-peineentre des palmettescong~nitalementtrop rapproch~es.
II.1..3. CHRONOLOGIE.
Dans ces conditions, et faute de trouvailles r~centes stratifi6es, nous ne pouvons
qu'indiquerles quelquespoints sur lesquelsla chronologied'E. Buschordoit etre abaissee.
I.1 .3.a. Les simasX/XI (P1.3:a)restentles plus anciennes,47dates des ann6es520-500
par un unique argument d'ordre strictementstylistique48: en raison de leurs palmettes a
multiplesfeuilleslongues, minces et tres serr6es,elles ne peuvent avoir 6tWconquesque dans
l'orbitede Psiax,49d'Epiktetos,50d'Oltos surtout;51ces palmettesdisparaissentdes la fin du
VIe siecle du repertoiredes vases, et ne subsistentque par copie dans celui des simas (parex.
P1. 3:f) et des antefixes. Deux traits de pure originality,sans module connu sur les vases:
l'alternancedes palmettes a feuilles nombreuseset serrees avec les palmettes habituelles a
feuilles peu nombreuses, larges et espacees, et l'arceau vertical qui relie les deux files de
45 Buschor 1929, pp. 34-35, fig. 37, pl. 5. Pour les simas megariennes en clair sur sombre, voir Le Roy
1967, pp. 121-127, pl. 43-45 et 102: Roebuck 1990, pp. 59-60, fig. 2, pl. 7. Ajouter une piece trouvee A
Sicyone et publi&eici meme par K. Krystalli-Votsis;Krystalli-Votsis1987, pl. 53:a.
46 Cf. n. 41. A d6faut de l'Hecatomp6don, un autre edifice archaYque d'Athenes a pu fournir le module,
ou meme une stele fun6rairecomme celle du New York MetropolitanMuseum, inv. 17.230.6, ouil'on note
que les groupes de s6palessont li6s par un bandeau: Richter 1961, pp. 19-20, fig. 73-76.
47 Font partie des memes series ou leur sont directement apparentes et proviennent du meme atelier les
fragmentsLouvre CA 1687 (Buschor1929, p. 23) et CA 1689, qui proviennentde l'Acropole,deux fragments
trouv6spar N. Platon tombs au flanc Sud de l'Acropole 'a 'Asclepi6ion(Platon 1963, p. 18, pl. 14:4 et y),
un fragmentconservea laJohns Hopkins University(ReederWilliams 1984, n? 67, p. 99), le fragmentd'angle
Mus6e de Agora A 3885 et une sima d'Eleusis.
48 BIest vrai que la plupartdes nombreuxfragmentsont 6t6trouv6sa grandeprofondeurdds 1835, m~l6sa des
restes de charpente carboniseeet qu'ilsportent eux-memes des traces d'incendie: Ross 1855, p. 102; Buschor
1929, p. 24. Sans aucun doute, 1'6dificea &6 incendie. Mais en raison des incertitudesde la stratigraphie
et de la chronologie, nous ne pouvons dire qu'il l'ait W en 480/79 (voiraussi supran. 25).
49 Boardman1975,p. 17,fig. 12.
50 Epiktetos: Boardman 1975, fig. 67; mais dans son oeuvre, les palmettes a feuilles trds nombreusessont
g~n~ralementun peu plus a&r&es, par ex. Boardman 1975, fig. 66.
51 Par ex. 1'embouchurede la c~lbre amphore nicosth6nienneLouvre G. 2: Charbonneaux, Martin et
Villard 1968, fig. 364; et sur la plupart des coupes, de part et d'autre des anses: par ex. Boardman 1975,
fig. 61; voir aussiune coupe de Pheidippos(Boardman1975, fig. 79), une autredu P.de D61os(Boardman1975,
fig. 83), mais sur cette dernidre,les feuillessont d~ji plus individualis~es.
10 BILLOT
MARIE-FRANQOISE
l'anthemion et qui, lui aussi, disparaitpar la suite. Du reste, ces trois disparitionsprouvent
par la negative A quel point le style des simas attiques est soutenu par celui des vases. Ne
subsisteraqu'une nouveaute decorative,la branche de lauriersur le listel de couronnement;
1'etroitessedes margeschronologiquesne permet pas de dire s'il s'agit,la encore, du transfert
aux simas d'un motif inauguredans le meme temps sur les vases, ou si, bien plutot, le peintre
des simas en a donnedl'idee a un atelierdu debut du Ve, d'Hermonax par exemple.52
A la fois tetes de liste et sans suite, les simas X/XI represententencore maintenant,
dans le contexte athenien, le premier transfertd'un profiledu marbre a la terre cuite, et,
avec la sima de Torone, la premiereadaptationdu repertoiredes vases en architecture,sous
l'influencedes simas <<m(gariennes>>.
I. 1.3.b. La sima XV (P1.3:d),les tuiles d'egout XVIII et les antefixesXIV qui lui sont a
peu pres sufrementassociees53represententune autre forme de compromisentre influence
corinthienneet le style athenien, entre le marbreet la terrecuite. Malheureusement,rien ne
permet de preciserleur date dans le premiertiers du Ve siecle.
I.1 .3.c. Alors que Buschorpensait pouvoir situer les simas XII a XIV dans le premier
quartou tiers du Ve siecle,54la facturede l'anthemionau lotus archaisantne les rend guere
possiblesavant 480 (voirci-dessus).
L'apparitiondu motif de lyres horizontalesgarnies de palmettes vers 490, notamment
sur les amphores de Nole (P. d'Eucharides,de Pan, de Dutuit etc.) etablit le terminus a quo
des simas XVI a XXI,55 ce qui n'impose evidemment pas que les plus anciennes d'entre
elles, XVI et XVII, doivent etre datees aussitot56: la file de lyres a palmettesdure au moins
jusqu'a la fin du troisieme quart du siecle et connalitson epoque de gloire sur les crateres
et les hydriesde 465 a 430 environ.
La palmette a coeur remplacepar une palmetteplus petite a feuillesmultipleset serrees
des simas XII a XVII et des antefixesXII a XVI57 est empruntee aux steles pariennes du
dernierquart du VIe et du debut du Ve sicle.58
a quodes simas XII a XVII et des antefixesXII a XVI se situe
En definitive,le terminus
dans la decennie 480-470.
52 Boardman 1975, pp. 193-194, fig. 352.
53 Buschor 1933, pp. 17-18, fig. 25 et pp. 48-49, fig. 61-63, pl. 9; Vlassopoulou 1990, n0s 58 a 60; ensemble
date entre 480 et 450 (Buschor1929, p. 39).
54 Buschor 1929, pp. 24-30.
55 Jacobsthal 1927, pl. 51, 52, 74; Boardman 1975, fig. 164 (P.d'Eucharides)et 339 (P.de Pan);Petit Palais
G. 203 du P. de Dutuit: Kurtz 1975, pp. 14 et 125-126, pl. 55:3. Cratdretrouve A Agrigente, du groupe
de Pezzino, 500-490 av.J.-C. : Schefold 1975, fig. 309.
56 Buschor 1929, pp. 35-38, fig. 38-43, pl. 6 et 7; datees par E. Buschor,p. 39, entre 480 et 450, ce que
n'autorisepas le soffitede la sima XVI, qui impose une date posterieureau milieu de siecle.
57 Buschor 1933, pp. 45-50, fig. 59-64, pl. 7-10; Vlassopoulou 1989, n0s 46-49; Vlassopoulou 1990,
nOS46-49 et 59. Hubner 1973, pp. 73-77, fig. 1 et 2, pl. 57:1.
58 Steles sculptees a Paros et exportees vers Naxos, Thera, Amorgos, Egine, Thebes, Samos, l'Eub&e
(Karystos)et Salamine. Certaines presentent une petite palmette superposee A la grande en guise de coeur:
Buschor 1933a, pp. 43-46, Beil. XV und XVI; Karousos 1934, p. 61, fig. 46; Marangou 1986, pp. 124-125,
pl. 49, 50. Deja sur une palmette d'angle du geison horizontaldu Tresor de Siphnos: Daux et Hansen 1987,
pl. 87, 88.
TERRES CUITES ARCHITECTURALES, PEINTURES ET MOSAYQUES 11
Ainsi, les simasXII a XV, les antefixesXII et des tuilesd'egoutXIII (ca 450), XIV et XV,
dont plusieursfragmentsont 6te trouvessous la couche de d(chets de marbredu Parthenon
(ci-dessusn. 28), sont de nature a renforcerdoutes et recherchessur le contenu et la date des
couches en place et des remblaissuperposesentre le Parthenonet le mur Sud de l'Acropole,
et sur les circonstancesreelles dans lesquelles ils se sont progressivementaccumules. Elles
confirmentque 1'ona recommencea construiresur1'Acropolepeu apresSalamineet Platees.
II.1.4. L'ARcHrrEcTuRE,DOMAINEDISTINCT.
Les simas constituentpar ailleursune productionspecifiqueet destinee Ades edificesoNu
la lisibilitedes decors est primordiale;aucun detail n'est donc laisse au hasardde 1'execution,
meme plus ou moins heureuse et soignee. Cette differencede nature et d'echelle avec les
vases stimule la precision du detail, et suscite par exemple la creation de calices de sepales
(simasXVII [P1.3:g], XIX, cf. XX),59 traditionnelspour les fleursde lotus libresdes vases de
la fin de 1'archaismeet du style severe,60mais inhabituels dans le type d'anthemion que
reproduisentles simas XII, XIII, XIV et XV (P1.3:b-d).
II.1.5. REFLEXIONS SUR LE DECOR ARCHITECTURAL PEINT A LA FIN DU VIe ET DANS LA
PREMIERE MOITIE DU ye SIECLE.
Nous venons d'observer l'ampleur des transfertsde decor des vases aux simas et aux
antefixesde terre cuite.
Nous avions aussi note deux exemples de transfertsdu repertoiredes vases A celui des
marbres,les epicranitisdu Tresor des Atheniens(I.1.3, P1.2:a, b) et les <<Palmettensimen>> de
l'Acropole (I.1.4, P1. 2:d, e): ils s'accompagnentd'une interpretationplus structures des
fleursde lotus et de leur insertiondans des types d'anthemionsqui, a meme epoque, ou bien
ne comportent pas de fleur (comparerl'anthemion canonique, de palmettes seulement, de
la cella du Tresor des Atheniens, a l'anthemion cree pour le pronaos), ou bien en offrent
des formes differentes.
Nous avons par ailleurs constate qu'a la fin du VIe et au premier tiers du Ve siecle,
la peinture sur marbre atteste des liens tant avec la sculpture, notamment dans l'orbite
des architecturescycladiques-sima du Temple des Alcmeonides(P1.1:d),sima et chapiteau
d'ante du Temple d'Aphaia(P1.1:e et f), simadu Thesmophorionde Delos-qu'avec le decor
des vases (palmettesdes simasdu Telesteriond'Eleusis,du Temple d'Athenades Pisistratides,
du Temple des Alcmeonides [P1.1:d], du chapiteaudu Temple d'Aphaia [P1.1:f], epicranitis
du Tresor des Atheniens [P1. 2:a, b], <<Palmettensimen>> de l'Acropole [P1. 2:d, e]). Le
repertoiredes simas et antefixesde terre cuite, lui, ne s'inspireque de celui des vases; leurs
anthemions archaisantsassurent du reste une tradition inattendue de la sima en marbre
du Vieux Temple d'Athena (ca 566, P1. 2:) aux simas XII et suivantes, notamment XV
(P1.3:b-d), par l'intermediairedes vases d'Euphronios,du P. de Pan et de beaucoup d'autres.
Toutefois, nous n'avons pas reconnu de liens privilegiesentre peintres sur marbre et
peintressur terre cuite. Du seul recoursdes uns et des autresaux decors de vases relevent et
les parents stylistiqueset les differencesd'interpretation: empruntesa la meme ceramique
a figure rouge de l'archaisme recent, les lotus des simas XII A XIX d'une part, ceux de
59 Buschor 1929, fig. 41 et 45, pl. 7-9; Vlassopoulou 1990, n' 63.
60 Par ex. encore sur la coupe de BerlinF 2547, du P. du Mariage, 460-450 av.J.-C.: CVA,Berlin3, pl. 109.
12 BILLOT
MARIE-FRANQOISE
II.2. 1. LEsCOMPOSITIONS.
file simpledepalmettes
I.2. l.a. Antlzmions etdefleursdelotus.
circonscrites
* Athnes, Agora, portique, epicranitis naguere attibuee a la Stoa Poikil,62 (P1.4:a,
restitutionaquarelleepar M. Welker).
Thompson 1950, p. 327; Meritt 1970, pp. 236-237 et 250-25 1, pl. 62 (A 1710) et 64:b;
AgoraXIV, pp. 90-91.
Le lotus s'ouvreau bout d'une tigelle adventice.
La modenature de l'ensemble architecturalauquel appartientcette epicranitisse laisse
daterpeu avant le milieu du Ve siecle,juste avant l'Hephaisteion.
* Athenes, Agora, linteau au lionnes peintes (P1.4:b, restitution aquarellee par Piet
dejong).
Stevens 1954, pp. 169-184, pl. A et 39-48, en particulierpl. 39, 41-46; AgoraXIV, p. 31
n. 32.
Date dans le 3e quart du Ve siecle. Parmi les emplois possibles, G. P. Stevens propose
qu'il provienne d'une fenetre ouverte dans le mur de refend qui spare le pronaos et la cella
de l'Hephaisteion.
* Athenes, Temple de l'Ilissos,architravedu pronaos, fasce superieure(P1.4:d).
61 On ne connait le decor de l'epicranitisdu grand temple d'Athena Sounias que par une mention de
Dinsmoor, Jr. s.d., p. 49: ((The crowningcourse of the wall had a delicate lotus-and-palmettedesign>.
62 Depuis que la Stoa Poikilda ete effectivementmise au jour, cette attributionn'est plus rappelee, car elle
mbritesans doute plus ample etude: Shear 1984, pp. 1-57, pl. 1-16; Camp 1986, pp. 68-72; Camp 1990,
pp.101-109.
PEINTURES
TERRESCUITESARCHITECTURALES, ETMOSAMQUES 13
Dessin de J. Stuart pour Stuart and Revett 1762, chap. II, p. 10, pl. VIII fig. 3
(cf. fig. 2A).63
II est probable que ce decor faisait partie des finitions proches de la mise en place
de la frise. Connue par la publication des Dilettanti, triesproche de celle du Temple
d'Athena Nike, l'architecturedu Temple de l'Ilissos a longtemps ete attribute au meme
architects, Callicrates. B. Wesenberg,faisant la synthese entre les donnees epigraphiques,
arch~ologiques,architecturaleset stylistiques,a recemmentmontre que le Temple d'Athena
Nike a &tecommence dans les annees 437-435, en tout cas en meme temps que les travaux
d'extension du Bastion et peu apres le debut de la constructiondes Propylees, sur un plan
qui reproduit le temple archaique et tient compte de l'environnementarchitectural,mais
n'a pas de rapport organique avec celui du Temple de l'Ilissos.64 Dans l'tat actuel de la
recherche, le sort des deux edifices semble donc devoir etre disjoint,meme si l'analyse des
formes architecturaleset des elevations les rapprochel'un de l'autre et des Propylees, et si
l'attributiondu Temple de l'Ilissosa Callicratesn'est pas serieusementebranlee.
La frise a longtemps oppose les tenants de la date haute proposee par E Studniczka
(annees 448-445) a ceux d'une date basse;recemment, C. A. Picon et A. Krug65ont plaid
les ann~es425-420. Par ailleurs,selon B. Wesenberg,le styledes baseslaisseraitsupposerque
l'edifice a ete commence lui aussi aux environsde 435, peu de temps apres celui d'Athena
Nike; acceptant la date recente de la frise, il suppose pour cet edifice ce qu'il essaie de
demontrer pour le Temple d'Athena Nike (voir ci-dessous),une interruptiondes travaux
par la guerre du Peloponnese en 432/1 et une reprise en 425/4. Dans le meme temps,
M. M. Miles defend la these d'une constructionmenee a bien en triespeu de temps dans
les annees 435-430, 6poque retenue par E. B. Harrisonpour la frise.66
Quelques annees plus tard,W A. P. Childs defend a nouveau une date haute de l'edifice
qui aurait e entierement construitdans les annees 445-440.67 Mais il faut bien dire que
l'argumentationqu'il developpea partirde l'analysearchitecturaletendraitplutot a soutenir
celle de M. M. Miles et Adater 1'&difice, au moinsjusqu'a 1'architrave,des annees 437-432,
car les parents de son elevationavec celle des Propyleeset du Temple d'Ath~naNike restent
indeniables.
La date du decor oscille donc, provisoirement,entre 432 et 420 av. J.-C.
* Athenes, Parthenon,sima (P1.4:f, g).
Paccard 1845-1846 (CollectionEcole des Beaux-Arts[EBA],prise en charge n? 2208):
1)le dossierd'etudespreliminairesqui ne fontpas partiede l'Envoiproprementdit contientun
relevesurcalqueinedit(iciP1.4:f): au recto, a la mine de plomb (en clairsurla photographic),
releve effectuesur deux ou troislotus a cinq petalesvariantlegerementde largeuret de place
par rapport A l'axe vertical de la fleur, et a deux sepales de hauteurs differentes d'une
fleur a l'autre; au verso, au crayon gras pour permettre un report sur un dessin definitif,
63 Redessin6dans 1'6ditionfransaisede C. P. Landon (FirminDidot 1808), chap. II, pl. XII, fig. 3, et dans la
deuxieme edition anglaise de 1825, qui reproduiten fait 1'&ditionfransaise.
64 Wesenberg 1981, pp. 28-54.
65
Picon 1978, pp. 47-81; Krug 1979, pp. 7-21, pl. 1-9; Ridgway 1981, pp. 88-89.
66
Miles 1980, pp. 309-325, pl. 91-96; bibliographicet 6tat de la question pp. 309-310; Harrison 1981,
p. 233.
67
Childs 1985, pp. 207-251, pl. 43-45; (tat de la questionpp. 207-211.
14 MARIE-FRAN9OISE BILLOT
le dessin d'E. Landron pour Le Bas 1847, pl. II, 6, no I, avec une attributionerronee au
Parth~non,79d'autrepart le dessin de T. I. Willsonpour Penrose 1851, pp. 59-60, pl. XXV,
en haut a gauche (ici P1.6:a en bas "agauche).80Sur le premier,les tiges qui <<encerclent>les
palmettesse rejoignentpour donner naissanceaux calicesdes fleurs: c'est, en quelque sorte,
un anthemion de palmettes et de lotus inscrit dans un carre; sur l'autre, elles se referment
au-dessusde la palmetteet les fleurs,depourvuesde sepales,ne sont pas relieesaux palmettes.
Cette deuxieme lecture a ete retenue par Bohn 1882, pp. 9-10 et 21, pl. XII, 4.
* Un caisson rectangulaireappartenantau passage occidental, decor de quatre pal-
mettes circonscritesdeux a deux opposees par le sommet: T. I. Willson pour Penrose 1851,
pl. XXV, au milieu Agauche (ici P1.6:a en bas a droite). Bohn 1882, pp. 21-22, pl. XII:3.
Peut-etreMusee de l'Acropoleinv. 8835, mais le decor est triesefface: voir infra11.2.1.d.
II.2.1.c. Anthemions a file doubledepalmettos
et defleursen correspondence,
relives
par destigesen
Sformantunefiledelyreshorizontales etdeux'adeuxantithitiques.
* Athenes, Agora, portique, chapiteaud'ante attributenaguere a la Stoa Poikile(P1.5:a:
restitutionaquarelleepar M. Welker).
Thompson 1950, pp. 327-328, pl. 103:a et b; Meritt 1970, pp. 236-237, 238-240 et
247-248, fig. 5, pl. 63:a et b et 64:a et c; AgoraXIE, pp. 90-91, pl. 49:a et b.
Ensemble architecturaldont provient l'epicranitisrecensee ci-dessus en 11.2.1.a. Peu
avant 450 av. J.-C. d'apresla modenaturedu chapiteau.
* Athenes, Hephaisteion,geison et epicranitis(P1.5:c, d).
* Epicranitis-assiseporte-caissons de la peristasis (P1. 5:d) : Stuart et Revett 1794,
chapitreI, vignettep. 1 (dessindeJ. Stuartou N. Revett, redessinepar W Reveley),81pl. VII
et VIII.82 Selon toute vraisemblance,c'est ce decor que E. Schaubert a releve en 1835 ou
1836 et dont H. Koch a public le dessin,83malheureusementsans commentaire : Koch
1955, p. 194, fig. 83, ici P1.5:d.
* Geison (P1.5:c): Paccard 1845-1846, no 9 (<<Entablement restaure>>,Paris-Rome-Atkines,
la
pp. 162 et 167), reporte sur face anterieuredu larmier du Parthenon un decor qu'il a
releve(sur le larmier de l'Hephaisteion(P1.5:c) : <<Onvoit donc qu'excepte 1'ornementque
j'ai mis dans la face du larmier, sequela le memedessinet les mimescouleurs queceluiqui existe
encoredansle larmier dela cornichze dutempledeThdsle,je n'ai presquerien a ajouterpour restaurer
entidrementla decoration exterieure du Parthenon>> (Memoire,p. 71 = Paris-Rome-Athenes,
p. 365). La rigueur de son travail preparatoireet 1'utilisationde calques pour relever les
motifs qu'il a reconnus (Memoire, pp. 67-68 = Paris-Rome-Athines,pp. 364-365; voir ci-dessus
pour la sima) pretent A ce dessin une grande fiabilite. Du reste, le decor est pratiquement
79 Repetee par Reinach dans Le Bas 1888, p. 134. Reproductiondans Paris-Rome-Alhznes,
p. 97, fig. 59.
80 Planche en couleur reproduitedans Paris-Rome-Atns,p. 101, fig. 62.
81 Reproduitdans Paris-Rome-Atns,p. 64, fig. 34.
82 Elles correspondentaux planches X et XI de 1'editionfransaise publi&echez Firmin Didot en 1812, et
de la deuxidmeedition anglaise qui reproduitl'edition fransaise.
83 K. Iliakissemble avoir cru qu'il s'agissaitdu d~cor d'un chapiteaud'ante de
1'H~phaist~ion:Iliakis 1976,
p. 256 et n. 55, avec renvoi AKoch 1955, pp. 99-100. Or H. Koch, reunissantles temoignagesauxquelsiljoint
le sien et l'analysedes dessinsd'E. Schaubert,rappelleque la restitutiond'un decor de ce genre sur le chapiteau
d'ante a bien et proposee, mais par G. Semper, et qu'elle n'est nullementjustifi~e par les rares vestiges de
couleur que P'ona observesou cru observerautrefois.
18 MARIE-FRANQOISEBILLOT
Selon A. Delivorrias, la frise et l'unique piece qui subsiste du fronton Est confirment
l'epoque de construction proposee par W B. Dinsmoor (444-440 av. J.-C.), tandis que
B. Sismondo Ridgway abaisseraitvolontiers la date des sculpturesjusqu'aux environs de
420, mais sans developpergrande argumentation.86Le profilearchaisantde la sima, proche
de celui de la sima du Temple d'Aphaia, et l'acrotere, que son travail et son style pariens
placent dans la descendance des acroteres d'Egine, sont anterieursau Parthenon. Ca 440
av.J.-C.
Ce petit corpus comporte quatre variantes structurelles. L'une concerne le trace des
tiges, qui, sur la sima de Sounion comme sur la sima XVII de l'Acropole87se rejoignent
sous le lotus, occultantle motif de lyre. Les deux autresconcernentl'insertionde la fleur: au
Tr~sor dorique de Marmaria et au Temple de Pos(idon a Sounion, elle est posee> dans
l'axe des spirales,sans lien organiqueavec les tiges: l'anthemionrepond triesprecisementa
la definition donn(e ci-dessus;sur le chapiteau de l'Agora d'Athenes et a l'H~phaist~ion,
la fleur s'ouvre a l'extremite d'un tigelle adventice : il s'agit en fait d'un anthemion a file
double de palmettes en correspondence,relies par des tiges en S formant une file de lyres,
avec fleurs en correspondence sur tigelles adventices (voir ci-dessous II.2.1.e). Le linteau
du Tr~sor des Atheniens est trop peu visiblepour qu'on en puissejuger. Enfin, au Tr~sor
dorique de Marmaria, comme sur les simas XVII et XX/XXI de l'Acropole, les lyres sont
garnies de palmettes.
Mais en tout &tatde cause, il parait improbableque cet ensemble, a lui seul, ne reflete
pas l'existencede <cahiersde motifs>>, de repertoires,ce dont le groupe 11.2.1.e nous donnera
confirmation.
II.2.1.d. Compositionsdesurface depalmettes,
defteurset delyresdeux deuxantitletiques.
* Athenes, Propylees,caissonsrectangulaires(P1.6:a).
Mus~e de 1'Acropole,inv. 8800, 8801 et peut-etre 8835. Mais le d(cor de ce dernier
est triesefface, de sorte qu'il pourraitreleverdu motif II.2.1.b (supra).
Dessin trdsfiddle de T. I. Willson pour Penrose 1851, pl. XXV, au milieu a droite (ici
P1.6:a en haut a droite).
II.2.1.e. AntkmionstzfilesimpledepalmettesreliespardestigesenS avecfleurssurtigesadventices.
(Pourl'anthemion a file double, voir ci-dessusen II.2.1.c le chapiteau d'un portique de
l'Agora et le geison de l'Hephaisteion.)
* Delphes, Pilier des Mess~niens,couronnementpeint.
Jacquemin et Laroche 1982, pp. 192-200, fig. 5.
Monument date par le profiledu couronnement,<prochedes chapiteauxd'ante dorique
des monuments attiquesdu troisiemequartdu V' siecle>> (p. 194), en particulierde celui dit,
encore en 1982, de la Stoa Poikile(p. 204), recens6 ci-dessusen II.2.1.c. Les circonstances
politiquesjustifieraientson erectiondans les ann~es450-440. Toutefois,pour autantqu'elles
soientlisibles,les feuillesdes palmettesa double courburepourraientindiquerle derniertiers
ou quart du siecle.
* Athenes, H(phaisteion, simas d'egout et de rampant.
86 D6ivorrias 1969, pp. 127-142, pl. 57-75, Beil. 1; DElivorrias1974, pp. 61-86; Ridgway 1981, p. 40
et n. 3 et pp. 84-85.
87 Buschor 1929, pp. 36-38, fig.41, pl. 7.
20 MARIE-FRANQOISEBILLOT
Dinsmoor 1940, pp. 32-37, fig. 11-14. Dinsmoor 1941, pp. 110-1 16, fig. 41 et 43.
Dinsmoor, Jr. 1976, pp. 232-240, ill. 3, 4, 7, 9 et 10, pl. 39 et 40 (tres importante n. 29).
Dessin de E. Schaubertreproduitpar Koch 1955, pp. 65-66 et 188, fig. 66-67. Les dessins
d'E. Schaubert et de W B. Dinsmoor, Jr. concordent: fleur a petales minces peu evases,
palmette e1anc&ede 13 feuilles a une courbe, assez raides. Sur les dessins publics dans
Dinsmoor 1940, p. 33 fig. 12 et dans Travlos 1971, fig. 344, les feuilles des palmettes ont
un point d'inflexion.
Les simas sont contemporainesau plus tat des frises, en tout cas des frontons (ou du
plus recent des frontons). Mais les dates de ceux-ci, et leur composition meme, sont tries
controversies.88Le d~batengage par E. B. Harrisonsur la divinitydu temple (H~phaistoset
Athena H~phaist~ia?Art~misEuclkia?)ne permet pas de consider 416/5, date d'acheve-
ment du groupe d'Alcamene, comme un terminus assure;or L. T. Shoe a date le profiledes
simas de cette 6poque.89 Les frises sont dates peu avant 431,90 ou dans les ann~es qui
suivent,91mais en tout cas avant celle du Temple d'AthenaNike. La seconde serie de lettres
qui marquent les caissons92peut etre date a partir de 430. Il paralt donc raisonnablede
placer la sima dans les annees 425-420, d'autant plus qu'elle est triesproche de celle du
Temple de N~m~sisa Rhamnonte.
* Rhamnonte, Temple de Nemesis, simas d'egout et de rampant.
Iliakis 1976, pp. 245-247, fig. 1-2, pl. 51; Iliakis 1987, pp. 17-20, fig. 4-6, 8, 12.
L'architecturedu temple indiquequ'ila ete construitdansla decennie 430-420, pendant
que la statue de N~m~siset sa base 6taient61abor~es.93
La sima, presque identique a celle de l'Hephaisteion, date donc de meme vers 420
av.J.-C.
IL.2.1.f. Compositions
desurface
depalmettesetfleursreliespardestigesenS.
* Sounion, Temple d'Ath~naSounias, caissonsdu plafond de la p~ristasis.
Dessins de N. D.J6annitis pour Stais 1899-1900, col. 127-130, pl. 9 a droite.
II s'agit a nouveau d'un anthemion inscrit dans un carre. L'edifice, dont une grande
partie a e transfereevers la fin du Ier siecle ap. J.-C. pour etre reedifi&eau Sud-Est de
l'Agora d'Athenes,94semble dater de la deuxieme moiti6 du Ve siecle.95
* Athenes, Parthenon,caissonsde la peristasis,types A (Nord et Sud) et B (Estet Ouest)
(P1.6:c).
88 D6livorrias 1974, pp. 16-60; en voir le compte rendu par Harrison 1976, pp. 209-210; Harrison 1979,
p. 220; Ridgway 1981, p. 29, n. 29 et p. 40.
89 Shoe 1936, p. 108.
90 Delivorrias 1974, pp. 49-51; Ridgway 1981, pp. 85-88: <<Adate in the late 430's or shortly afterwards
should not be wide of the marko.
91 Von Bockelberg 1979, pp. 23-50, pl. 10-48; compte rendu par Harrison 1981, p. 233.
92 Wyatt et Edrnonson1984, p. 161.
93 Miles 1989, pp. 226-227, avec bibliographicant~rieure.Completerpar P~trakos1987, pp. 277-282.
94 En dernierlieu Dinsmoor,Jr. 1982, pp. 429-431; Camp 1990, pp. 147-150, fig. 96 et 97.
95 Dinsmoor,Jr. s.d., pp. 40-49. L'auteuravait bien voulu me dire, il y a quelques ann~es, que les caissons
devaient etre compl~temmentr6etudi6s,mais que la compositionde leur decor, telle qu'elle avait 6t6 (fortmal)
dessin6e par N. D. J6annitis, etait fiable dans ses grandes lignes. H. A. Thompson et W. B. Dinsmoor, Jr.
projetaientde republierle temple.
TERRES CUITES ARCHITECTURALES, PEINTURES ET MOSAIQUES 21
Sounion (P1.5:e) et Pilier des Messeniens oui les sepales sont presque aussi epais que les
p~tales. Le calice peut aussi presentertrois s~palesdistincts(Tr~sordorique de Marmaria,
P1.5:b) ou plus ou moins jointifs (linteaudu tresor des Atheniens, simas de l'Hephaisteion
et du temple de Nemesis a Rhamnonte)
* F 4 : vers 440 apparait aussi une fleur a 5 petales longs et effiles, deux filiformes et
les trois autres plus larges (caissonsA et B du Parthenon, P1. 6:c; anthemion peripherique
du Temple d'Ath~naNik6), ou bien tous les cinq d'importancepresque 6quivalente(simas
du Parthenon [P1.4:f et g] et du Temple d'Athena Nike). C'est, a plus grande echelle,
interpretation d'un nouveau modele de fleur cree dans les ateliersdu P. d'Altamuraet du
P. des Niobides : sur leurs vases, plusieursvariantes(P1.4:e) coexistent avec 1'ancienmodele
archaisantet le remplacentprogressivement.104
II.2.2.c. Syntaxe.
Ici encore se melent survivances,empruntset nouveautes.
* S 1 : la palmette circonscriteperdurejusque vers 420, comme sur les vases et tres
probablementsur les simas de terre cuite (P1.3:). Les tiges dessinent autour des palmettes
des ogives plus ou moins etirees dont les exemples abondent aussi sur les vases; l'effet
s'apparente a celui des compositions de lyres couchees garnies de palmettes. Ici encore,
harmonies et correspondencesse rencontrent d'une composition a l'autre et du domaine
de la ceramiquea celui de architecture.
* S 2: mais l'anthemiona palmettescirconscrites--picranitis de l'Agora(P1.4:a),linteau
aux lionnes peintes (P1.4:b), architravedu Temple de l'Ilissos(P1.4:d)-est toujoursa simple
file, contrairementa ce que P'onobserve sur les simas de terre cuite.
104 Par ex. Louvre G. 341, Arias et Hirmer 1960, fig. 173-181; Arias et Hirmer 1957, fig. 13, 37, 42, 43;
Caskeyet Beazley 1954, no 108, cratereinv. 33.56, p. 77, pl. LVIII; CVA,Bologna,MuseoCivico4, III, 1, cratdre
en cloche Pell. 286, p. 13, pl. 75 (P.de Blenheim).
24 BILLOT
MARIE-FRANQOISE
* S 3 : l'anthemionde palmettesseules n'a plus cours, alors qu'il est toujoursen vigueur
surles vases: l'epicranitisde la cella du Tresordes Atheniens(P1.2:b)et les <<Palmettensimen>>
(P1.2:e) restent sans posterity.
* S 4: il n'y a pas, pour autant, ruptureavec les compositionsde palmettes qui s'offrent
quotidiennement aux yeux de tous sous les anses de skyphoi, de stamnoi, d'hydries, de
crateres: un caissonet le soffitedu geison d'angledu Parthenon105ainsique deux caissonsdes
Propylees(P1.6:a, b et d) y ont leursmodeles trop evidents(cf. P 2), seulementcontraints,ici,
par le cadre quadrangulaire.Comme au Tresor des Atheniens et sur les <<Palmettensimen>>,
il s'agit de transfertsdirectsdu repertoiredes vases a architecture.
* S 5 : le repertoiredes marbresignorel'anthemionAdouble file en alternance(palmettes
opposees aux fleurs et viceversa).
* S 6 : la sima du Temple de Poseidon au Sounion (P1.5:e) est jusqu'a present la seule
sima de marbre Aporter un anthemion Adouble file : son profil archaisant,proche de celui
de la sima du temple d'Aphaia(supran. 22) et parentde ceux des simasde terrecuite, favorise
peut-etrel'empruntde cette structuredecorativeparticulierementbien adaptee, nous l'avons
vu (II.1.2.b), a ce genre de profils.
* S 7 : tous les autresanthemionsa double file se developmentsur des partiesdifferentes,
offrantdes faces planes verticales: chapiteaud'ante de l'Agora(P1.5:a), geison et epicranitis
de l'Hephaisteion (P1.5:c et d), assise decore(edu Tre(sordorique de Marmaria (P1.5:b),
linteau du tresor des Atheniens. De la terre cuite au marbre, les lieux d'applicationd'un
meme schema changent.
* S 8 : l'anthemion a double file en correspondencen'estjamais autrementcon~u (voir
S 2) que comme le redoublement,par effet de miroir,d'un anthemion simple reliefpar des
S couches; d'ou'le motif de lyre, garni (Tresordorique de Marmaria, P1. 5:b) ou non de
palmettes.
H. Busing consider que les decors du Tresor des Atheniens et du Tresor dorique sont
contemporainsdes edifices,du moins de leur achievement.C'est une hypothese economique
et vraisemblable,d'autantplus malaisee a discuterqu'il faut en bonne regle dater un decor
par le contexte archeologiqueet non ce dernierpar le style d'un decor.106Encore le cas est-il
differentd'un grand edifice dont la polychromiene sauraitetre completedou rafraichiesans
nouveaux et coufteuxechafaudages,et d'un tresor toujourssusceptibled'embellissementsa
peu de frais.
Une fois etabli que la constructiondu Tresordes Atheniens a ete interrompue(ci-dessus
n. 11), le problem de la chronologie absolue, du debut des travaux a l'achevement de
l'6difice,reste entier.
Etant donn6 que la file de lyres horizontales,deux a deux antithetiqueset garnies de
palmettosapparaltsur la ceramique des annees 480 (supraII. 1.2.a et II. 1.3.c), il n' y a pas
105 La comparaisonavec la ceramiquea ete faitemaintesfois, d'abordpar Paccard(Paris-Rome-Athhnes,
p. 365),
en dernierlieu par Orlandos 1976-1978, pp. 249-253, fig. 161.
106 Busing 1979 admet, ce que avons fait pour d'autresedifices,que la peinture du linteau relievedes ultimes
finitions;il montre que le decor ne peut, pour des raison stylistiques,dater qu'apres480. Dater des finitions
superficiellesne permet pas de dater la constructiondu Tre'sor,le deroulement des travaux anterieurs, leur
debut avant ou apr~s490, ni meme de l'achevementde l'essentiel,pose des sculpturestympanalescomprise,
avant ou apres 490.
PEINTURESET MOSAIQUES
TERRESCUITESARCHITECTURALES, 25
surlessimasdeterrecuiteetsurlesmarbres
II.3. Regardsynoptique peintsdu Ve siecle.
Lesparcoursseparesdu repertoiredes simaset de celui des marbres,tous deux redevables
aux peintres de vases, peuvent ainsi reveler,au-dela de leurs particularites,des structureset
des formes communes. Ils invitenta une presentationsynoptique,quinevautpaschronologie, ni
mais qui met en evidence leur parents originelleet, partant,des harmonies
ni absolue,
relative,
110
CVA,Bologna,MuseoCiuko4, III, 1, inv. Pell. 300, pp. 16-17, pl. 86-87 (420-410): noter le contrast
entre la zone sup~rieure,dont la formule remonte au dernier quart du VI' siecle, et la zone inf~rieuredes
anses outles tiges en S et surtoutla fleur a calice de sepalesreproduisentles motifsde la peinturesur marbredes
annees 450-420; detail dans Boardman 1989, fig. 179.
111 Par ex. Beazley 1963, inv. 03.821, p. 81, pl. CII et supply.27; CVA,Bologna,MuseoCivico4, III, 1, inv.
Pell. 301, p. 16, pl. 83-85.
112
Hubner 1973, p. 119.
113 aveclyresgarnies sur le cratereSchloss
etafteurssurdgellesadventices
Andthnion a doublefile
depalnettes, depalinettes
Fasanerieinv. 77 (CVA,Adopliseck, SchlossFasanerie,pp. 32-36, pl. 46-48) et sur le crateredinoide Mus~eJ. Paul
Getty inv. 87.AE.93. Antliniona simplefile depabnettes, avectges enS etfteurssurtgellesadventicessur le crat~rede
Bologne, Museo Civico inv. Pell. 301 du P. de Cadmos (CVA,Bologna, MuseoCuico4, III, 1, p. 16, pl. 83-85); sur
le craterede 1'Ermitageinv. St. 1807 du meme peintre(Metzger1951, pl. XXVII, cf. pl. XXV); sur 1'hydriedu
BritishMuseum inv. E. 205, du P. de Nicias: CVA,BritishMuseum6, III, 1, p. 6, pl. 93:1a (cf. Hibner 1973,
fig. 15). Ces changes dans le strictdomaine de la peinturene sont qu'un aspectde la circulationdes motifs,par
des itinerairesdivers, entre tous les arts qui, en dernierressort,releventdu dessin: le meme anthemion simple,
mais A palmette flamm&e,se rencontre sur une hydrie du P. de Modica (CVA,Syracusa,R. MuseoNazionak,
inv. 38031, p. 12, pl. 26-28), sur un cratere recemment mis au jour a Corinthe (Williamset Zervos 1991,
p. 6, no 2, pl. 1), et naturellementsur le couronnement de la <<Steleau chat> (Athenes, Musee National inv.
715); il n'est pas (encore)attests en architecture.
PEINTURESET MOSAIQUES
TERRESCUITESARCHITECTURALES, 27
decorativesindependantesdes differencesmaintenuespar la separationdes activitieset des
traditionsartisanales.
C'est ainsi que pouvaient s'offrirsimultanementaux regards,sans heurterle gouit
* la sima XII de 1'Acropoleavec ses lotus triesetroits(P1.3:b)et l'epicranitisd'un portique
de 1'Agora(P1.4:a);
* la sima XVII (P1.3:g), celle du temple de Pose(idona Sounion (P1.5:e) et les caissons
A et B du Parthenon (P1.6:c),"14tant pour la structuredes anthemions que pour la forme
et le detail des fleurs;
* la sima XXIII (P1.3:i) et l'anthemion peripheriquedu Temple d'Athena Nike pour
les arceaux, le calice des fleurs115et la libertydes palmettes;116
* les simasXIX et XX/XXI (P1.3:e et h), l'assisedecoree du Tresordoriquede Marmaria
(P1.5:b),les anthemionsd'unportiquede 1'Agora(P1.4:a et 5:a),ceux de l'Hephaisteion(P1.5:c
et d) et le chapiteau d'ante du temple de Nem~sis a Rhamnonte, pour les lyres garnies de
palmettes. Les simas de terre cuite accusent le statut ambigu de la fleur sur tige adventice,
statutassociedu reste a la nouvelleforme de fleuretroitea troispetales(F 3): surla simaXIX,
cette <fleurde lis>>est effectivementun element surajoutequi se glisse en surnombreentre
la palmette et la fleur de lotus canoniques1 17 (de meme entre les palmettesde la sima P1.3:f);
sur la sima XX, elle occupe au contrairel'espace disponible entre deux palmettes et tient
la place normalementdevolue a la fleur de lotus;118les panneaux de la sima XXI (P1.3:h)"19
constituentdes unites isolees, a la maniere des caissonsrectangulairesdes Propylees(P1.6:a)
ouiles tigelles adventices divergentvers les angles; la sequence attendue de l'anthemion est
rompue. De meme que les sculpteursdes acroteresmedians du Parthenon,le peintre de ces
trois simas a ose un rendu en trois dimensions de la <fleur de lis>>.Il n'y a pas de raison
objective de situer ces simas necessairementdans la dependance de l'Erechtheion,120ni de
penser qu'elles soient posterieuresa 430-420. Des cette epoque, comme plus tard, elles
peuvent anticiperle decor floral des mosaiquesdu IVe siecle.121
II.4. Copiesetpastiches.
Mais un moment vient ouil'unitestylistiquedu decor architecturaltient surtouta la copie
et au pastiche.
114 La sima XVII et la sima du Temple de Poseidon a Sounion ont e tres tot rapprochees:par ex. Buschor
1929, p. 38; Hubner 1973, p. 112. Les calices des fleurs de la sima XVII et des caissons du Parthenon sont
remarquablementproches. La sima de Sounion et les caissonsdu Parthenonsont par ailleurscontemporains,
ca 440 av. J.-C.
115 Sur Buschor 1929, pl. 12, les sepales sont separeset tourmentesA l'exces. E. Buschor propose de dater
cette sima du debut du IV' siecle, ce qui parait trop tard.
116 I1paraitdu reste indeniableque le peintrede la sima XXIII frequentaitles decorateursde vases Al'epoque
des P. de Cadmos et de Cecrops.
117 Buschor 1929, pp. 40-41, fig. 45, pl. 8.
118 Buschor 1929, p. 41, fig. 46, pL.9.
119 Buschor 1929, pp. 41-44, fig. 47, 50 et 51, pl. 10. Vlassopoulou 1989, no 66; Vlassopoulou 1990, no 65.
120 Le relief utilise la troisieme dimension, mais n'a pas le monopole de son rendu. Les fleurs sur tiges
adventicesqui enrichissentplusieursanthemionsde l'Erechtheionpeuvent driver de creationspeintes.
121 Par ex. Salzmann 1982, Sikyon 3, no 118, pl. 20 et 21:3; Pella 12, no 105, pl. 38:4; Vergina 1, no 130,
pl. 40:3.
28 BILLOT
MARIE-FRANQOISE
Elle est evidemment parfaite lorsque, par exception, la morphologie specifique des
6lments, palmetteset lotus, passe d'un supportAl'autre. Ainsi, la sima IX de l'Acropole'22
ne s'interpreteque comme un pastiche de sima en marbre,peinte en sombre sur le ton du
materiau,comme les <<Palmettensimen>> de rampantet d'egout (ci-dessus1.1.4, P1.2:e);la file
de coeurs anticipe sans doute le decor du larmier au fronton du Parthenon123(P1.4:g et
5:g). De meme, plusieurssimas de terre cuite a decor d'oves,124dont celle du petit distyle in
antisdu sanctuaire d'Athena Sounias,125 imitent tres evidemment la sima a oves sculpts
ajour des Propyleesde l'Acropole.'26Et l'on observeencore a la fin du Ve siecle et au debut
du IVe d'autrestransfertsde decors completsdu marbrea la terre cuite, ainsi l'anthemionde
palmettescirconscriteset de <fleursde lis>> au soffitede tuiles d'egout d'Eleusis.127
En sensinverse,les antefixesen marbredes propyleesdu sanctuairede Sounion (ci-dessus
et n. 101) reproduisentles modeles contemporainsen terre cuite, attests par exemple au
Pompeion (ci-dessuset n. 102).
Issusdes memes orignes ceramiques,arrivesau terme de parcoursdistinctsmaisproches,
les repertoires du marbre et de la terre cuite fusionnent et s'usent. Les sculpteurs de
l'Erechtheionviennent d'inventerun style nouveau.
Pour prolonger ces observations,il faut soulignerque le repertoiredes marbrespeints
atheniens a eu la vertu d'aider les ateliers corinthienset corinthianisantsa s'affranchirdu
style severe, non seulement en proposant des compositions A reproduire,ainsi par la sima
d'Olympiedite <du type de l'Hephaisteion>> 128mais en allegeantet en redressantles formes,
notamment en creant cette fleur a trois petales (F 3) qui allait transformer,pour la dur~e
du we siecle, l'apparencede toutesles o<fleurs de lotus>>
(voirP1.5:) : un ou plusieursateliers
corinthiensont su s'en inspirer,qui ont produitdansles annees 425-380 la sima du Ptoion,129
celle du temple d'Apollon a Tegyra,130deux simas ineditesd'Elatee,plusieurssimas d'Argos
(voirP1.7:b)et celle du Temple d'Apollon MaleatasAEpidaure.131
132
L'iconographiesort du cadre de nos investigations.
33 Salzmann 1982, Olynth 4, no 79, pp. 11, 23 et 99-100, p1. 15:2.
134 Salzmann 1982, p. 23.
135 Boardman 1989, fig. 306, 310, 323, 324.
136 Salzmann 1982, Olynth 12, n? 87, pp. 11, 24 et 102, p1. 14:2 et 15:1.
137 Salzmann 1982, Olynth 9, n? 84, pp. 11, 22-23 et 101, p1. 12:3.
138 Salzmann 1982, p1. 15:2.
139 Salzmann 1982, Eretria 3, n? 38, pp. 12, 27 et 91, p1. 27:3 et 4; Ducrey, Metzger et Reber 1993, p. 86,
fig. 45, 96 et 97, pl. I:5, 6.
140 Salzmann 1982, Eretria 2, n? 37, pp. 12, 27 et 90-91, p1. 26; Ducrey, Metzger et Reber 1993, p. 88,
fig. 45, 101 et 102, p1.II:1 et III:1, 2.
141 Par ex. sur la pyxide UniversitatHeidelberg Inv. 69/5: Hampe 1971, n? 80, p. 50.
142 Jacobsthal 1927, p1. 110-117; Boardman 1989, fig. 244, 245, 289, 291, 292, 337, 340 etc.
143 Salzmann 1982, Athen 5, n? 22, pp. 30-31 et 87, p1.42:2-4 et 43:3.
144 Sur cette question en general, Salzmann 1982, pp. 56-57. Toutefois, l'origine des motifs decoratifs
ne permet pas de regler la question fondamentalede la significationde la mosaYque,succedane de tapis ou
30 BILLOT
MARIE-FRANQOISE
d'Eretrie et meme d'Olynthe est pure creation athenienne, a verser au dossier de l'unite
ou de l'ht~rog~n~it6 ornementale de architecture d'Athenesdans le premier tiers du IVe
siecle.
III.2. Mais nos perspectives se diversifient lorsque certaines compositions se reverent
transf~r~esnon plus de la cdramique, mais de l'architectureelle-meme. D. Salzmann a
montre qu'entre autrespavements a panneau circulaireinscrit dans un carre, la mosaique
aux Arimaspes d'Er~trie,avec ses oiseaux vus de dessous, <<reflete>>, au sens figure sinon
propre du terme, une peinture de plafond145 et n'est pas la seule de ce genre ni a Eretrieni
en Gr~ce. Depourvues d'indicesoptiques,d'autrescompositionss'annoncentcependant du
meme ordre;ainsiles deux panneaux carr~sde la mosaiqueau gorgoneiond'Eretrie,a decor
vegetal centre, refl~tent-ilscertainementdes caissonsde plafond146 comparablesA ceux du
Parthenon et des Propylees. Sans epuiser la question, triesvaste,'47 l'exemple d'Eretrie,
justement relatif au repertoirevegetal qui retient ici notre attention, a le merite de mettre
l'accent sur la circulationdirected
des motifs decoratifsd'une partie a l'autre, d'un materiau a
l'autre de architecture d'une meme region ou de regions differentes. Exemple d'autant
plus eloquent que les fleursde l'un des panneaux suscitentun rapprochementimmediat avec
celles de la <<simapale>>d'Olympie, qui echappe tout a fait aux stylespeloponnesiens.148
III.3. Dans le P6loponndse, riche de terres cuites architecturales-a la fois parce qu'il
en produisait beaucoup et que les circonstances ont permis d'en decouvrir un certain
nombre-, riche aussi de mosaiques, et singulidrementdans le Nord-Est, plusieurs cas
s'imposentde comparaisonstriesprecisesentre les deux productions.
III.3. 1. Surl'une des mosaiquesles plus anciennesde Sicyone,la mosaiqueau gorgoneion
(P1. 7:c et 8:c),149 les lotus de la bordure circulaire du tapis suscitent des reminiscences
immediates150 : une sima de Delphes (P1.7:a) et une autre, a anthemion identique, de
Nem~e151 pour la forme gen~rale de la fleur et du petale central, et pour les s~pales tres
minces et remontants. L'une et l'autre ont ete fabriqueesa Corinthe ou par des artisans
corinthiens. La sima de Delphes a W datee de 460-450, la mosaique ca 400 ou du premier
quart du IVe siecle, independammentl'une de l'autre et sur appreciation stylistique. La
duree d'un motif peut expliquer l'6cart chronologique; la vraisemblance inviterait A le
reduire, d'autant plus que les palmettes de la mosaique n'affichentaucune caracteristique
pavement polychromepercu comme tel: la natureet l'itinerancedes decors sont independantesde la fonction
des objets ou succedanesd'objetsqu'ilsviennent orner.
145 Salzmann 1982, p. 57, Eretria 2, n? 37, pl. 26; Ducrey, Metzger et Reber 1993, p. 88, fig. 45, 101 et
102, pl. II:1 et III:1-5.
146 Salzmann 1982, Eretria 1, n' 36, pp. 12, 27, et 90, pl. 27:2; Ducrey,Metzger et Reber 1993, pp. 85-86,
fig. 44 et 95, pl. I:1-4.
147 Guimier-Sorbetset Barbet 1994; Salzmann 1982, p. 58.
148 OlForsch V, pp. 114, 124 et 131, fig. 38, pl. 41 :c.
149 Votsis 1976, pp.575-581, fig. 1-7. Salzmann 1982, Sikyon 4, n? 119,pp. 11,22 et 112, pl. 10:1 et 2; 11:1
et 4.
150 En particulierP1.8:c au fond Adroite.
151 Le Roy 1967, toit 59, pp. 134-135, pl. 49-50. Nem~e, AT 55 et AT 75, 2: Miller 1978, pp. 83-84,
fig. 4 et pl. 26:c et d; Miller 1979, pp. 89-90, pl. 32:a.
TERRES CUITES ARCHITECTURALES, PEINTURES ET MOSAIQQUES 31
du IVe siecle. Nous retrouvonsici tous les delicatsproblems d'estimationque posent l'assise
d~cor~e du Tr~sor dorique de Marmaria et le linteau du Tr~sor des Ath~niens, au sein
d'un corpus pourtant plus nombreux. Ici, les trois documents ne se rapprochentque par
un seul element, la fleur integree dans deux schemas d'anthemionsdifferent, a tiges en S
jointives sur les deux simas, A arceaux sur la mosaique. De surcroit,la mise en oeuvre de
l'anthemion est triesdifferent sur les deux simas, la moulurationde celle de Nemee etant
exceptionnellement complexe. On peut affirmer seulement que la fleur a W creee puis
copiee deux fois, ou copiee trois fois, dans un ordre triesincertain.
III.3.2. La mosaique aux tritonsdu pronaos du Temple de Zeus a Olympie (P1.7:d)'52
a e datee du milieu ou de la seconde moitie du IIIe siecle av.J.-C. par les monnaies trouv~es
dans le support. La restitutiondessinee par A. Blouet pour l'Expeditionscientifique
deMoreeI,
pl. 63-64,153 a une epoque ou la mosaique etait moins delabree, et les vestiges de bordures
des panneaux laissent envisager,au moins pour les bordures, une epoque plus ancienne,
soutenue par une sima d'Argos(P1.7:b)datee vers 400 av.J.-C.
III.3.3. La compositionvegetablequi occupe les e&ointonsde la mosaique aux centaures
de Sicyone154 (P1.7:g)utilisedeux petitsfleuronsqui figurentaussisurles antefixesdu Temple
d'Apollon Maleatas (Fig. 1),155 vers 390-380. Sur mosaique et antefixes,la meme palmette
s'elance, etroite aux feuilles raides. On a propose pour la mosaique le deuxieme tiers du
IVe siecle, sans argumentprecis.
III.3.4. La fouilledu <<Building
IV>>sous la Stoa Sud de Corinthea reveled
un fragmentde
mosaique (P1.8:d)du IV' siecle, necessairementanterieurea 300 av.J.-C., mais sans terminus
post quem.156Nous y retrouvonsle meme petit fleuron, mais allege, tel qu'il se rencontre
au soffitedes simas d'egout du batiment de scene du Theatre d'Argosdans le premier quart
du IIIe siecle (P1.8:e et Fig. 2).
III.3.5. La mosaique circulaire trouvee dans le Quartier du Theatre de CorintheI57
recense presque tout le repertoiredes simas (voir P1. 5:f et 7:e, f) et des antefixes de 370
environ Ala fin du siecle, dont la grande et la petite sima aux acanthes d'Olympie (voir ici
meme la communicationdeJ. Heiden) et toutesles antefixesdu type de la Stoa Sud158ou qui
lui sont apparentees,159et dont la productioncommence au milieu du siecle. Cet anth~mion
inscrit dans un tapis circulaireresume,jusqu'aux enjolivures,la morphologie et la syntaxe
152
Salzmann 1982, Olympia 2, no 138, pp. 63 et 117-118, pl. 71:5 et 6, 72 et 102:4.
153
Salzmann 1982, pl. 72:1.
154
Votsis 1976, pp. 582-583, fig. 8; Salzmann 1982, Sikyon 2, no 117, pp. 11, 25-26, 111-112, pl. 22:1.
155 Lambrinoudakis1978, p. 120, pl. 96:8; la serie 69 de Delphes, egalement fabriquee A Corinthe, est
identique: Le Roy 1967, p. 140, pl. 53 et 127 (restitutionerronee).
156 Williams 1980, pp. 114-116, pl. 18:b et 19:a. Salzmann 1982, p. 128. On prendra garde cependant
que la disparitiond'un ou deux galets A mi-hauteur du petale central ne doit pas faire croire A existence
d'un sepale, comme sur les simas d'Argos.
157 Williams et Fisher 1976, p. 114, pl. 24; Williams et Zervos 1983, pp. 18-19, pl. 6:c; Salzmann 1982,
Korinth 2, n? 64, pp. 26 et 95, pl. 23.
158
CorinthI, iv, pp. 86-87, pl. 20:1 et 21:1.
159
Epidaure(dont une serie plus ancienne),Perachora,Dodone etc.
32 MARIE-FRAN4OISE BILLOT
174
0 1 5cm
0 1 5cm.
BIBLIOGRAPHIE
AgoraXIV = H. A. Thompson et R. E. Wycherley,TlheAgora XI), Princeton 1972
ofAthens(Agora
Alexandri,0. 1968. <<XpovLx& tr3v &vaaxapcOv.rF 'Epopeta KXaaOLxxav 'ApXaLoTlTCV AeXt
'AOvv&5v>>,
23, 1968, BW1[1969], pp. 39-42
Amandry,P. 1984. <<Notesde topographicet d'architecturedelphiques>>, BCH 108, pp. 177-198
ApXala Maxbovwla-Ancient Macedonia= ApXata Maxbovwca-AncientMacedonia(Cat. d'exposition tenueau
Museumof Victoria, Melbourne!QjeenslandMuseum,Brisbane/Australian
Museum,Sydney1989), Athenes 1988
Arias, P., et M. Hirmer. 1957. Spina,Munich
. 1960. Tausendjahregriechische
Vasenkunst,
Munich
Audiat, J. 1933. LseTrsordesAthniens(FouillesdeDelphes,II, LeSanctuaire
d'Apollon),
Paris
Bankel, H. 1986. HallervonHallerstein in Griechenland,
1810-1817, Architekt-Zeichner-Bauforscher,
H. Bankel, ed.,
Berlin
. 1990. <The Athenian Treasury as Dated by Its Architecture>,dans Aktendes XIII. Internationalen
Kongressesfir klassische Berlin1988, Mainz, pp. 410-412
Archdiologie,
- 1993. Derspdtarchaische TempelderAphaia 19), Berlin/New York
aufAegina(DenknilerantikerArchitektur
Beazley,J. D. 1963. AtticVasePaintingin theMuseumofFineArts,BostonIII, Boston
- 1974. DerPanMaler,3eme ed., Mayence
163 Salzmann 1982, Sikyon 3, no 118, pp. 11, 18, 25-26 et 112, p1. 20 et 21, feuilles dentees et feuilles
ocreneleesosur p1. 21:5; comparer les 4 feuilles lobees a contour continu (p1.20 et 21:1), semblables a celles
de la sima, aux 4 feuilleslobees Acontour ocr~nelb>des angles (p1.20 et 21:4).
34 BILLOT
MARIE-FRANQOISE
Billot, M.-F. 1976. oTerres cuites architecturalesdu Mus&eEpigraphique?, A&ex 31, 1976, A' [1980],
pp. 87-135
. 1991. <<Terrescuites architecturalesd'Argolide>>, RA,pp. 199-209
Blouet, A., A. Raviosie, et A. Poirot. 1831. Expedition deMorkeI, Paris
. 1838. EvplditiondeMorieIII, Paris
Boardman,J. 1975. Athenian RedFigureVases.TheArchaicPeriod,Londres
. 1989. Ahnian RedFgureVases.TheClassicalPeriod,Londres
Boersma, S.J. 1970. AtenianBuildingPolsfrom 561/60 to 405/4, Groningue
Bohn, R. 1882. Die PropylhnderAkropols zuAthen,Berlin
Broendsted,P. 0. 1830. Voyages dansla Greceaccompagnis derecherches
archdologiques II, ed. fransaiseFirmin Didot,
Paris
Brouskari,M. 1989. <<Ausdem Giebelschmuckdes Athena-NikeTempels>>, dans Beitrdge zur Ikonographie und
Hermeneutik, Nikolaus
Festschrfiflir Hinmelmann, H. U. Cain, H. Gabelmann, et D. Salzmann, ed., Mayence,
pp. 115-118
Bruneau,Ph., et J. Ducat. 1983. GuidedeDilos, 3em ed., Ath~nes
Buhlmann,J. 1872. DieArchitektur desklassischen undderRenaissance,
Alterhtms Stuttgart
Busing, H. H. 1969. <Vermutungenuber die Akropolisin Athen?, MarbWPr,pp. 1-30
.1979. <<EinAnthemion in Delphi?, dans Studiesin ClassicalArtandArchaeology. A Tributeto P. Heinrich
vonBlanckehagen, G. Kopcke et M. B. Moore, ed., LocustValley,N.Y., pp. 29-36
.1988. <<Athener Schatzhausund Parthenon?,dansIpaxxem& TOU XII Al 0CVO51uve8ptou KXaalXw5
ApXtLoXoytaq,AOvat, 4-10 1cntel3pppou1983, IV, Athenes, pp. 29-32
. 1990. <<Zur Bemalungdes Nike-Tempels?,AA [Jdl 105], pp. 71-76
Burn, L. 1991. <<ADinoid Volute-kraterby the MeleagerPainter:An Attic Vase in the South ItalianManner?,
GreekVasesin theJ.PaulGett Museum5, Malibu, pp. 107-163
Buschor,E. 1929. Die TondiichffrderAkropolis, I, Sinen,Berlin/Leipzig
- 1933. Die TondcherderAkropolis,II, Stirnziegel,Berlin/Leipzig
. 1933a. <<Altsamische Grabstelen>>,AM 58, pp. 22-46
Cambitoglou,A. 1978a. <<Topdv>>,'Epyov 1978 [1979], pp. 26-30
. 1978b. <<'AvacxxacqTopdvis?, IIpaxxLx&1978 [1980], pp. 80-92
Camp, J. M. 1986. TheAhnian Agora.Excavatonsin theHeartofAthens,Londres
. 1990. TheAthenian Agora.A GuidetotheExcavations andMuseum,4emeed., Athenes
Carpenter,R. 1970. Die Erbauer desParthenon, Munich
Caskey,J. D., et J. D. Beazley. 1954. AtticVasePaintingin theMuseumofFineArts,BostonII, Oxford/Boston
Cavadias, P., et G. Kawerau. 1906-1907. Die Ausgrabung derAkropoisvomJahre 1885 bis zumjahre 1890,
Berlin
Charbonneaux,J., R. Martin, et F. Villard. 1968. Grace archafque,
Ath~nes
Childs, W. A. P. 1985. ((In Defense of an Early Date for the Frieze of the Temple on the Ilissos?,AM 100,
pp. 207-251
Cockerell, C. R. 1860. The Temples ofJupiterPanhellenius at Aeginaandof ApolloEpicuriusat BassaenearPhigalia
inArcadia,Londres
Collignon, M. 1912. LeParthnon,Paris
Corinth I, iv = 0. Broneer, TheSouthStoaandItsRomanSuccessors (CorinthI, iv), Princeton 1954
Courby,F. 1927. la terrasse dutemple(FouilesdeDelphes,II, Lesanctuaire d'Apollon), Paris
Daux, G. 1923. Lesdeuxtrisors(Fouilles deDelphes,II, Lesanctuaired'Atna Pronaia),Paris
Daux, G., et E. Hansen. 1987. Ie trisordeSiphnos(FouilksdeDelphes,II, Lesanctuaire d'Apollon),Paris
De La Coste-Messeli~re,P. 1942-1943. <<Geison et sima au temple des Alcm~onides>>, BCH 66-67, pp. 66-67
. 1953. <<Troisnotules delphiques. (.. .) Tresor des Atheniens: les palmettes?,BCH 77, pp. 179-182
De La Coste-Messeliere,P., et G. De Mire. 1957. Delphes,Paris
De Laborde, L. 1848. It Parthnon.Documents pourserzira unerestauradon,Paris
PEINTURESET MOSAIQUES
TERRESCUITESARCHITECTURALES, 35
Delivorrias, A. 1969. <Poseidontempelauf Kap Sunion. Neue Fragmente der Friesdekoration>,AM 84,
pp. 127-142
. 1974. AttischeGiebelskuipturen undAkrotere desftnflen Jahrhunderts (TlbingerStudienzur Archdologie und
Kunsigeschichte 1), Tubingen
Despinis, G. 1974. oT& yXutx&'c&v&c'pcat&-cav 'toGvaou 'riq 'AOivdq Nixrjo, AeX'r29, 1974, A' [1977],
pp. 8-24
Dinsmoor,W. B. 1940. <The Temple of Ares at Athens>>, Hesperia 9, pp. 1-52
* 1941. ObservationsontheHephaisteion (HesperiaSupplement5), Princeton
- 1946. <The Athenian Treasuryas Dated by Its Ornament>,AJA50, pp. 86-121
Dinsmoor,W. B., Jr. s.d. Sounion (Keramos Guides),
Athenes
- 1974a. vNew Fragmentsof the Parthenonin the AthenianAgorao, Hesperia43, pp. 132-155
- 1974b. <The Temple of Poseidon: A Missing Sima and Other Matterso,AJA 78, pp. 212-238
1976. <The Roof of the Hephaisteion>,AJA80, pp. 223-246
. 1982. <(AnchoringTwo Floating Templeso,Hesperia 51, pp. 410-452
Dorpfeld, W. 1881. oQber die Verwendungvon Terrakottenam Geison und Dach griechischerBauwerke>>,
BWPr41, Berlin
Drerup, H. 1981. <Parthenonund Vorparthenon.Zum Stand der Kontroverseo,AntK24, pp. 21-38
Ducrey, P., T. R. Metzger, et K. Reber. 1993. Le Quartier de la Maisonauxmosaiques (Eretria,fouilleset recherches
VIII), Lausanne
Ehrhardt,W. 1989. <Der Torso Wien I 328 und der Westgiebeldes Athena-Nike-Tempelsauf der Akropolis
in Athen>, dans Beitrdgezur Ikonographie undHermeneutik, FestschfftflirJNikolausHimmelmann, H. U. Cain,
H. Gabelmann et D. Salzmann, ed., Mayence, pp. 119-127
Epifanio, E. 1982. <Un cratereimerese dell'officinadel Pittoredei Niobidio, dans Aparchai, Nuovericerche e studi
sullaMagnaGreciae la Siciliaantcain onoredi P E. Arias,Pisa, pp. 347-356
Euphronios = Euphronios, peintre AtkAnes au VI' sick av.I.-C. ExpositonMuse du Louvre,Paris, 18 septembre-31
dicembre 1990, Paris 1990
Fenger,L. 1886. Die dorische Poychromie, Berlin
Furtwangler,A. 1906. Aegina.Das Heilzgtum derAphaia I, Munich
Ginouves, R., et R. Martin. 1985. Dictionnaire mIthodique de architecture
grecqueet romaine, I, Matkiaux,techniques
deconstruction,techniquesetformesdudecor,Athenes/Rome
Guimier-Sorbets,A.-M. 1990. Lesbasesdedonniesenarckdologie. Conceptionet miseenoeuvre, Paris
Guimier-Sorbets,A.-M., et A. Barbet. 1994. <Le motif de caisson dans la mosaique, du IVe siecle av.J.-C. A
la fin de la Republiqueromaine: ses rapportsavec l'architecture,le stuc et la peintureo,dans IVInternationales
Mosaikolloquium, Tner,8-12 August1984, Paris, Aparaitre
Hampe, R., ed. 1971. CatalogderSammiung antikerKiinkunstdesarchdologischen InstitutsderUniversitdt Heidelberg,
II, Neuerwerbungen 1958-1970, Mayence
Harrison,E. B. 1976. Recension de Delivorrias 1974, dans AJA80, pp. 209-210
. 1979. <The ArchitecturalSculpture of the So-called Theseumnv,dans Greece andItly in theClassical
World, Actaof theXI InternationalCongressof Classical
Archaeology, London,3-9 September 1978, Londres,p. 220
. 1981. Recension de AntP 18, 1979, dans AJA85, pp. 232-234
Heiden,J. 1987. Korinthische Dachziegel.ZurEnticklungderkorinthischen Ddcher,Frankfort/Berne/New York/Paris
Herrmann, K. 1974. <Die Giebelrekonstruktiondes Schatzhausesvon Megara>,AM 89, pp. 75-83
Hesperia59 = Proceedings of theFirstInternationalConferenceonArchaicGreek Architectural Athens,December
Terracottas,
2-4, 1988 (Hesperia 59, 1990), N. A. Winter,(d.
Hubner, G. 1973. <Dachterrakottenaus dem Kerameikosvon Athen>,AM 88, pp. 67-143
Iliakis, K. 1976. <H ZxTpampLxK apXrext-xcoX btaxoaV tou vao6 tv5 Niaeaq a-co Paqvo6vtao>,
AeXt 31, 1976, A' [1980], pp. 244-259
.1987. oH a-cyt tou v=o6 tv5 N4esav a-co Paoqvoiv-a j ?itaan16netpaavanv pa&patcaa tv>>,
Avaa-cXa-Euv-4parIIpooacaa Mv .tiLcavxat Euv6Xcv 2, pp. 11-29
36 MARIE-FRAN9OISE BILLOT
Credit d'illustrations
PLATE 1:a. Cl. M.-E B. d'apres BCH 103, 1979, p. 590, fig. 149 (reproduitavec la permission de Torone
Excavations)b. Cl. EFA;c. Cl. M.-F. B. d'apresEuphronios, no 5, p. 90; d. Cl. EFA, Ph. Collet; e. Cl. ENSBA
d'apres Cockerell 1860, pl. XIII, fig. 3 et 4; f. Cl. M.-F. B. d'apres Bankel 1986, p. 130, fig. 4:10 (reproduit
avec la permissionde 1'auteur).
PLATE2:a. Cl. M.-F. B. d'apresAudiat 1933, pl. XXII; b. Cl. EFA, Ph. Collet;c. Cl. M.-F.B. d'apresEuphronios,
n? 12, p. 114 (reproduitavec la permissionde MetropolitanMuseum of Art);d. Cl. DAI Athen, AKR 1959;
e. Cl. DAI Athen, AKR 1968; f. Cl. M.-E B. d'apresTravlos 1971, fig. 331.
PLATE 3:a, b, c. Cl. PremiereEphorie, Acropole; d. Cl. DAI Athen, AKR 951; e. Cl. DAI Athen, AKR 936;
f. Cl. Troisieme Ephorie no 1151; g. Cl. PremiereEphorie, Acropole;h. Cl. DAI Athen, AKR 938 i. Cl. DAI
Athen, AKR 942.
PLATE 4:a, b. Cl. Agora Excavations;c. Cl. ENSBA; d. Cl. ENSBA, d'apres Stuart et Revett 1762, chap. II,
p. 10, pl. VIII fig. 3; e. Cl. M.-F.B. d'apresAriaset Hirmer 1960, fig. 173;f. Cl. ENSBA;g. Cl. EFA,Ph. Collet,
d'apresOrlandos 1978, pl. 74.
PLATE5:a. Cl. Agora Excavations;b. Cl. EFA; c. Cl. ENSBA; d. Cl. EFA d'apres Koch 1955, fig. 83; e. Cl.
M.-F. B. d'apresOrlandos 1953-1954; f. Cl. EFA, Ph. Collet;g. Cl. ENSBA.
PLATE 6:a. Cl. EFA, Ph. Collet d'apresPenrose 1851, pl. XXV; b. Cl. ENSBA; c, d. Agora Excavationsd'apres
Dinsmoor,Jr. 1974a, pl. 24 et fig. 5.
PLATE 7:a, b. Cl. EFA, Ph. Collet; c. C1. Quatrieme Ephorie, Nauplie; d. Cl. D. Salzmann; e. Cl. Corinth
Excavations;f. Cl. EFA;g. Cl. DAI Athen, 78/429.
PLATE 8:a. Cl. DAI Athen, 01. 5393; b. C1.D. Salzmann;c. Cl. Quatrieme Ephorie, Nauplie; d. Cl. Corinth
Excavations;e. Cl. EFA.
FIGURE1: EFA, Planotheque,n? 8442.
FIGuRE 2: EFA, Planotheque,n? 5318.
MARIE-FRANyoISEBILLOT
CNRS-IRAA
Adressepersonnelle:
38, rue Lacepede
75005 Paris,France
PLATE 1
-; x
R. Cockerell)
(C.,
M- EF os BILT TEE cur, ARHTCUAE,; PENTRE ET MOSAIQUE
PLATE 2
;U
<4 , !
,~~~~~~~~~~~~~~~~~~~~S-
r *_
f. AtlheTneores,
H idon,simnaos
~atomp co
de rapntdue de 1DAcroeol3948de tenescla
T(P. cu.~
M'iRaiE-RNQIEBILT ARCITCTRAES deT
PEIITRE MOSicanitEs(.Tunie
I4
71~~~~
Cd~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~U
4~~~~~~~~~~~~~~~~~~~-
Cd~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.
CA,
W~~~~~~~~~~~~~o
00
I~~~~~~~~~~~~~~~l
Jr~~~~~~d dC
j ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~e
PLATE 4
f C- _ 8 ~ ~ ~~ P I
t ~ ~ ~ == ~~~- -- = --Il
2.35
P-rheon
c. Athenes, Parthenon, decor de regula(A. Paccard) @
. ,
( i
A'-
.tA f
V\
BILLOT:TERRESCUITES
MARLE-FRANQOISE ARCHITECTURALES,PEINTURES ET MOSAIQUES
PLATE 5
.t
N41 ~ ~ ~ ~ ~ ~ 4
I..,-,,,,,
A!#
F!--, bi
g. Athenes,Parth6non,d6corde la mouluredu
larmierrampant(A. Paccard)
f. Heraion d'Argos,sima Mus&eNational SA 277 p (.Pcad
BILOT: TmuuESCUITES
MARIE-FRANCOISE ARcHrrEcTURALs,
PEINTURES
ET MOSAIQUJES
PLATE 6
1 *
.-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 4!7T.-"
A ~ ~ ~ ~ ~ A
11#41
a. Delphes, smoaS.44(A Tougrgnei eaile
AM.FRAN9OSE BILLOT TERRES CrES
d. Olrpe
ARC~l-ECTURA=S
eml
PEINTURS
eZus oaqe epoas
dealdu nlet
~~~~~ "4~~~~~~~~4
0-~~ 0 ~
* i; ** @; r
;fikiA
'-p4'~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~:
MD
d'Epidaure,ssm~Ag-
f. Musee as
au gogon~iondetailed
c. Sicyoe, mosa~ue TepedaZumsaqedpoas
Oslympe,
g.~~~dti 'u nl
PLATE 8
IV>>,mosaique
d. Comthe, <<Building e. Argos, The~tre spmad gout 73/31.1
BILLOT:TERRESCUITESARCHImTECURALES,
MARIE-FRANCOISE ET MOSAIQIJES
PEINTURES
ARCHITECTURALTERRACOTTASFROMCLASSICAL
CORINTH
AND HELLENISTIC
(PLAMS 9-19)
IV, i, FA 2, pp. 12-13, fig. 5:c, p. 47; also a seriesof unpublishedantefixes,FA 125-FA 140, FA 142,
l Corinth
with the exception of FA 129, FA 135, and FA 136, which are a little larger. Other similar pieces are also
noted by Ida Thallon-Hill.
40 MARY C. ROEBUCK
Temple Hill. The regular leaves of the palmette of the antefix forming a nearly circular
arc are clearly an extension of the shape used in the Late Archaic period. The lower part
below the palmette is more elongated than before and less stiff. An extra leaf has now been
added on either side between the spiraland the lotus calyx. On the face of the eaves tile is a
meander. The narrowspace between the blackborderof the center squareand the meander
border is characteristicof the 5th century. The antefixesresemble those from Roof 56, the
Lesche of the Cnidians at Delphi, dated 475-460 B.C.,2 and those of the Classical temple
at Kalapodi.3 The Corinthtiles are perhapsa littlelaterthan those from Delphi but stilldate
to the firsthalf of the 5th century.
Another antefix fragment4(FA 417, PI. 9:b) of equally good quality,which preserves
most of the palmetto, is probably somewhat later. Its leaves are more attenuated and its
spade-shapedcenter longer and narrower. This piece was built into the wall of the later
kiln at the Tile Works.
Two sets of tiles from rakingsimas also belong to the second quarteror middle of the 5th
century. Some of these were publishedby Hill. Additionalpieces were later discoveredin the
area to the north of Temple Hill. These simas are on a smallerscale than the combination
tiles already discussed and cannot be from the same building. Nonetheless, the meander
preservedon the bottom fascia of one set5 (FS 5, FS 6 [P1.9:c], FS 425) is very similar to
that used on the combinationtiles. The uppermolding of these sima pieces is decoratedwith
a bead and reel above a Lesbianleaf. The cymatiumhas a typical alternatingeleven-leafed
palmette and three-petaledlotus. The lotus and palmettes are connected below by tendrils
ending in down-curvingspiralsunder the palmettes and looping up under the calyx of the
lotus. This is a pattern found throughoutthe Classicalperiod. The outer petals of the lotus
flower curve over the top of the palmette as in Roofs 55 and 56 at Delphi, dated about
480 B.C. and 475-460 B.C. respectively.6This form of lotus is also found in the sima from
Ptoion dated toward 450 B.C. or mid-5th century.7
The second set of tiles8 (FS 9, FS 428 [PI. 9:d]) resemblesthe first except for the lotus
whose outer petals do not extend over the palmette. In this characteristicthis set recalls
those tiles from Roof 58 at Delphi, dated 460-450 B.C.9 Otherwisethe two sets are virtually
identical. This is also true of their profiles (Fig. 1, profile nos. 1 and 2). Both still have a
fairly strong curve and project out some distance at the top. The cyma reversa is still in
the processof development. These tiles shouldbe dated about the middle of the 5th century.
The Delphi roofs they resemble are also of Corinthianworkmanship.
2 Le Roy 1967, pp. 128-132, 137-139; A 186, p. 130, pl. 48:17; p. 139 for the date.
3 Hubner 1990, p. 167, pl. 15:a.
Unpublished.
4
5
For FS 5 and FS 6, see Corntlh 1, i, p. 66 and pl. V; Heiden 1987, pp. 117-118; also Van Buren 1926,
p. 97, no. 104, and Shoe 1936, I, p. 78, II, pl. 33:9. The date of the end of the 5th century given by Van
Buren and by Shoe is too late. FS 425, found in the area north of Temple Hill, is unpublished.
6 Le Roy 1967, pp. 127-128, 139, pis. 45, 103 for Roof55; pp. 128-132, 139, pls. 46, 103, 110 for Roof 56.
7 Le Roy 1967, p. 145 and pl. 56; Heiden 1987, p. 118.
IV, i, p. 67; FS 428, found to the north of Temple Hill, is unpublished.
8 For FS 9, see Corinth
9 LeRoy 1967, S44, pp. 133-134, 139, pl. 48.
TERRACOTTASFROMCLASSICAL
ARCHITECTURAL AND HELLENISTIC
CORINTH 41
l
L~
1 2 3 4 5 6 7 8 9
0 4 S 6 7 8 9 40
? ? l , 1- ' ' ' ' l CMS.
10
FIG. 1. (1) FS 6; (2) FS 9, FS 428; (3) FS 900; (4) FS 885; (5) FS 876; (6) FS 865; (7) FS 860; (8) FS 867;
(9) FS 875; (10) FS 877
The Tile Works at Corinth has provided us with a wealth of material, but we are
concerned here only with the roof tiles. The greaterpart of the architecturalterracottasdate
to the period around 430-420 B.C. There were two kilns. The earlier dates to the Archaic
period and on into the 5th century. A cistern (formerlydesignatedWell C) appears to have
been filled when this kiln went out of use. The materialin it dates from the late 6th century
and first half of the 5th century,with some pieces as late as 460-420 B.C.10 The second
kiln was largely built of tile pieces of various sorts. Some forty-six pieces, mostly simas,
from the wall of the kiln were inventoried. Other tiles were also found in fill in the area,
some of which evidentlywashed in when the Tile Workswent out of use. Presumablythe
pieces built into the wall of this kilnwere left lying aroundwhen the firstkilnwas abandoned.
There is no evidence of a third kiln that might have produced other material. Probablythe
second kiln was built soon after the first ceased to be used. While there is an occasional
Archaicpiece in the wall, most of the materialis generallyhomogeneous, from about the end
of the thirdquarterof the 5th century,providinga good sourceof informationfor thisperiod.
10
Merker 1988, p. 193 and note 3, p. 198 and note 10; Weinberg 1954, p. 130, fig. 2:c and note 132 and
pp. 132-133; Weinberg 1957, pp. 290, 292 and note 22; Corinth
VII, iii, p. 205, deposit 26 (WellC Tile Works).
42 MARYC. ROEBUCK
One tile11 (FA 415, PI. 9:e) built into the wall of the second kiln very possibly was
intended to be used as an akroterion. Only one side is painted and the surface is flat, not
molded. The scale is rather large. It is unlikelythat it was meant to be an antefix. The
design consistsof a palmettewith a diamond-shapedheart,below which are the upper parts
of two large spiralsheld together by a darkred band. The leaves of the palmette are quite
straightand stiff,except for the lowest ones, which curve somewhatto conform to the line of
the spirals.The decorationresemblesthat of the ridgepalmettesof the Archaicperiod. This
piece may stillbelong to the firsthalf of the 5th century.
Two antefixesgive an idea of the variationsfound in the second half of the 5th century.
One12 (FA 425, PI. 9:f) comes from the Tile Works cistern. It should not, therefore, be
later in date than 420 B.C. Its design differs from that of the earlier antefixes of the first
half of the 5th century in several respects. The lowest leaves project out as far as those
above. It is this that results in the altered profile. The design of the lower part of the
antefix under the palmette also is unlike that used earlier. The lotus of the earlier pieces
is replaced by down-curvingspirals. Their tendrilsfollow the line of the sides, then curve
upwardonce more to form double spiralsalong the edge of the tile. Below the center spirals
is a seven-leavedpalmette suspendedfrom a diamond-shapedheart.
A second antefix13(FA200, PI. 10:a)from the Potters'Quartershows the same delicacy
of treatmentfound in FA 425. At the top is the usual palmette. The lower part, however,
representsanother variation. The central pattern is the same as before, but the tendrils
end in single rather than double spirals. In the center below the spiralsis a lotus flower,
but unlike that of the earlier antefixesthe calyx of the lotus is curved rather than angular,
giving quite a differentappearance. There are eyes in the spiralsand half palmettesbetween
the tendrils and the lotus. Both FA 425 and FA 200 show some affinity to the antefixes
from Delphi of Series 63 and 69, the latter dated between 450 and 420 B.C. 14
Many rakingsima pieces were built into the Tile Workskiln wall and thus are discards.
The quality of the painting varies, being very good in some, quite sketchy in others. One
excellent piece, from a Tile Workswell ratherthan the kiln, although only a small fragment
of the upper part of the sima15(FS 885, PI. 10:b),might be said to representthe next stage
in sima development after FS 9 and FS 428 (P1.9:d). The moldings at the top, decorated
with a bead and reel and a Lesbian leaf below, while not identical to the earlierpieces, are
very similar. The lotus is of the same type, but the outer petals flare less. The palmette,
however, is quite different. Its leaves are fewer and more widely spaced, resemblingsome
of the Delphi pieces. The profile(Fig. 1, profileno. 4), as far as it is preserved,has a definite
curve, although this is not as pronounced as in FS 9 and FS 428. FS 885 appears to be
intermediatebetween them and the next group of simas.
The workmen who painted the raking simas seem to have relieved the monotony of
the typical design at this time by minor variations. A few examples will serve to illustrate
l FA 415 is unpublished.
12 FA 425 is unpublished.
XV, ii, pp. 272, 281, pl. 58, no. 63.
13 Corinth
14 Le Roy 1967, series 63: p. 136, A 57 and A 58, pl. 51; series 69: pp. 141, 148, A 95, A 109, and A 110,
pl.53.
15 FS 885, unpublished,comes from a 4th-centurywell in the Tile Works.
TERRACOTTASFROMCLASSICAL
ARCHITECTURAL CORINTH 43
AND HELLENISTIC
this effort. Certain elements remain fairlyconstant, such as the upper molding or moldings
with egg and dart or bead and reel or both and the meanderon the fasciaat the bottom. The
palmettes,too, are fairlystandard,eleven leaves with tendrilsbelow ending in down-curving
spirals below them and a loop under the lotus. There is an occasional difference in the
treatmentof the palmette center. The variationscome mainly in the form of the lotus. In
FS 90016(PI. 0O:c) the calyx has a teardrop-shapedcenterflankedon either side by a pointed
leaf curving downward. Above these latter are tall rectangularboxes, probably intended
as leaves. The floweris made up of threepetals. FS 87617(P1.IO:d)is basicallythe same, but
instead of straight-sidedrectanglesit has curved leaves of white lined in darkred. FS 86518
(PI. IO:e)varies this same design by giving the lotus flower five petals, the lower two of
which are thin and narrowand curve outward. The upper side petals also spreadout more.
FS 86019(P1.11:a)has a differentversion. Here the space in the loop of the tendrilsunderthe
lotus is filled with a teardrop or dagger. The two side leaves of the calyx are smaller and
more widely spaced but still lined with red. The lower two petals, also more widely spaced,
curve out and down. The heart of the palmette, unlike the others, is reserved and has a
small dart. The palmette leaves of all these vary somewhat from tile to tile, FS 900 and
FS 865 having more of a tendency toward a double curve than the others. As far as the
profiles(Fig. 1)are concerned, FS 900, which is not from the Tile Works,has a strongercurve
with greaterprojectionthan the others. It is much flatter,however,than those of the middle
of the 5th century. The others, all from the Tile Works, one (FS 865) from the kiln wall,
are similarand resemblethe profileof Mallwitz's"Blutensima"from the PheidiasWorkshop,
dated ca. 430 B.C.20 For palmettes with the slight double curve, compare those from the
Temple of Ares.21
A differenttreatmentof the cymatiumoccurs in some other pieces of raking sima, also
built into the wall of the Tile Works kiln. In these the lotus calyx is made up of three
tulip-shapedleaves. In one example22(FS 867, P1. 1 :b) the lotus flower has five petals.
They are not unlike those of FS 860, though less widely spaced. Each spiral under the
palmette has an additionaltendril at the side. The palmette leaves are thickerand rounder
at the ends than those in the previousgroup of simas.
A much larger sima23(FS 877, PI. 11:c) again differsin design. The calyx is made up
of the same three tulip-shapedleaves and is carefullyoutlined. There are three petals in
6 FS 900, unpublished,is from the Asklepieionarea.
17 FS 876, unpublished,comes from the firstaccumulationof fill that washed in afterthe Tile Workswent out
of use.
18 FS 865, unpublished,was built into the wall of the second kiln of the Tile Works.
19 FS 860, unpublished,came from a Byzantinewell in the Tile Works.
20 OlForsch V, pp. 112-113, 121, 124-125, 131-132, pls. 15, 38, 39, 40: 1; Heiden 1987, pp. 109, 119, 123,
profile 7:5; Shoe 1936, I, pp. 78-79; II, pl. 33:5; Mallwitz 1980, no. 41, p. 150, pl. 102:4.
21 Dinsmoor 1940, p. 33, fig. 12 and p. 51 for date of erection, between 440 and 436 B.C.; Hubner 1973,
p. 117, fig. 13.
22 FS 867 was built into the wall of the second kiln of the Tile Works. It is unpublished. For others with
IV, i, pp. 25, 78, fig. 25, for FS 128, and pp. 78-79 for FS 129; Roof 80 at
a tripartitelotus calyx, see Corinth
Delphi (Le Roy 1967, pp. 154, 168, S 48 and 49, pls. 59:1, 2 and 104) probably should be put here rather
than with the late 4th-centuryroofs of the South Stoa type.
23 FS 877, unpublished,was built into the wall of the second kiln at the Tile Works.
44 MARYC. ROEBUCK
the flower,very like the upper three of FS 867. Beneath the lotus calyx is a small triangle.
Because of the height of the sima, both the lotus and the palmette are elongated and the
spade-shapedpalmettocenter is very tall. The palmetteleaves show a slighttendency toward
a double curve. Running under the lotus and palmette band are S-shaped spiralsthat turn
upward under the lotus, downwardunder the palmette. The palmettes resemble those on
the Hephaisteion-typesima at Olympia, dated around 430-420 B.C.24 This sima is from a
Corinthianworkshop(the PheidiasWorkshop).It does, however,show some Athenian influ-
ence in the decoration. The spiralsare used in the sameway as those on the friezeofthe South
Hall of the Erechtheion.25The profilesof FS 867 and FS 877 (Fig. 1, profile nos. 8 and 10)
are flatterthan the otherswe have discussed.In FS 877 there is virtuallyno curve;it is almost
concave. The palmetteis very similarto that of Roof 70 at Delphi, dated by Le Roy between
420 and 400 B.C.26Again the palmettesand lotusflowersof Roof 70 are somewhatelongated.
The same designs used on the cymatia of the raking simas were often reproduced on
the soffitsof the eaves tiles. A few illustrativeexamples come not from the Tile Worksbut
from other parts of the excavation. The tripartitelotus calyx with five flower petals and
a regularlyshaped palmetteis found on FT 27727(PI.11:d);it can be comparedwith FS 867
(P1.1 :b). Another type is found on FFT15828(P1.12:a,b), whose palmettesresemblethose of
FS 865 (PI. 10:e) and FS 876 (P1. 10:d). While the lotus is not exactly the same, all have
side leaves of much the same type. FFT16829(P1.12:c, d) presents another version of this
design. A final example30(F' 156, PI. 13:a,b) is interestingbecause of the acanthusleaves
on either side of the calyx center. Such leaves occur on the South East Building3l and the
Hephaisteion-typesima,32but they turn upward in the Corinth piece and downward on
the Olympia simas and the palmettes are not the same. The meander of the face of these
eaves tiles is typical of the 5th century.
Two other pieces built into the kiln wall, a raking sima fragment33(FS 875, PI. 13:c)
and an eaves tile34 (FT 170, PI. 13:e), do not seem to fit with the others. At first glance
their thick, rather blunt palmette and lotus leaves and the straight sides of the palmettes
suggest a later date. The calyx of the lotus of FS 875 is rather sketchy and careless, but
a close examination shows that the side leaves are of the same basic type as on other simas.
The meander,too, is very poorly painted and has a wide borderaroundthe squaresuch as is
found later. It is difficult to place this piece, but because it is a discard and is so poorly
24 OlForschV, p. 115, fig. 39, pp. 132-133, pls. 42, 43; Heiden 1987, pp. 119, 121, pl. 15:2.
25 Hibner 1973, pp. 120-121, pl. 69:1.
26 Le Roy 1967, Roof 70, pp. 141-144, 148, pls. 54, 103.
27 FT 277, unpublished, was found in a well with Roman Arretine ware in the Lerna section of the
Asklepieion.
28 FT 158, unpublished,from the South Stoa area. FT 26, Corinth IV, i, p. 103, belongs to the sames series.
29 FT 168 is from the Asklepieionarea; Corinth XIV, p. 89 and pl. 22:6, from the bottom of the Northeast
manhole of the Lerna plateia drain. The date given in the text is too late.
30 FT 156, unpublished,was found in Well IV of the South Stoa. For the well, see Corinth VII, iii, p. 226.
31 Heiden 1987, pp. 127-131, 160, pl. 16:2;Mallwitz 1980, no. 42, pp. 150-151, pl. 102:4;Mallwitz 1972,
pp. 202-203, figs. 163-164.
32 Heiden 1987, p. 119, pl. 15:2;Hubner 1973, p. 120, fig. 16, pl. 69:2.
33 FS 875 is unpublished.
34 FT 170 is unpublished.
AND HELLENISTIC
TERRACOTTASFROMCLASSICAL
ARCHITECTURAL CORINTH 45
executed, while its profile (Fig. 1, profile no. 9) fits in with others of the late 5th century,
it seems likelythat it is a carelessor trialpiece. Much the same can be said of the fragmentof
eaves tile, whose workmanshipis also very poor. This type of careless painting occurs in
another sima piece built into the kiln wall35(FS 871, PI. 13:d). The meander is fairly neat
and well done, but the leaves of the palette, only nine in number, are thin and widely
spaced. The outline of one side of the palmette is straight, the other curved. The calyx
of the lotus also is crudelydrawn.
Other pieces of interestwere built into the kilnwall. One, an antefix36(FA414, PI. 14:a),
is importantbecause it is the prototype of that used in the South Stoa and other late 4th-
century buildings. The palmette leaves, eleven in number, are quite regular with a slight
double curve. The center leaf is angular at the top. The palmette is rather tall and less
rounded than those from north of Temple Hill of the earlier5th century. The center of the
palmetteis spade-shapedas before. On the lower part of the antefix,just below the center of
the palmette, are grooved spiralswith eyes. These turn inward. They curve downward
and outward to the bottom, then turn upward again along the sides, ending in a double
spiral. In the space between the two is a small diamond-shapeddart. Below the central
spiralshangs a lotus, the outer petals of which extend well out to the side. On either side
of the lotus is a half palmette. The lower part of FA 414 did not take the mold as sharplyas
it should have. The lower part of an antefix of the same type37(FA 440, PI. 14:b) shows
how it should look. The quality of this piece is quite fine. It has been suggested that it
belonged to the 4th-centurytemple of Asklepios,38but this fragmentwas a surface find in
the later excavationsof 1946-1947 and cannot be connected with the temple. As we shall
see, the antefixesof the South Stoa and the Asklepiostemple differfrom these in severalways
(pp. 48-49 below). The palmette leaves are thicker and blunter at the ends and the lotus
petals do not spread out as much.
From the kiln wall also comes a small lateral sima or perhaps a crowning molding39
(FS 1077, PI. 14:c)only 0.10 m. high. One side and part of the center where a spout was
attachedare preserved.The decorationis raisedand consistsof an S-shapedspiralon its side.
Between the spirals,at the bottom, is a lotus flower;between them, above, is a heart-shaped
flower. In the center of each spiral is an eye. The upper edge is cut out to follow the line
of the design. Is this perhaps a forerunnerof the plastic lateral simas? The work is quite
delicate. The tendrilsand spiralsare round in section. One is remindedof the workmanship
on the antefix from the cistern in the Tile Works(FA 425, p. 42 above, PI. 9:f) and some
of the antefixesat Delphi.40 Part of a finial4l (FM 38, PI. 15:a,b), also from the kiln wall, is
importantbecause of the questionsit raisesabout the firstuse of plasticdecorationon lateral
simas. It is decorated on both faces with fluted plastic tendrilswith small acanthus leaves
35 FS 871 is unpublished.
36 FA 414 is unpublished.
37 FA 440 is a surfacefind from the Asklepieionarea.
38 Coulton 1964, pp. 128-129, fig. 13.
39 FS 1077 (formerlyFM 46), from the collapsed section of the wall of the second kiln at the Tile Works,
is unpublished.
40 As, for example, series 63 and 69, note 14 above.
41
FM 38, unpublished,was built into the west wall of the second kiln at the Tile Works.
46 MARYC. ROEBUCK
along the stems. These stemsin some cases divideinto grooved spirals.At the end of some of
the tendrils is what appears to be the open end of a bell-shaped flower. This piece is on
a large scale; the preserved height is 0.32 m., comparable in size to a large lateral sima.
Comparisonwith pieces of a lateral sima42(FS 104, FS 105, FS 776, FS 1047, PI. 14:d, e)
showssimilaritiesin design and treatment. The flutedtendrils,grooved spirals,and acanthus
leaves all occur in the sima in a better-executedform. Even the bell-shapedfloweris there in
a differentform. In the finialwe seem to have the same componentpartsas in the sima. The
upper molding of the sima is decorated with a Lesbian leaf, the fascia at the bottom with
a meander of the type normally found in the 5th century. The workmanshipis very fine,
like that of the South East Buildingat Olympia, a productof a Corinthianworkshop.43If, as
Heiden argues,44the firstlateralsimaswith plasticdecorationcome afterthe constructionof
the South East Building, this sima from Corinth must belong at the very beginning of the
series. Could it be earlierthan the Olympiabuilding?One piece, FS 776, is from the Potters'
Quarterat Corinth.
An eaves tile45(Fl 172, PI. 16:a, b) froma Tile Workswell is an unusualpiece, interesting
because of the use of a regularpan tile as an eaves tile. The wide groove on the soffitindicates
that this pan tile was substitutedfor a normal eaves tile. On the face is an ivy pattern, on the
soffit what resembles a wave pattern. There are other examples of this usage at Corinth.
One comes from the Gymnasium46and is decoratedwith a similarivy pattern. A sima piece
in the Nauplia Museum47also has an ivy design. A date in the latterhalf of the 5th centuryis
indicatedfor the Corinthianpieces. The sima seems to be earlier.
Another 5th-centurypiece48 (FM 27, PI. 16:c) is a fragmentarycorner, from either a
sima or a molding. On one side is painted a large spiral from which extend two smaller
spiralsborderingthe edge. On the other side there is a meander on the lower part that is set
off from the rest. Above is a group of tendrils ending in spirals. The tile curves upward
and inward at the corner. The meander indicatesa 5th-centurydate.
42 Corinth IV, i, p. 29, note 4, p. 31, fig. 34 and p. 75; Heiden 1987, pp. 165-166, pl. 23:1. On the trip
that followed the conferenceDr. PetrosThemelis discovereda join between FS 105 and the lion's head spout,
FS 1047 (formerlyin the National Museum in Athens). This now gives us a fine nearly complete sima for
furtherstudy. The currentsuggesteddate is the firsthalf of the 4th century.
43 See note 31 above.
44 Heiden 1987, pp. 127-131.
45 FT 172 was found in Well A, a 4th-centurywell at the Tile Works;Weinberg 1954, p. 133, note 138a;
Corinth VII, iii, no. 27, p. 205.
46 Wiseman 1969, pp. 99-101, FA 519, FA 520, fig. 17, pl. 31:f. This piece is from an eaves tile. For the
ivy pattern, see Weinberg 1954, pp. 132-133, fig. 3, citing a lebes with ivy pattern found in the Tile Works;
Talcott 1935, no. 8, pp. 500-501, fig. 19, a black-glazedkantharoswith an ivy pattern. The well (1alcott
1935, p. 497) was flourishingaround 440 B.C. and abandoned around430 B.C. Other 5th-centurypotterywith
ivy decoration comes from the North Cemetery at Corinth: Corinth XIII, pp. 121, 164-165, white-ground
lekythoi with ivy pattern, dating to the middle and early third quarter of the 5th century. The ivy on the
roof tile seems more like the 5th-centuryexamples than those of the later 4th century.
47 Hubner 1975, sima no. 17352, pp. 127-128, pl. 67:4-6, fig. 6; dated by Hubner ca. 460 B.C.
48 FM 27, unpublished,has no context.
ARCHITECTURAL
TERRACOTTASFROMCLASSICAL
AND HELLENISTIC
CORINTH 47
49 FA 420 is possibly from the first accumulation of fill washed in after the Tile Works went out of use.
FA 423 comes from Late Roman fill.
50 Le Roy 1967, Roof 70, pp. 141-144, A 97 + A 174, A 98, A 99, pl. 55:18-20; also series69, note 14 above.
51 For information on workshops, see Heiden 1987, pp. 108-109, 135, 153-169, 196-199; OlForsch V,
pp. 133-134.
52 For a general account of the roof of the South Stoa, see Corinth
I, iv, pp. 83-88, pls. 19-22. For the raking
sima, ibid.,pp. 85-86, pl. 20:3. Some twenty-ninepieces were inventoried, too numerous to be listed here.
For the lateralsima, pp. 84-85, pl. 20:1 and 2 with twenty-fivepieces inventoried.
53 CointhI, iv, pp. 86-87, pls. 19:1, 20:1, 21:1a. The following antefixeswere inventoried: for eaves tiles,
FA 247, FA 248, FA 346-FA 360; for simas, FA 212, FA 249, FA 345, FA 361.
4 CorinthI, iv, p. 86, pl. 21:lb and c, pl. 21:3, right. Inventoried: rounder type, FR 54, FR 56-FR 58,
FR 60, FR 61, FR 63-FR 65, FR 67, FR 69, FR 70, FR 73, FR 83; second type, FR 55, FR 59, FR 62,
FR 72.
48 MARY C. ROEBUCK
1 2 3
repairs55(FR 66, FR 68, PI. 17:c). It seems that the tile makersfor the South Stoa copied
the earlier tiles to some extent and that there is an archaizingtendency here, revealed by
comparisonwith a sima from the Tile Works56(FS 867b, PI. 18:a). In the profiles of the
rakingsimas of the South Stoa and FS 442 from the Asklepieion(Fig.2, profilenos. 1 and 2)
we also see a return to the earlierform, rounderand projectingout more.
One group of tiles from the Asklepieion,belongingeither to the temple or to some other
building of similar date, is very like those of the South Stoa. Pieces of the raking sima57
(FS 442, P1. 18:b), the lateral sima58(FS 433, PI. 18:e), and the antefixes59(FA 184 and
FA 434, P1. 18:c, d) have all been recovered. The leaves of the palmettes of the rakingsima
CorinthI, iv, p. 88, pl. 21:3a and b; inventoried:FR 66, FR 68, FR 71.
5 FS 867b, unpublished,was built into the wall of the second kilnof the Tile Works. The palmette has more
regularlyrounded leaves than others such as FS 860 and FS 865 (pp. 42-43 above, Pls. 10:e, 1 L:a). Both
types occur built into the kiln wall and are of approximatelythe same date.
57 FS 442, unpublished. Another possible set for the temple is made up of FS 439, FS 440, FS 444.
M
CoringXWV, p. 36, pl. 11:5;Heiden 1987, p. 155. Inventoried:FS 433-FS 436.
59 FA 184 and FA 434 are unpublished. Other antefixesinventoriedare FA 143, FA 144, FA 146, FA 181-
FA 188, FA 192, FA 577.
TERRACOITASFROMCLASSICAL
ARCHITECTURAL CORINTH 49
AND HELLENISTIC
are less rounded than those of the South Stoa and tend to have a double curve again, while
the acanthusleaf on the fluted stem of the lateralsima does not curve upward.
An entirelydifferentgroup of simasfrom the 4th centuryhas also been found at Corinth
as well as at other sites such as Olympia, Delphi, and some smaller sanctuaries.60 These
are raking simas with a painted egg-and-dartband on the cymatium. Several have been
found at Corinth.61 I show one here62(FS 10, PI. 18:). There is some variation in style
and date, but they seem to fit in the middle to the end of the 4th century,this being one of the
latest examples,perhaps even 3rd century.
In conclusion,I mention a few pieces from the Hellenisticperiod to complete the picture
of Greek Corinth. In them one sees a return to greater vitality and imagination in the
treatmentof the designsused over severalcenturies.
The area to the north of Temple Hill contained a series of stoas in the Greek period.
This part of the excavation was uncovered in 1930.63 In the most recent of the stoas and
at the north side of a street that bordered it were found many architecturalterracottas.
Among them were two fine complete simas, one a cornerpiece, the other a lateralsima. The
excavatorhas suggestedthat afterthe destructionof 146 B.C. some Sikyoniansclearedpart of
the area of the blocks and any other material of use to them. Presumablythe tiles would
have fallen on the blocks so they were moved aside and stacked separately alongside the
piers of the basement. The two sima pieces were placed beside Pier 17. It is presumed
that the tiles formed the roofing of the buildingsin the area at the time of the destruction
of the city by Mummius.
The corner sima piece64 (FS 382, PI. 19:a, b) is particularlyinterestingbecause of the
way in which the tile makersmade the transitionfrom the plastic decoration on the lateral
side to the painted decoration of the raking sima. On one side of the lion's head spout
are the customaryplastic tendrils with an acanthus leaf from which grow two spirals. In
the fork of the spiralsis a bud. On the bottom projecting molding is a painted meander
pattern. At the top is a torus molding with a painted egg and dart. The lion's head is very
highly modeled, with a deep groove down the center of the forehead and wild locks of hair.
On the other side of the spout the upper molding is plain at the top with a Lesbian leaf
below. On the cymatium, next to the spout, is half of a painted palmette. At the corner
is a flower with five petals which extends across to the raking sima side. On this side the
upper border continues unchanged, but the meander at the bottom continues on a recessed
fascia. The cymatium has an alternatingeleven-leafedpalmette and a nine-petaled flower
with notched ends, giving a flame effect. The palmette has a diamond-shapedcenter, the
flower a bell-shaped center. Between the two of them at the bottom is an S-shaped spiral.
60 Le Roy 1967, pp. 153-154, 168, series 78-79, pls. 58 and 104; Heiden 1987, pp. 175-177; Bomme-
laer 1978, pp. 173-197; Mallwitz 1980, pp. 149-150, no. 45, pl. 102:4.
61 Corinth i, pp. 67-68, 70, 73-74, 76, FS 10, FS 12, FS 19, FS 38, FS 53, FS 59, FS 106-FS 108;
IV,
also FS 902, FS 1024, and FS 1070, all unpublished.
62 FS 10, Corinth IV, i, p. 67.
63 De Waele 1931, pp. 410-411, 416-418.
64 De Waele 1931, p. 417, fig. 10; with this also go FS 414 and FS 1014 as well as rakingsima pieces FS 40
and FS 51 (Corinth1V,i, pp. 71, 73) and unpublishedpieces FS 384, FS 429, and FS 903. See also OlForsch
IV, pp. 84, 127, pl. 94.
50 MARYC. ROEBUCK
This sima measures0.28 m. in height and 0.69 m. in length for the lateralsima and 0.76 m.
in length for the raking sima; overall it is not quite as large as the tiles of the South Stoa.
Probablyit belonged to the roof of the stoa in which it was found.
The other sima piece65 (FS 381, PI. 19:c)is a complete section of a lateral sima. The
design is basicallythat used in the 4th century,but it shows some differencesfrom the South
Stoa. The upper molding is decorated with a Lesbian leaf pattern. The meander on the
base molding is raised rather than painted. The acanthus and spiral design on either side
of the lion's head has an acanthus leaf alongside the spout, not found on the South Stoa
but found on some other 4th-century simas. There is a small spiral added to the top of
the end spiral, as well as a bud. In the fork of the spiralsbelow is another bud. The hair
of the lion is rathermore formalizedthan in the other sima. With this sima can be connected
a numberof pieces of rakingsima66(FS 50, P1.19:d)found both to the north and to the south
of Temple Hill. The upper molding is the same and the meander is also raised, not painted.
The principal decoration is similarto that on the rakingsima of the corner piece, but the
palmette has fewer and narrower leaves and there is a Mycenaean-type lily between the
palmette and the flower. The profile (Fig. 2, profile no. 3) once more is fairly flat. With
these should be put some very large ridge tiles with palmettes67(FR 39, FR 41, P1. 19:e)
found in the same area. These pieces are all on a large scale. The lateral sima measures
0.39 m. in height and 1.13 m. in length while the preservedheight of FS 50 is 0.36 m. and the
height of the ridgepalmetteis 0.31 m. Tiles of this size, found in this area, can only belong to
one building, the Temple of Apollo. I think we can safely say that the temple and the stoa
below were roofed with these impressivetiles at the time of the destructionof the city in
146 B.C.
BIBLIOGRAPHY
65 FS 381, De Waele 1931, pp. 416-417, fig. 9; Heiden 1987, p. 155 and note 713; OlForsch
C, pp. 83-86,
pl. 79, WillemsenevidentlymisreadDe Waele. His date of the 1stcenturyB.C. for the two simas(pp.84,86, 127)
I, i, pp. 80,89) and the unpublished
is incorrect. Others of the same seriesare FS 141-FS 143, FS 250 (Corinth
pieces FS 417, FS 421, FS 422.
66 FS 50, Corinth IV, i, pp. 72-73, fig. 30, p. 28 and note 6. Others belonging to this series are FS 49
(p. 72), FS 58 (p. 73), and the unpublishedpieces FS 819, FS 915, FS 916.
67
The ridge tiles FR 39-FR 43 are unpublished.
AND HELLENISTIC
TERRACOTTASFROMCLASSICAL
ARCHITECTURAL CORINTH 51
MARYC. ROEBUCK
P.O. Box 65
Eastsound,WA 98245
PLATE 9
1,~~~~~~~~~~~~i
W~~~~~~~~~~~~
ITTM
a. FA 125. Combinationeaves tile and antefix b. FA 417. Antefix from kiln wall in Tile Works
~ppzs p
A~~~~~~~~~~
sa *'h
A~~~~~~~~~~
c. FS 6. Raking sima
e. FA 415. Akroterion(?)from kiln wall in Tile Works f. FA 425. Antefix from cistern in
Tile Works
0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~p
tv~~C
-~~~~o
m
:
IL~~~~~~~~~~~~~~~~
O Q
Cd~ ~ ~ ~ t
__~~~~~~~bO~~~~~~~~~~~~~~~~~L H
co~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~c
~~~~~~~~~~~~~~~~~~~~j..
v, t...
Cho~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
. ............................... 0
Y - _ _ r~~~~~~~~~~~~
Urn. ~~~~~PILATE
11I
co~~~~~~~~~~~~~~
bo~~~~.
PLATE 12
'~~~~~~~~~~~~~~
_a._..
a. FT 158. Eaves tile from South Stoa area, soffit b. FT 158. Face
~>-<-*
~~~rr~~~
O. ~~~-~
*)
-
3,'."
* Fp
-3 T n a- J'
d. FT 168. Face
MARY C. ROEBUCK:ARCHizTECURALTERRACOTTAS
FROMCLASSICAL
AND HwIEIsNITC CORINTHr
PLATE 13
AV,~~~~~~~~~~~~~~U
LXI,~~~~~~~~~~~~~~~~~~~~~~~~~I
4)
_4_ _
V.~~~~~~~~~~~~
H ~ ~ ~ ~ ~ ~~~~* 7
CU~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~l
U,
C-C
I-
d
I~~~~~~~~-
CAR
* M
~ _C
4. ~ ~ ~ ~ 2
C '~LO .A!4I3U~
*t_ _ _ _ -
* _ 1 -_ _ H
PLATE 14
a. FA 414. Antefix from kiln wall in Tile Works b. FS 440. Antefix fragmentfrom Asklepieionarea
AI~~~~~~~~~~~~~~~~
d.1 r ~ ~
I0sis _ _ __ U
......t_'
MAuRYC. ROEBUCK:ARCHITECTURAL
TERRACOrAS FROMCLASSIC.AL
AND HnmasnIc CORINTH
PLATE 15
7;.
-I-___~~~~~~~~~~~~
b. F:M38. SideB
C. ROEBUCK:
MAR~Y ARclirECmnw..TmuRAcorrASFROMCLASSICAL.
ANDiHEuEIsnIc CORINTH
PLATE 16
C14~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-
C14~~~~~~~~~~~~~~~~~~~~~~~~~C
C-i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-
C,,~~~~~~~~~~~~~~~~~~~~~~C
- CU~~~~~~
PLATE 17
N
-
b. FR 62 (left),FR 65 (right). South Stoa ridge palmettes c. FR 66 (left), FR 68 (fight). South Stoa ridge palmettes,
from originalroof Hellenisticreplacements
MARGYC. ROEBUCK:
ARCHITzc uRALTEltAcorrAs FROM CLASSICAL.
ANDtHmnsnc COR~INTH
PLATE 18
I'd A~~~~~~~~~~~~~~~~~~~
m - m
V_ r
MARYC. ROEBUCK:
AitcmmEcnnAL TmuucoACrs FROM CLASSICAL
ANDHNamnsnIc COluNTH
PLATE 19
- AieJ ml
%
L A&f V% MAL
c. FS 381. Hellenisticlateral 77 -
d.FS5O. Hellenus- Lw
v~ ~~~~~
usedwithlateral s<
e. FER41(left),FR 39 (right).Hellenistic_
ridge palmettesfrom Temple of Apollo
MARY C. ROEBUCK:ARCHizTECURAL TERRACOTTAS
FROMCLASSICAL
AND HwrLNISnIC CORINmH
ROOF TILES FROM TWO CIRCULAR BUILDINGS
AT CORINTH
not deform the visual effect of the overall diamond pattern of the roof once the tiles are
in place and overlapped. Indeed, the dimensions of the lateral edges are not only small
but may vary by tile. Moreover,the lateraledges are not cut parallelto one another. Rather,
they were angled 2 degrees from the longitudinalaxis of the tile. If the lateral edges were
extended upward, they would meet at the apex of the roof. Assuming that all lateral edges
angled 2 degrees off the longitudinalaxis (as they do in the preservededge of FM 2 1) and
assuming that all tiles at the level in the roof in which this tile lay had their lateral edges
abutting,a maximum of 90 tiles could be restoredat this ring. Tile FM 21 is ca.0.54 m. thick
at its midpoint. If all tiles of this one ring were similar,then the ring of roof tiles would have a
circumferenceof ca. 48.60 m.
The finished downside vertical edges of the tile are about 9 degrees off the vertical
when the tile is laid horizontal. The 9 degrees probablycan be consideredan approximate
indicationof the general slope of the roof.
All these tiles are made of a Corinthian buff to tan clay, going pink at the core. All
have liberal amounts of mudstone in the mix. Except that the tiles were designed for a
conical roof, they are typical products of the Corinthian roof-tile industry. Unfortunately
no decorated eaves tiles have been discovered to go with the building for which the tiles
were designed.
With this evidence at hand, one can restore this specific Corinth roof with every tile
having the same angle on all four of its diagonal sides. If all tiles should ever prove to
have equal corner angles, the roof could be restored with small, relatively square tiles at
the apex of the roof with a resultantawkwardnessof design there and largertiles of the same
shape at the eaves. The more satisfactoryconclusion, but not yet attested as the correct
one, is that the roof tiles graded from a long narrow hexagon at the top to broader and,
in proportion,shorterhexagonaltiles as the horizontalrows of diamondtiles approachedthe
eaves. This latter solution, as restored on the roof of the Tholos of the Athenian Agora,
is also suggestedas the better solutionfor this Corinthianroof.4
Seven rather unassumingfragmentsof fictile roof tile attest the existence of a second
round building at ClassicalCorinth. No sima fragmentsor roof palmetteswere found with
the fragments. All the tiles found had been dumped onto bedrock under Building 5 of
East Theater Street, apparentlyas part of an Early Roman cleanup by the new Corinthian
colonists.
Despite the paucity of fictile evidence, a number of interesting principles can be
established about the design of these tiles and roof. The fragments are all small; none
has a length of more than 0.22 m. or a width of more than 0.24 m. The single most
importantfragment,FM 134, is a flat, thick (0.04 m.) tile that preservespart of both lateral
edges, the top edge of the tile, and a hole for nailing or pegging the tile to the roof timber.
The top and side surfacesare slipped;the bearingundersurfaceis not only not slippedbut is
quite rough, although flat. This tile reveals a number of differentfacts about the roofing
system of the specificbuildingit covered.
First, the tile has sides that flare outward from a flat top edge that is 0.146 m. wide
(Fig. 1). The flareis ca. 9 degreesto a side and indicatesthat the roof from which it came was
4 Miller 1988, pl. 48:1, 2.
ROOF TILES FROM TWO CIRCULAR BUILDINGS AT CORINTH 55
0.146
0 5 cm
FIG. 1. FM 134
either conical or multifaceted. Twenty such tiles ([9 + 9] x 20 = 360 degrees) would be
needed, laid side to side, to complete the tile ring at this level of the roof.
Second, a reservedband (0.018-0.020 m. wide) is defined along the longitudinalaxis of
the top surfaceof the tile by two parallelblack-glazedlines, each between 0.015 and 0.017 m.
wide. A distinct broadening of the glazing startsjust above the break at the lower end of
the tile. This type of decoration,at this moment unique to these fragments,and the findspot
stronglysuggestthat all pieces came from a single roof.
Third, the top surface of the tile is totally flat and has neither lip nor channel along
any preservededge, such as are to be found on all normal Corinthian pan tiles. A design
without such lips could allow leakageat the overlap,a fact that was apparentlyof no concern
to the designerof this circularbuilding.
Fourth,a hole 0.013 m. in diameterhas been drilledinto the tile afterfiring. This hole,
on the longitudinal axis of the tile and at ca. 0.05 m. from its top edge, apparentlywas
once used to peg the tile to the wood substructure.
Five other fragments,FM 127, FM 129, FM 136, FM 137, and FM 138 (Fig.2:a-e), add
one more fact. Of the group only FM 136 does not preserve any part of a lateral edge.
One sees a numberof genericcharacteristicswithin this group. As with the firsttile, FM 134,
the top slipped surfaceof all the fragmentshas a reservedcentralband between glazed lines
or areas running along its longitudinalaxis. These two glazed bands widen as they progress
to the bottom edge of the tile. Where the four tiles preserve part of a lateral edge, that
edge flares at ca. 11-12 degrees from the centralreservedband of the tile.
56 II
CHARLESK. WILLIAMS,
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.
a~ ~ ~ ~~~~
t. *r 1,, i,~~~~~~~~~tt
*~~ I
0 5
/ 1 :'. / i:-:.}
I:;/ 2
d e
FM 127 (Fig. 2:a) adds two facts. First, it preservestwo lateral edges, but those edges
do not have the conventional raised lip found on the usual Corinthianpan tile. The flare
of the lateral edges is ca. 26 degrees between the sides. The glazed bands that flank the
central reservedstrip here are broad, ca. 0.032 m. wide, and curve out to the lateral edges
of the tile. On one preserved side (left, when looking down on the top of the tile) one
can see that the glaze is continued from the top surface down the vertical side edge.
If two such tiles were fitted together side by side, the resultantreserved area would define
a pointed scallop or scale.
The bearing surface, or underside, of this tile adds still more information to the
understandingof this special roofing system. Although the undersurfaceis rough, as on
all the tiles of the series, the surfaceis articulatedby two parallelchannels cut into the clay,
apparentlywhile the clay was wet or damp. There are also signsthat the ends of the channels
were enlargedthereafter,perhapson thejob. The channelsare 0.06 m. wide, almost 0.02 m.
deep, running lengthwise, and about 0.05 m. apart. The heads of the channels start at
ca. 0.045 m. above the point where the painted top surface goes down the lateral face of
the tile.
From the evidence of this tile, then, we have the following information. The painted
design of the top surfaceproducedpointed scalesor scallops. In addition,certainroof tiles of
the overall roofing system did have raised lips on their lateral edges, at least at some point
or points where certain scale tiles overlapped other tiles. The lower series was laid with
a 0.05 m. space between adjacent lateral edges, attested by the channels described above.
One could then restore at least one row of lip-edged tiles under the series of scalloped tiles
representedby FM 127.
Pan tile FM 128 (Fig. 3) has no original side edge preserved. Its top surface as
preservedis solidly glazed except for the customaryvertical reservedline along the axis of
the tile, here 0.015-0.017 m. wide. The upper left-hand corner of the top surface may
possibly present a small area of reserved scale. If not, a brush stroke in the glaze at
this point indicates that the reserved scallop area of the upper portion of the tile was
just beyond the preserved break.
It is the bottom surface of this fragment that offers the important information. The
undersurfacepreservestwo lines scratchedinto the rough clay, crossingeach other at right
angles, as well as the top end of a channel that was meant to receive the lipped side edge
of an underlying tile. The cutting is ca. 0.095 m. from the incised line that demarks the
longitudinalcenter of the tile. The cutting here is not only positionedmore to the side of the
center of the tile than are either of the cuttings on FM 127 but also appears to be lower
(closer to the front of the scalloped tile by a good 0.15 m.) than are the channels in the
undersurfaceof FM 127.
The clay of all fragmentsof this group is greenishcream with a pinkishcore. The clay is
comparable to that used in a large series of Classical Corinthian roof tiles and also close
to the fabric of the roof tiles of the South Stoa but with more grit than is usually found in
Hellenistictiles manufacturedat Corinth. It has a heavy admixtureof large-sizedfragments
of mudstone. The surfaceis covered with a fine clay slip.
58 CHARLES K. WILLIAMS,II
o s
0 5
FIG.3. FM 128
The above information permits a proposed restorationof the roof following certain
design principles.A roof with scallopedterracottatiles (Fig.4) apparentlyfollowsthe rulesof
design governing the representationof feathersin terracottasculptureand pottery.5
According to these rules, once the pattern is adopted, each superimposedrow of tiles
must overlapso that the axis of each tile of any overlappingring is on axis not with that of the
tile immediatelybelow or above it but with the tile two rows below or above, that is, in every
other ring. This means that axis lines descend from the apex to the eaves both along the
centralaxis of each tile in the firstring below the apex and from thejoint between tiles of that
ring. The joint axes would, of course, descend in a line that runs along the axis of each
tile in the second ring below the apex. Since the roof is a cone, these axis lines flare as
they descend to the eaves.
Two inconveniences arise from such a design. One is the increasingsize of the tiles in
every ring as they approachthe eaves. The tiles of every ring must be largerthan those that
overlap them. Such a design allows only limited scope for mass production. Second, for a
roofthat is designedto cover a buildingof largediameter,the tilesas they descend can become
out of scale with the other architecturalelements and impossibleto manufactureor lay. At
the bottom edge of the roof the large scale of the tiles could easily become inappropriate
to the scale of the sima. This problem was avoided in the LysikratesMonument by the
addition of a secondary sima ring close to the edge of the roof at the point where the rings
of tiles would have become so large as to become a design handicap; below this point the
tiles could be started once more in reduced size. The result is that tiles of reasonable size
could be continued from there to the eaves, with the terminalring of tiles relatingdirectlyto
5 The influence of featherson the design of scallopedroof tiles can be seen in the marble roof of the Late
ClassicalLysikratesMonument and in Roman structuressuch as the BabbiusMonument at Corinth (Corinth I,
iii, pl. 12:2) and a circularbuilding immediatelywest of the Odeum (Corinth X, p. 109, fig. 100). A circular
foundationthat lies immediatelywest of the Odeum, discoveredafterthe publicationof the Odeum itself,is of a
size and date that are appropriatefor this roof slab. The two small monopteroiof the Nympheum of Herodes
Atticus at Olympia use the feather-patternedroof but with a creased central axis, not a rib. See OlForsch I,
p. 75, fig. 15. The terracottaroof from Corinth under discussionin this paper should be equated with the
designs in marble, except that here the central reservedline representsthe central quill of a natural feather.
Domed roofs have been excluded from this list.
ROOF TILESFROMTWOCIRCULAR ATCORINTH
BUILDINGS 59
0.146 m.
s-s
_I 0 .i
II S~I
the sima tiles, which in turn would be related to the scale of the facade. The solution of a
secondary sima near the bottom of the scalloped roof may not have been needed-on the
Corinth roof if the building was small enough. We do know, however, from the cuttings
in the undersurfaceof FM 128 and FM 127 that the width of the row of tiles at the eaves
got out of synchronizationwith the scale-patternedtiles. This row probably was of sima
tiles with a more or less conventional Corinthian pan-tile design, with a raised lip on the
side edges.
Another disadvantageof this roof design arises from the lack of interlockinglip-and-
channeljoints. In the AthenianTholos and the first-mentionedCorinthiancircularbuilding
the tiles have interlockingjoints. In this second example, however,the builder(s)apparently
hoped that a generousoverlapwithout interlockswould protectthe monument from leakage
and water damage. Perhaps the amount of calculation necessary for designing accurate
lip-and-channeljointing was too much effort for the project, whatever the intended use of
the building.
In any case the scale-patternedroof appearsto have been a success for small buildings,
for the design was used into the Roman period, although such roofs in Corinth were then
executed in marble, not terracotta. The Babbius Monument and at least one other such
60 CHARLESK. WILLIAMS,
II
monument attestto the use of this form into the 1stcenturyafter Christ. Such marble roofs,
however, resolve the problems posed by this design. Leakage between large numbers of
small tiles is avoided and the tiles need not be speciallymanufacturedin small groups; the
whole terracottadesign is simplycarved onto large, wedge-shapedmarble slabs.
BIBLIOGRAPHY
CHARLEs K. WILLIAMS,II
CoRqiTH EXCAVATIONS
American School of ClassicalStudies
54 Souidias Street
GR-106 76 Athens
Greece
GREEKARCHITECTURALTERRACOTTAS
FROM THE SANCTUARYOF POSEIDON AT ISTHMIA
(PLATEs
20-22)
ARCHAIC TILES
Figure 2 is the reconstructedplan of the temenos for the period 575-550 B.C., soon after
the inaugurationof Panhellenic games at Isthmia. Of particularimportance is the shape
of the temenos plateau and the surroundinggullies, which are oriented from southwest
1
Fora summaryof the Isthmiansanctuaryat the end of Broneer'sexcavations,see Broneer 1976. Broneer's
descriptionand interpretationof the remains,exclusiveof the Temple of Poseidon, are describedin IsthmiaII.
Additional work was done in the Roman bath and the Late Antique wall and fortress by Paul Clement
(1967-1976). Elizabeth Gebhard succeeded Broneer as Director in 1976, and Timothy Gregory followed
Clement in 1986.
2 Cf. Gebhard and Hemans 1992, pp. 1-77. The reader should refer to this source for a description of
the stratigraphyand the new interpretationsof the chronology of the temenos up to the destruction of the
Temple of Poseidon, ca. 470 B.C. The next report will describe the sanctuaryfrom ca. 470 to the 3rd century
after Christ.
3 I would like to thankall my colleaguesfor their continuedassistancein my studiesat Isthmia,in particular
the Directorof the project,ElizabethGebhardof the Universityof Illinoisat Chicago. I would also like to thank
Nancy Winter for her invitation to take part in this conference as well as for her observationson the Isthmia
materialmade during the summer of 1991. Recent work at Isthmia has been made possible by a three-year
grant from the National Endowment for the Humanities (#RO-21847-89). See Gebhard and Hemans 1992,
p. 1, note 3, for a list of staffmembers in the currentproject.
4 The amount of material from buildings other than the Archaic and Classical Temples of Poseidon is
relatively small. Thus, we are more in a position at the present stage of the excavations to benefit from
previous scholarshipthan to contribute a great deal of new information on the subject of Corinthian roof
tiles. The exception is the roof of the Archaic Temple of Poseidon, from which thousandsof fragmentshave
been recovered. See Hemans 1989.
62
0~~~~~~
0 0~~~~~~~~~~~~~~~~~~
0,~~~~~~~~~~~~~~
0~~~~~~~~~~~~~
(I >7'>
I,,0 0
61~ S/ 4
C's
2,~~~~~~~~~~6
09~~~~~~~~~~~~
I'z,~~~~~~~~~~~~~~O
4,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~c
63
z *-=
00
0
~~~~~~~~~~~LJ
X, XOn
LO)
, A/ ' E
LC)
0s LOi:
LC)
0 : X' 2
LO
CFS
CUS
N, ~ N /
in -77
/n
D-/
~ ~~~~~~
~ ~ ~
0
zm
~
it~
~ ~ ~ ~
~ ~ ~ ~~
Ld~~~~~~~0/C
N ~~~~~~~
/U
NX ~ ~ ~ ~ N
/ 0~~~~~~~~~
64 FREDERICK
P.HEMANS
to northeast. With the exception of the Temple of Poseidon, the topography determined
the orientation of all the Greek buildings. Not until Roman times was the temenos made
rectangularto follow cardinal directions (Fig. 1). The earliest well-preservedfeatures are
the Archaic Temple (completed ca. 650 B.C.) and the temenos wall, which defines limits
that were retained until the 5th century.5 The eastern part of the temenos, surrounding
the Long Altar,servedas the place of sacrificethroughoutthe historyof the sanctuary;to the
southeastlay the early stadium that was the center of the celebrationof the games.6 Along
the northernedge of the plateauwas a highwaythat carriedtrafficbetween Corinth and the
Isthmus. Although the position of the road appears to intrude awkwardlyinto the space
surroundingthe temple, no other place in the vicinitypresenteda gradient shallow enough
to allow constructionof an alternativeroute at a greaterdistancefrom the temple. Thus, the
road was alwaysan importantfeatureof the site.
At the present stage of the project, a large area within the temenos, at the northeast
corner of the plateau, has not been excavated, and the areas immediatelysurroundingthe
temenos remain largely unexplored. Of the Archaic tiles that have an excavated context,
most were discoveredin the NorthwestReservoir,suggestingthat buildingsmay once have
lined the road which led from Corinth.7 The small scale of the terracottasindicates that
these were treasuries(oikoi),small stoas, or propylaia.
The earliestpainted tiles at Isthmia are fragmentsof two types of pentagonal antefixes
and a raking sima. These are late examples of their types and can be dated stylistically
to the second quarter of the 6th century, suggesting that at least two buildings were
constructed soon after the traditional foundation date of the Isthmian games in 582 or
580 B.C. (1-5).8 No architecturalfragments or building foundations of comparable date
have yet been recognized, but excavation in 1989 has provided evidence that there were
extensive additions to the sanctuary during this period. At the southeast corner of the
temenos the firststadiumtrackwas created as well as a walkwaythat linked the trackto the
temenos.9 Across the north half of the temenos the terrace between the temenos wall
and temple was resurfaced. It seems likely that the buildingsrepresentedby the terracottas
formed part of the same building program.
During the latterhalf of the 6th century,the stadiumwas expanded by the addition of a
spectatorembankmenton its north side. Another, larger terrace was laid over the eastern
side of the temenos, which soon afterward,duringthe last quarterof the century,was paved
with sea pebbles to providea durablesurface. During the same period formal gatewayswere
added to the temenos at the north and east. The north propylon is the better preserved
CLASSICALAND HELLENISTICTILES
Much larger numbers of Classical and Hellenistic tiles have been recovered than Archaic
ones, but the vast majority come from only three roofs. Thirty of a total of 33 antefix
fragmentsare from these three roofs: 3 from Roof 1, 13 from Roof 2, and 14 from Roof 3.12
The eaves tiles show a similar distribution(Roof 1, 13 and 14; Roof 2, 19; Roof 3, 21).
Other types of tiles are present in very small quantities,often single examples from a roof,
and the buildings they representwere probably located at some distance from the center
of the sanctuary:for example, the theater,located below the temenos to the northeast.
Afterthe ArchaicTemple burned ca.470 B.C., the layout of the temenos was significantly
altered (Fig. 3). The temenos wall was dismantledacross most of the north and east sides
of the sanctuary;by the end of the centurya new beddingfor the road to the isthmuswas laid
10 IsthmiaII, pp. 10-11. For the date of the North Propylonin the 6th century,its relationshipto the Archaic
temenos wall, and the identificationof the East Propylon,see Gebhard and Hemans 1992, pp. 47-51, 73-74.
l l Deposits after 470 B.C. will be describedin the next report (see note 2 above).
12 The catalogue includes only Classical/Hellenistictiles for which
more than one example has been found.
Thus, of the 33 Classical/Hellenisticexamplesof antefixes,only 30 are listedin the catalogue,underRoofs 1-3.
I-J~~~~- C
C' 0~~~~~~~~
Ii -,~~~~~~~~~~~~~~~~~~~0
Cl
1~~~~~~~~~~cd
crU~~~~~~~~~~~~~~~~~~~~~~~~~c
CL~~~~~~~~~~~~
\? 0~~~~~~~~~~~~~~~
ARCHITECTURAL
TERRACOTTASFROMTHE SANCTUARYOF POSEIDON 67
along the north side of the plateau and the stadium had been completely rebuilt. One of
the new buildingsconstructedat this time is representedby a group of three finely executed
antefixfragmentsand six eaves tiles (Roof 1, 12-14). The styleof the antefixesis comparable
to examplesfrom Delphi (Groups67-70) that ChristianLe Roy datesto the second half of the
5th century.The most closely dated examplesof the style are from Olympia, recoveredfrom
Pheidias'workshop,and are dated by AlfredMallwitzto ca.430 B.C. All are characterizedby
volutes in shallow relief with a very smooth contour and fine painted detail. The Isthmia
type, however, has a simpler design than those found either at Olympia or Delphi and is
probablysomewhatearlier.A cornice block and triglyph-metopeblock found on the slope of
the NorthwestGullydate to approximatelythe same time period and might also be associated
with the same building.'3 Both blocks are in good condition with fine work, including a
painted meander on the fascia at the bottom of the cornice block.
Two unusual fragmentsof combination pan/cover tiles made of poros stone probably
also date to the 5th century(10 and 1 1). Unfortunately,neitheris well enough preservedfor
us to be certain of all the characteristicsof the roof or even to know whether they both came
from the same building. The largerpiece (10) preservesnearlythe fullwidth of the cover and
a small portion of the pan. Although this is a combination tile, the lack of undercutting
beneath the cover suggeststhat the tile was used at the edge of the roof. A stucco coating
is preservedacrossmuch of the top surface.
The smallerpiece (11) preservespart of one side of a cover risingfrom a finished edge.
Part of the cover has been cut away from the vertical edge of the tile at an angle of ca. 20
degrees. On the underside,towardthe frontof the tile, a cuttingwas made at a rightangle to
the preserved finished edge, either to overlap the tile beneath it or the edge of the roof.
Behind this cutting a sufficientportion of the undersideof the tile remains to show that the
cover overlappedthe adjacentpan tile. The excavationcontext of the pieces, from deposits
of Late Hellenistic and Early Roman date, suggests they were made after the destruction
of the Archaic Temple, ca. 470 B.C. 14
Between the second half of the 5th century and the latter half of the 4th century there
appearsto have been little constructionactivityat Isthmia. The next group of tiles (Roof 2,
15-19) dates to the end of the 4th century, based on their similaritiesto those from the
South Stoa at Corinth.15 Antefixes, ridge palmettes, and a raking sima (17, 18, and 15
respectively)all have the same designs and workmanshipas the South Stoa types and must
be close in date. In addition,the eaves tiles of the roof (19) have been identifiedbased on the
similarityof the workmanshipof the meander design with that of the rakingsima.
Another type of 4th-centurysima with a large egg-and-dartmotif on the cymatium can
also be assigned to the same building (16). The attributionis made because the meander
3 IstimiaII, nos. 117 and 118, p. 131.
14 Only Classicaland Hellenistictypes have been found in the debrisfrom the early 2nd-centurydestruction
of the sanctuary.
15 A well-preservedlion's head corner sima with the same design on the cymatiumas IA 707 (15) was found
in the debris of the Rachi settlement, located on the ridge south of the sanctuary. The fine condition of the
sima is anomalous in comparisonwith materialfrom the sanctuary.If this sima was used on the same roof as
the fragmentsfound in the sanctuary,and then reused on the Rachi, it must have been removed from the
sanctuarybefore the destructionof the Rachi (ca.200 B.C.) in orderto account for its presence in the settlement.
68 FREDERICKP.HEMANS
pattern on the lower fascia and the bead and reel on its soffit have the same workmanship
as those on the sima and eaves tiles. If a greaternumberof roofswere representedin the tiles
that have been recovered at Isthmia, this attributionwould seem less likely. The fact that
so few roofs are representedmakes their shared characteristicsseem a stronger reason to
relate them to the same building.
Fragmentsfrom Roof 2 were found in every section of the temenos (17 and 19), but
most were found outside the temenos, with a concentrationto the south and southeast. This
distributionsuggeststhat Roof 2 representsa buildinglocated outsidethe southeastcorner of
the temenos. Since the tiles are similar to those used on the South Stoa at Corinth, one
might speculatethat a stoa was erected adjacentto the stadium.
The remaining tiles, grouped under Roof 3 (20-22), do not share common features.
The meander pattern of the sima is quite differentin both color and draftsmanshipfrom
the eaves tile. Nor can any feature of the antefix be directly associatedwith the eaves or
sima. All these tiles, however, are stylisticallylater than the pieces from Roof 2. The sima,
in particular,is carelesslymade in comparisonwith the earlier roof. While it is possible
these pieces do not belong to the same roof, all are roughlycontemporaryand representedin
large numbers. It may be the case that by the 3rd century carefullymatching the designs
of differentcomponents of a roof was no longer of any great concern. Another possibility
is that the variouselementsof the roofwere not designedfor a specificbuildingbut ratherthe
roof was created from pieces taken from an inventory.
There are several poros architecturalfragmentsthat belonged to small buildings con-
structedin the late 4th and early 3rd centuries. FoursmallDoric capitalshave been found in
the temenos and two outsideit; they were datedby Broneerto the 4th centuryand assignedto
three separatebuildings.16Fourlargefragmentsof Doric cornicesof the same date were also
recoveredby Broneer in the temenos area; three of these belonged to the same structure.17
Although none of these poros architecturalfragmentscan be assigned to specific building
foundations,they confirmthe picturerepresentedin the tiles that at least two smallbuildings
were erected near the temenos duringthe late 4th or early 3rd century.
CONCLUSIONS
Building activity at Isthmia appears to have been concentrated within relatively discrete
periods of time. The dates suggested for the tiles on stylisticgrounds correspond to the
periods of constructionat the sanctuarywhich have been documented by deposits cleared
in the old and new excavations. This correspondencesuggests that the smaller buildings
were conceived as part of largerbuildingprogramsthat encompassedthe entire sanctuary.
16 IsLhmiaII, nos. 100-102 and 104, p. 130, were found within the temenos. The two found outside the
temenosbelonged to the same buildings;no. 99 is comparableto 100, and 103 (foundin the Rachi settlement)is
comparableto 104.
17 IsthmiaII, nos. 120-122, p. 132, from one building. No. 126 was from a second building. An additional
cornice, no. 123, found in the NorthwestReservoir,was thoughtby Broneerto be from a thirdbuilding. Three
other fragmentsfrom other buildingswere found at greater distancesfrom the temenos. Nos. 124 and 125
were found in the Sacred Glen, and no. 127 was found in the later (Hellenistic)Stadium.
TERRACOTTASFROMTHE SANCTUARYOF POSEIDON
ARCHITECTURAL 69
During the first period in the second quarter of the 6th century, small buildings were
completed following the inaugurationof the games. The second period falls near the end
of the 6th centurywhen at least three additionalsmall structureswere added. Remains from
each of these buildingprograms,however,are few and the only foundationsyet identifiedare
the propylaialocated on the north and east sides of the temenos.
After the fire of the 470's B.C., during the third quarter of the 5th century, the north
temenos wall was dismantled and the boundary moved closer to the Temple of Poseidon.
At least one small building was added within or near the temenos at about the same time
as the new temple was being constructed.The last majorbuildingperiod representedby the
tiles occurs during the late 4th and early 3rd centuries, when at least two small buildings
were constructednear the temenos.
CATALOGUE'8
All measurementsare in meters unless otherwiseindicated.
ARCHAIC TILES The height of the tile, the angle of the preserved
1. IT 160 Pentagonalantefix Fig. 4, PI. 20:a upper edge, and the slope of the tendrilindicate that
H. 0.13, W 0.077, Depth 0.066. the tendrils would have turned upward to connect
Fabric 2.5Y 8/2, slip 10YR 7/4, brown lOR 3/2, with a smalllotus at each side of the tile, that is, there
red 10R 4/4. would have been too much space at each side of the
PentagonalantefixofCorinthianclay,which is porous tile for the tendril to have ended in a simple volute.
with small to large inclusions. Only a portion of the The use of a small lotus at each side of the antefix
rightsidewith the upperand lower edges is preserved; is characteristicof this type of pentagonal antefix, a
the left and right sides are broken. The antefix was type that was among the last made before the outside
originallyattached to a combinationeaves/cover tile profile of the tile began to follow the contour of the
at the left. decoration.19
At the bottom is the right-handside of an inverted Several examples are known from Didyma and
lotus that occupied the center of the antefix. A tendril Miletos, and it was once believed that the type devel-
passesover the top of the lotus at the upperright. The oped in Ionia. Le Roy concluded that the origin was
relief is deep, and small bits of brown and red color Corinth,based on a comparisonof the style of the lo-
are preserved on the edges of the tendril and lotus tus with decorationon Corinthianpottery.20Two of
respectively.At the top center of the tile there would the threeexamplesLe Roy publishedfromDelphi are
have been a small palmette. described as having Corinthian fabric.21 The more
18 The tilesincludedin the cataloguewere selectedfor theirgood stateof preservation,except for the Archaic
examples which are all included. With the possible exception of IT 20 (3), the tiles are of Corinthian man-
ufacture. Context, listedat the end of each entry,includesthe fieldnotebook(NB)reference,the trenchnumber
(Tr.),and the lot number. Trenches can be located from the trench plans publishedin Gebhard and Hemans
1992, figs. 2 and 3, pp. 10-11. Lots are the units of excavation and storagefor ceramicsand other materials.
19 Nancy Cooper, in her descriptionof the development of the antefix, places the group in phase IV of
the attached antefixes;Cooper 1989, pp. 59-60, pl. 20. See also Billot 1990, pp. 110, 123.
20 Le Roy 1967, pp. 87-89.
21 Le Roy 1967, pp. 87-89, pis. 30 and 123, series 33 and 34. Series 33, antefixes
A.2 and A.8 are of
Corinthianfabric. AntefixA. 1, in series34, is not Corinthian. See also Van Buren 1926, p. 133 (no. 45), pl.89.
70 FREDERICK P. HEMANS
FPH, 1991
FIG. 4. Restored drawingof pentagonalantefix IT 160 (1), based in part on IT 732 (2). On this and the other
restoreddrawings,diagonal hatched lines indicate areas on the tile where paint is preservedor where
it can be restoredfrom fragmentsof paint. Closelyspacedlines indicateshadesof darkbrown, black, or
dark red; less closely spaced lines indicate lighter shades of brown or red. The outline surrounding
the areas where paint is indicatedshows the extent of the preservedsurfaceof the tile.
recent recognition of an example from Corinth by deep relief. The surface is abraded but small bits
Marie-FransoiseBillot as well as this example from of brown/red and red/purple are preserved in the
IsthmiasupportsLe Roy's conclusion.22 crevices of the relief. The back of the tile shows
Le Roy dates the Delphi seriesto 570-560. A date paring marks.
for IT 160 in the second quarterof the 6th century, Date: second quarterof the 6th cent.
perhaps as early as the 570's, seems likely (see p. 64 Context: none.
above).
3. IT 20 Pentagonalantefix Fig. 5, PI. 20:b
Date: second quarterof the 6th cent.
Context: NB 11 (1956) p. 4, Early Stadium Tr. 7; H. 0.098, W 0.105, Depth 0.04.
associatedfinds are Roman. Fabric 1OYR7/4 (burned), slip 1OYR8/3, pur-
ple/red lOR 4/3, red 2.5YR 4/6.
2. IT 732 Pentagonalantefix Fig. 4
Pentagonal antefix, possibly Argive. Clay is homo-
H. 0.08, W 0.113, Th. 0.068. geneous, with small to medium inclusions, not Co-
Fabric2.5Y 8/2, slip IOYR7/4, brown IOR 3/2, rinthian in appearance. The red appearance of the
red IOR4/2. fabricmay be a resultof laterburningof the tile. Most
Pentagonal antefix of Corinthian clay, of the same of the left half of the antefix is preserved,with a red
type as 1. Only a part of the bottom center of the invertedlotus at the center and a purple/red volute
tile is preserved, and contains a large portion of the at the left in deep relief. The figuresare surrounded
inverted lotus and a small piece of the left tendril in by a thin reliefline, possiblyan Argivefeature.23The
22 Billot (1990, p. 110, note 68) identifiesa piece from Isthmiaas a member of the same group, citing Isthmia
inventory number IT 20 (3). This is probably a confusion of inventory numbers since IT 160 (1) is clearly
the tile that belongs in the group. The Corinth piece is identifiedby Billot as FA 555.
23 Nancy Winter,personal communication.
ARCHITECTURAL TERRACOTTAS FROM THE SANCTUARY OF POSEIDON 71
- I
purple/red border is preservedon three sides. The of ca. 0.041. Surroundingthe tongues are a reserve
tile was burned after it was broken. band, which is in turn surroundedby black.
The straightsloping upper edges and the horizon- The guilloche is poorly preserved. Black/brown
tal lower edge of the tile indicate that this piece is circlesat the center are surroundedby an alternating
a late example of the pentagonal type and should red and black braid, bands in reserve, and a black
be dated to the second quarter of the 6th century, frame. The pattern repeats at intervalsof ca. 0.025.
perhapsclose to 550 B.C. Circles were incised to guide the painting of the
Date: second quarterof the 6th cent. guilloche.
Context: NB 6 (1954) p. 108, from outside the The cavetto profile does not closely resemble any
excavations. of the examples from Delphi published by Le Roy.
It is closest to examples late in his typology (type b,
series 19 and 23).25 Draftsmanshipon IA 663, how-
4. IA 663 Cavetto rakingsima Fig. 6, PI. 20:c
ever,is better than the examplesLe Roy places at the
H. 0.118, L. 0.131, W 0.061, slope 1:12.24 end of the Delphi series,and the Isthmiapiece makes
Fabric 1OYR8/3, slip 1OYR8/4, dark red 1OR extensive use of incised lines to guide the painting.
4/3, black/brown 1OYR4/1. Date: second quarterof the 6th cent.
Corinthianrakingsima with cavetto profilefrom the Context: NB 20 (1959) p. 356, Northwest Reser-
left side of the gable. Brokenand worn on both sides voir, Tr. GW 2; associatedfinds are 1st century after
and bottom. The lower edge was thin, overlapping Christ.
the edge of the roof member underneathit, as shown
5. IT 1075 Cavetto rakingsima
by a smallpreservedportion of the inside surface(see
section, Fig. 6). H. 0.097, L. 0.092, Th. 0.059.
The fascia has a tongue pattern above a single- Fabric lOYR 8/2, slip 5YR 8/2 (burned),reddish
braidguilloche, with tips of tongues on the top mold- brown 5YR 3/2 (burned).
ing. Tongues are dark red with a verticalband in re- Corinthianrakingsima fragmentwith cavettoprofile,
serveat the center. The patternrepeatsat an interval same type as 4. Broken and worn on both sides and
24 The slope measurementsfor the simas are for the sloping rear surfacewith respect to the top and bottom
edges of the tile. The actual slope of the roof would have been greater.
25 Le Roy 1967, pp. 57-59, pls. 98-100. IA 663 (4) has a ratioof upperfasciato the depth of the cavetto of ca.
1:1.3, closest to S.202.
72 FREDERICK P. HEMANS
II
IA663 10cm
FPH, 1991
bottom. Decoration is like that of 4, but colors are 0.042. Red braid has been charred to olive-brown
differentas the resultof burning after it was broken. color.
Date: second quarterof the 6th cent. The single-braidguilloche is typical of Corinthian
Context: none. eaves tiles from the late 7th centuryto approximately
the last quarterof the 6th.26
Date: probablysecond quarterof the 6th cent.
6. IA 4063 Eaves tile Fig. 7:a, P1.20:d
Context: none, found in tile pile, 1989.
H. 0.0545, L. 0.115, W 0.155.
Fabric 2.5Y 8/2, slip 2.5Y 7/4, red 2.5Y 4/4 7. IA 689 Raking sima Fig. 8, P1.20:e
(burnedto olive brown),black/brown 1OYR3/1. H. 0.116; L. 0.159, W 0.11, slope 1:20.
Corinthian eaves tile with single-braid guilloche. Fabric lOYR 8/3, slip lOYR 8/3, red lOR 5/8,
Smooth and slipped on top and fascia. Rough and black lOYR 3/ 1.
unfinishedon bottom, showingimpressionsfromhav- Corinthianrakingsima, "Megarian"type, from right
ing been dried on matting or ground but somewhat side ofgable, brokenat both sides. Verticaldowel hole
smoother for ca. 0.04 toward front edge. Surfaces at rear,probablyfor the attachmentof an acroterion
and broken edges are charred. at the corner of the gable (see section, Fig. 8).27
Incised outlines and compass points at center of Double chain of lotus flowers and seven-leaved
circles. Pattern repeats at interval of 0.033. Braid palmettosin red and blackon fasciaand ovolo. Upper
alternates black and red within circle diameter of molding broken away. Pattern repeats at interval of
26 An early example is Corinth FA 101, lOl a, with a guillocheon the eaves; Corinth
C, i, pp. 1, 57-58, fig. 1.
A date in the last quarter of the 7th century is proposed by Roebuck (1990, p. 51) based on comparison of
the antefix to examples from the pre-PeisistratidTelestereionat Eleusis,dated ca. 610 B.C. A date in the third
quarter of the 7th century is proposed by Williams (1978, pp. 347-348, note 13) based on the comparison
made between the decoration of Protocorinthianpottery and the roof tiles by Payne (1931, p. 252) and the
lower chronology of Protocorinthianpottery. The use of the guilloche gives way to the meander in the last
quarterof the 6th century as, for example, in the Treasuryof the Megariansat Olympia, ca.510 B.C.
27
Acroterionattachmentidentifiedby Nancy Winter.
ARCHITECTURAL TERRACOTFAS FROM THE SANCTUARY OF POSEIDON 73
a. IA 4063
Red |
b. IA687 Band
Bead
ad Red
Reel Band
d.~~~~~~
IT 92 / Ree l Band
FIG. 7. Restored drawingsof eaves tiles in chronologicalorder: a. IA 4063 (6); b. IA 687 (8); c. LA704 (13);
d. IT 729 (19); e. LA662 (21)
74 FREDERICK P. HEMANS
ca. 0.093. There is a red band 0.014 wide at front Similarto CorinthFS 1010 but with single petal at
edge of the soffit. center of lotus and a more rounded torus profile.30
The curveof the torusis pronounced,more so than Similar in design to Delphi Roof 45 but smaller in
most simas found at Corinth.28The Isthmiapiece is size.31Isthmiapieces IA 719 and IA 545 also belong
also very short, and yet the overalldepth of the piece to the end of the dark-on-lightseries. IA 719 is from
fromface to rearis comparableto examplesof greater the same buildingas IT 173. IA 545 is smallerin scale
height.29 The date is based on the comparisonwith and from another building. The profile of IT 173 is
the Treasuryof the Megariansat Olympia. quite close to Corinth FS 28, a light-on-darksima,32
Date: last quarter6th cent. indicatinga late date for the Isthmiapiece, ca.500 B.C.
Context: NB 20 (1959) p. 360, Northwest Reser- Date: ca. 500 B.C.
voir, Tr. GW 2. Found with IA 687 (8). Ceramic Context: none, from outside excavations.
context is 1st centuryafter Christ.
8. IA 687 Eaves tile Fig. 7:b, PL.20:f
CLASSICALAND HELLENISTIC TYPES
H. 0.035, L. 0.136, W 0.097.
Fabric 2.5YR 8/2, slip 2.5Y 8/2, red 5YR 5/6, POROSCOMBINATION
TAus
black 1OYR4/1.
The earliestmeander-patternedeaves tile at Isthmia 10. IA 4086 Poros combination tile Fig. 10
and the only tile with such a small height. Single- H. 0.113, W 0.194, L. 0.204.
stoppedmeander,with red used for the top horizontal Poros combination tile preserving nearly the full
line and hook and black at the bottom. Border of width of the cover and part of the attached pan, bro-
checkerboardis black, and interior squares are red. ken on all edges and along the bottom. The orig-
Ground is painted white. inal width of the cover was ca. 0.174. Lack of un-
The meander repeats at an interval of ca. 0.065. dercuttingbeneath cover suggests it was used at the
Both horizontal and vertical guide lines are visi- eaves. Stucco coating preserved across much of the
ble, and the painting is precise within those lines. cover. Front edge of the cover is darkfrom exposure
Checkerboardoutline is 0.023 wide and 0.019 high. to weather.
On the soffitthere is a plain red band ca. 0.03 wide. Date: probablymid-5th cent. (see p. 67 above).
Date: last quarter6th cent. Context: Lot 89-460, NB 70 (1989) p. 66, east
Context: Foundwith IA 689 (7). temenos, ceramic context is late 3rd or early 2nd
9. IT 173 Raking sima Fig. 9, P1.20:g cent. B.C.
H. 0.144, L. 0.26, W 0.083, slope 1:29. 11. IA 3642 Poros combination tile Fig. 11
Fabric 1OYR8/2, slip 1OYR7/4, red 1OR4/4, H. 0.095, W 0.102, L. 0.119.
black lOYR 3/1 Porostilepreservingone edge with partofthe cover
Corinthian raking sima, "Megarian"type, from left rising above it. Rear edge of the cover has been cut
side of gable, broken at right edge. Five-leavedpal- away at an angle of ca.20 degrees. Bottom frontedge
mette and lotus in red and black on fascia and torus, has cutting to overlap the edge of the roof or the tile
with pattern repeating at interval of 0.13. Upper below it. Right half of the underside(away from the
molding not preserved. Incised lines surroundlotus preservededge)has been cut away beneath the cover.
leaves. Date: probablymid-5th cent. (see p. 67 above).
0 0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~
00
co~~~~~~o
V CC
rmE
R 0 G
L ~~~~~~~~~~~
0) L .
LL ~ ~ ~ ~ ~~ ~ ~ ~~~~~~~~~~~~~~~~~L
co ci
00
(0)
<o
/ /~~~~~~~~~~~~~~~~~~~~~~~~~0 0
0
K --X-om t
<a~~~~~~~~~~~~~F
E
-
O 5<0
00
0~~~~
0~~~~~~~~~~0
00
OoT
- so
) -
L ED
a
O I~~~~~~~~~.n
S: U
f
TERRACOTTASFROMTHE SANCTUARYOF POSEIDON
ARCHITECTURAL 77
Context: Lot 89-555, NB 80 (1989)p.6 1, northeast acorn, and on eyes within tendrils. Outside edges are
temenos, ceramic context is 1st century after Christ. orange/red.
IA 3094 is similar in workmanshipand style, but
Roof1 simpler,to a group of terracottasfrom the workshop
The types of tiles describedhereafterhave been re- of Pheidias at Olympia dated to 430 B.C.33 The use
covered in largernumbersand are grouped by roofs. of elaboratefloral relief on the Olympia pieces does
Three types of palmetto antefixes have been recov- not appear in the Isthmia pieces. Thus a somewhat
ered, and while the decorativemotifscannot be com- earlierdate, towardthe middle of the century,is likely.
pared directlywith the simasand eaves tiles, the qual- Two smaller fragmentsof the same type are IS 262
ity of the draftsmanshipprovides a good indication and IT 29.
for matching these to the appropriateroofs. Thus Date: third quarterof the 5th cent.
the Roof 1 antefixes and eaves tiles have been as- Context: Lot 89-138, NB 68 (1989) p. 96, Tr. 89-
signed to the same building based on their excep- 19, north temenos, part of the sanctuarydestruction
tionally fine workmanship. No sima from this roof debristhat was deposited as road fill along the south
has been identified. Roof 1 eaves tiles displaycareful edge of the Classical/Hellenisticroad duringthe first
line work in the meander and checkerboardpatterns, quarterof the 2nd centuryB.C.
with incised lines visible. The checkerboardis par- 13. IA 704 Eaves tile, Roof 1 Fig. 7:c, PI. 21:b
ticularlydistinctive,with a fairlywide, purple border
surroundingcarefullymade squares. The soffithas a H. 0.048, L. 0.315, W. 0.216.
narrow bead-and-reeldesign, a feature that appears Fabric5YR 7/4, slip 2.5Y 8/4, red (soffit)2.5YR
to have chronological significance. On Roof 1 the 5/6, purple (fascia)IOR3/6, brown IOYR4/3.
design is ca. 0.02 wide, on Roof 2 ca. 0.05, and on Meander repeats at intervals of 0.084, lines 0.005-
Roof 3, ca. 0.06 (Fig. 7:c-e). The greatest number 0.006 wide; checkerboardhas purple outline 0.0225
of fragmentsfrom Roof 1 come from the north and wide and 0.0195 high with lines 0.003 wide. Bead
northeastpart of the temenos, in contextsof the early and reel on the soffit is 0.019 wide and repeats at
2nd century B.C. and 1stcentury after Christ. intervalsof 0.037; beads are ca. 0.0 19 wide and ca.
0.0 14 high; red band is 0.03 high.
IT 4061 has the same workmanship,but the bot-
12. IA 3094 Palmette antefix, Roof 1 PI. 21:a
tom has no bead-and-reeldesign.
H. 0.183, W. 0.16 1, Depth 0.057. Locationsof fragments:north and northwesttem-
Fabric 5YR 8/4, slip 10YR 8/4, red 2.5YR 6/4, enos IT 309, IT 704, IT 1025, IA 4061; northeast
purple IOR3/3, black 2.5YR 3/0. temenos IT 522.
Nearly complete, missingportionsat upperand lower Context for IA 704: NB 31 (1960) p. 744, north-
left. Cover behind antefix was ca. 0.165 wide, with west temenos, Tr. NW c.
sides ca. 0.05 high rising to ca. 0.08 at center.
14. IA 705 Eaves tile, Roof 1 PI. 21:c
Eleven-leaved palmette with spade-shapedheart.
S-shaped vine tendrils below the palmette enclose H. 0.0445,L.0.I1,W 0.11.
eyes at each end. At the outsidecenter of each tendril Eaves tile like IA 704 (13) but with lotus and palmette
is a small three-petaledlotus. At the bottom center on the soffit. Total depth of band on underside
there is a palmetto with nine or eleven leaves, but is 0.096 with lotus and palmette 0.07 high at the
the bottom edge is broken away. Right of center at front edge followed by narrow bead and reel and
bottom is an acorn. Relief is shallow. purple band. Lotus and nine-leaved palmette on
The fine detailsare extremelywell executed. Red/ brown/black ground with purple outline of leaves on
purple is applied around the heart, at the center of lotus;pattern repeats at intervalsof 0.101.
the heart, and in a row of dots along the edge of Context: NB 31 (1960) p. 662, Northwest Reser-
heart. Fine red dots and lines on lotus, lower half of voir, Tr. GW 5; associatedfinds are Roman.
34 Corinth I, iv, 83-88, pls. 19-21. The date of the South Stoa has been the subjectof much recent discussion,
and the importance of the date to the chronology of Classicaland Hellenistic roof tiles has been emphasized
by Le Roy 1990, pp. 38-39, note 29. Broneer(Corinth I, iv,pp. 156-157) places the constructiondate soon after
the battle of Chaironeia(338 B.C.).
35 Van Buren 1926, pp. 99-100, rakingcornices 117-123, nos. 201-207.
36 Van Buren 1926, p. 47, no. 206.
37 Le Roy 1967, pp. 153-154, 165-166, pls. 58 and 104 (S.34-38, 99, 108, 169, 170).
38 Bommelaer 1978, pp. 173-197.
z d5
Lo cD
m -
U
-l
>~~~~~~~~~~~~~~~~
0~~~~~~~~~~~~~~~~~~~0
V%
- a
LL
MM~~~~~~~~~
I l
cr
(L)0~~~~~~~~~~~~~~~~
0)0
00)~~~~~~~~~~~~~~~~~~~~~~~1
m cr
wL 004)
C)~ ~ ~ ~ ~~~~~00
NO~~~~~~~~0 ~ ~ ~ ~ ~ ~ 0
0)~~~~~~~
o -
0 ~ ~ ~~~~~
/~~~~~~~~~~
do 0
E<i E i
#fe
M1,i.~~~
E~~~~~~~
TERRACOTTASFROMTHE SANCTUARYOF POSEIDON
ARCHITECTURAL 81
Date: fourthquarterof the 4th cent. Lower right corner of eaves tile. Meander repeats
Context: Lot 89-358, NB 76 (1989) p. 46, Tr. 89- at an interval of 0.097. Horizontal incised guide-
39, north temenos, part of the sanctuarydestruction lines. The two preserved checkerboardsare 0.024
debris that was deposited as road fill along the south and 0.021 wide. Bead and reel on the soffitis ca. 0.05
edge of the Classical/Hellenisticroad duringthe first deep and repeats at intervalsof ca. 0.091; beads are
quarterof the 2nd centuryB.C. roughly circular, ca. 0.045 in diameter; purple/red
band is ca. 0.04 high.
17. IT 168/234 Palmette antefix, Roof 2 PL.22:a Another fragment, IT 350, has lotus and eleven-
H. 0.225, W. 0.17, Th. 0.065. leavedpalmettedecorationon the soffitwhich repeats
Fabric lOYR 7/4, slip lOYR 8/2, red (outside at an intervalof ca.0.125. Purpleis used for the heart
edges) lOR 4/4, red (face)2.5YR 5/4, black lOYR of the palette and the center of the lotus.
3/1. The paintingon the eaves tilesfromRoof 2 is much
Completeexcept for chip offthe top. Coverbehind more carelesslydone than on Roof 1. The meander
antefixhas sides 0.048 high, sloping to 0.08 at center. on the front face is generallymore widely spaced but
Eleven-leaved palmette with spade-shapedheart. variesfrom an intervalof 0.085 to 0.103. Round dots
Below are vine tendrilsflanking three inverted lotus or irregularblobs are often found in the checker-
leaves on the center axis. Five smaller leaves hang board rather than squares. Vertical guidelines are
from tendrils to each side of central lotus. Tendrils generally not used and the vertical painted lines of
have raised edges. Black ground with relief in buff the meander are often not perpendicularto the hor-
reserve;red applied to heart, between lower leaves, izontal. Severalfragmentshave relativelydeep claw-
at centers of upper tendrils,and on outside edges. chiseland flat-chiselmarkson the bottom side beyond
Locations of fragments: north temenos IT 249; the painted pattern.
northeasttemenosIT 481, IT 482; southwesttemenos Locations of fragments: north temenos IA 3095;
IT 304; outsidecentralsanctuaryor no contextIT 23, northeasttemenosIT 350; southwesttemenosIT 287;
IT 25, IT 26, IT 27, IT 169, IT 168/234 (17), IT 735, south temenos IT 270; southeast temenos IT 272;
IT 834, IT 980. outside central sanctuary or no context IA 4079,
Date: fourth quarter of the 4th cent., based on IT 1029, IT 1030.
South Stoa at Corinth. Date: fourthquarterof the 4th cent.
Context: fragments are concentrated outside the Context: NB 19 (1961) p. 67, Theater, channel
central sanctuary toward the south and southeast, trench.
associatedceramicsare Roman.
Roof3
18. IT269/278 Ridgepalmette,Roof2 Pl.22:b The workmanshipof the tiles grouped together as
H. 0.20, W. 0.16, Th. 0.105. Roof 3 is markedlyinferiorto that of Roof 1 and 2.
Fabric 5YR 8/3, slip lOYR 8/4, red/purple lOR As describedin the text, the tiles assigned to Roof 3
4/4, black 5YR 3/ 1. have no features that link them to a single building
Ridge palmette,brokenat center,top, and bottom. and are grouped together on the basis of their date.
Nine-leaved palmette has red on the spade-shaped
heart and on the outside edges. 20. IA 4033 Palmette antefix,Type 3 P1.22:d
Another fragment, IT 248, is from the north H. 0.268, W 0.194, Th. 0.072.
temenos. Fabric SYR 7/3 to lOYR 8/3, slip lOYR 8/3,
Context: IT 269 and 278, both from outside the red (outsideedges) 2.5YR 4/4, red (face) lOR 4/4,
excavations, 1958. black lOYR 3/ 1.
Cover tile with antefix, complete. Cover behind
19. IT 729 Eaves tile, Roof 2 Fig. 7:d, PI. 22:c antefix has sides 0.068 high, sloping to 0.09 high at
H. 0.054, max. H. 0.074 at right, L. 0.213, W. center.
0.238. Nine-leavedpalmettewith diamond-shapedheart.
Fabric7.5YR 8/4, slip 2.5Y 8/4, red (soffit)2.5YR Below are vine tendrils flanking inverted lotus with
4/6, black 5R 3/1. three leaves on center axis. Five smallerleaves hang
82 FREDERICKP.HEMANS
from tendrils to each side of central lotus. Relief west IT 662 (21); Palaimonion area IT 172, IT 271;
is higher than on Roofs 1 and 2. Tendrils are in no context IT 240, IT 1028.
flat relief. Brown/black ground with relief in buff Date: firsthalf of the 3rd cent.
reserve;red applied to heart, lotus, at centers of the Context: NB 20 (1959) p. 334, Northwest Reser-
upper tendrils,on upper side of hanging leaves, and voir, Tr. GW.
on outside edges.
Locations of fragments: north temenos IT 28,
22. IA 4071/4080 Raking sima Fig. 14, P1.22:f
IT 233, IT 4033 (20); northeast temenos IT 483,
IT 484, IT 485; southwest temenos IT 24; temple H. 0.123, W 0.268, Th. 0.18, three joining frag-
IT 17, IT 34; outside centralsanctuaryor no context ments.
IT 21, IT 104, IT 148, IT 285, IT 819. Upper taenia 0.0 15 high with bead and reel, re-
Date: probablyfirsthalf of the 3rd cent., based on serve on black ground, repeats at intervalsof 0.043;
the comparisonwith Roof 2 antefixes. reels are diamond shaped, beads 0.02 wide. Mold-
Context for IT 4033: Lot 89-154, NB 68 (1989) ing below upper taenia has egg and dart on purple
p. 182, north temenos, Tr. 68; joining piece from ground ca. 0.015 high; darts are diamond shaped,
outside the excavation. pattern repeatsat intervalsof 0.0225.
Lotus-and-palmettefrieze0.054 high on cymatium
21. IA 662 Eaves tile, Roof 3 Fig. 7:e, PI. 22:e
with purple highlightsat heart of palmette and along
H. 0.053, H. right edge 0.079, L. 0.204, W. 0.16. central edges of lotus leaves. Incised lines spaced
Fabric 5YR 8/3, slip 5YR 8/3, red 2.5YR 5/6, 0.117 at middles of palmettes.
black 5YR 3/1. Meander frieze on lower fasciarepeats at intervals
Meander repeats at interval of ca. 0.08, lines ca. of 0.075, checkerboard outline is 0.017 wide and
0.0045 wide; checkerboardoutline 0.013 wide and 0.013 high, with round blobs for squares.
0.016 high, entirely in red with lines ca. 0.001 wide. Soffit has bead and reel 0.038 deep, with circles
Bead and reel on soffit 0.06 high and repeats at in- 0.033 in diameter. Purple band is preserved 0.029
terval of ca. 0.09 with circularbeads ca. 0.057 in di- wide but is broken away.
ameter;red band is ca. 0.043 high. Other fragments:IA 4039 fromthe northtemenos;
Eaves tiles from Roof 3 are carelesslymade. They IT 808 from the later stadium; IT 425 from the
are distinguishedfrom Roof 2 tilesby the painted red theater.
checkerboardoutline and by the difference in size Context: Lot 89-345, NB 76 (1989) pp. 51, 54,
between the diametersof the beads on the soffits. north temenos, Trs. 89-39 and 89-41; Lot 89-358,
Locations of fragments: north temenos IT 11, NB 81 (1989)p.21, north temenos, Tr. 89-40; Lot 89-
IT 247, IT 51/161; northeasttemenosIT 411; north- 343, NB 75 (1989) p. 75, Tr. 89-37.
BIBLIOGRAPHY
Billot, M.-F. 1990. "Terrescuites architecturalesd'Argoset d'Epidaure. Notes de typologie et d'histoire,"in
Hesperia59, pp. 95-139
Bommelaer,J.-F. 1978. "Simaset gargouillesclassiquesde Delphes,"BCH 102, pp. 173-197
Broneer, 0. 1976. "The Isthmian Sanctuary of Poseidon," in KeueForschungen in griechischen
Heiligtimern,
U.Jantzen, ed., Tubingen, pp. 39-62
Cooper, N. K. 1989. The Development of RoofRevetment (Studiesin Mediterranean
in the Peloponnese Archaeology.
Pocketbook 88), Jonsered
Corinth I, iv = 0. Broneer, TheSouthStoa(Corinth I, iv), Princeton 1954
Corinth IV, i = I. Thallon-Hill and L. S. King, Decorated Architectural (Corinth
Terracottas C, i), Cambridge, Mass.
1929
Gebhard,E. R., and F. P.Hemans. 1992. "Universityof Chicago Excavationsat Isthmia, 1989: I," Hesperia 61,
pp. 1-77
Hemans, F. P. 1989. "The Archaic Roof Tiles at Isthmia:A Re-examination,"Hesperia58, pp. 251-266
Hesperia59 = Proceedings of the First InternationalConference on ArchaicGreekArchitectural Terracottas,
Athens,
December 2-4, 1988 (Hesperia 59, 1990), N. A. Winter,ed.
Hubner, G. 1973. "Dachterrakottenaus dem Kerameikosvon Athen,"AM 88, pp. 67-143
IstkmiaII = 0. Broneer, Topography andArchitecture(Isthmia II), Princeton 1973
83
FREDERICKP. HEMANS
WIcHrrASTATE UNIVERSrrY
School of Art and Design
1845 Fairmount
Wichita, KS 67260
PLATE 20
a. Pentagonalantefix & 3 _ i
b. PentagonalantefixIT 20 (3)
~~~ta -
IA 689 (7)
e. Rakingsima
-
A 3
-~~~~~ uc- . :*.;~~~~~-:~~
-x
i;Se~~~S
g. Racing simnaif 173 (9)
E. P. HnwAzs: Gain ARcrnnc'ruitu. TERuucorrAs FROM
THESANCTUAnY
OFPOSEIDON
PLATE 21
I
I
r ~~~~~~~~~~~~~r~~~~7
I
Forthe late 4th-centurybuildingprogram,see Stephen G. Miller 1990, pp. 65-67; see also pp. 23, 43, and
61-62.
2 Although official excavations by the University of Californiaat Berkeleyhave not taken place for some
years, a varietyof workin removingroads throughthe archaeologicalzone has continued. The last and largest
of these took place in 1989 and 1990 when the Ministry of Public Works officiallyclosed the highway that
had bisected the ancient stadium and the Ministryof Cultureasked that we undertakethe task of completing
the work in the stadium and providing for drainage to protect the antiquities. To be thanked particularly
for supportof this effort are the Ephoreia of Antiquitiesin Nauplion and especiallyFani Pachyianne (Ephor),
Elsi Spathari, and Zoe Aslamatzidou. It was the legal advice of Georgios Pachyianneswhich resulted in the
formula allowing this road to be closed after fifteen years of discussionand futile attempts. I dedicate this
study to him.
3 Similarstructures,althoughnot heretoforeidentifiedas such, existat Epidaurosand Olympia and probably
elsewhere. For the building at Epidauros,see Kavvadias 1902, p. 83 and pl. A, Building A, and Tomlinson
1983, p. 42, fig. 5, buildingno. 18. Forthe situationat Olympia, see OlForsch
XIV, p. 84, where the court behind
the Echo Stoa is recognizedin a most tentativeway as possibly"derWasch-,Umkleide-, und Warteplatzfur die
Athleten.... "
86 STEPHENG. MILLER
7/PASSAGEWAY CUT
THROUGH BEDROCK
STADIUM
ENTRANCE
TUNNEL
...
1Dm.
M.
11 - 11-91
of its south side (Fig. 1).4The centralareaof the buildingwas hypaethraland was surrounded
by Doric columns on three sides. These colonnades supporteda roof, from which a great
number of tiles have been recovered.
Unlike most ofNemea, where EarlyChristianfarminghas dug throughand disturbedthe
more ancient material,5the destructiondebrisof the apodytrion was essentiallyundisturbed,
and many tileswere found smashedwhere they had fallen. It was thuspossibleto reconstruct
several of them and to make specific observationsabout these tiles as a group. The tiles
are all Laconian,6 specifically,Laconian pan tiles that work in the typical way, the wider
end at the top with a groove and a thickenededge on the concave upper surfaceservingto
preventwater from runningunder the next tile. The upper,or weather,surfaceis noticeably
smoother than the convex lower surfaceand always bears traces of a thin wash. As in the
example chosen (Fig.2 and PI. 23:b, c),7 they are all about 0.96 m. in length with a width at
the upper end of about 0.49 m. and at the lower of about 0.42 m. They average about 15
kilogramsin weight, although the variation is as much as a full kilogram above or below
that average.
At this point a slight digression is necessary to make two points observed during the
excavations. First, very few traces of carbon and only two iron nails were found in the
destructiondebris. Since metallic objectsare generallywell preservedby the soil of Nemea,
this must mean that much of or all the wood of the buildinghad been removed. This is not
particularlysurprisingsince we know that such removal of wood from a building occurred
and could even be a recognized part of a lease.8 Nonetheless, it is strikingto document
such activityin the archaeologicalrecord, especiallysince we can estimate that the quantity
of wood involvedwas about 13 cubic meters.9 A restoreddrawingof the timberingnecessary
indicatesthe minimum amount of wood in the building (Fig.3).
4 The precise location and nature of the doorwayscannot be establishedbecause the walls of the building
have been completely robbed out in these areas.
5 Forsome particularinstancesof such disturbance,seeJ.emea I, pp.5-6, 19-20, 23 (note 60), 25, 33,53,57,
70, 79, 83, 87, 91, 93, 95, 192, 220, 228, 232 (note 639), 234, 236, 239-243 (esp.note 661), 250, 257, 301-302.
6 Although some fragmentsof Corinthiantileswere scatteredaround the area, the actual destructiondebris
of the building produced only about two dozen Corinthian pieces, which may well have been used in the
buildingfor some purpose other than the roof.
7 AT 316. The variationsin size are slight, although the texture and color of the clay have wide variations
that are, however,of no apparentsignificance.These range from a friablegreenishwhite clay (2.5Y 8/2 on the
Munsell scale)to a hard light-redclay (IOR8/3 on the Munsell scale),alwayswith heavy inclusions. The wash
on the weathersurfacealso variesfrom reddishbrownto nearlyblack,again withoutany seeming correlationto
differencesin stamps.
8 Forexample, in IG II2 2499 of 306/5 B.C. the shrineof Egretesis to be leased to Diognetos, who at the end
of his ten-year lease is to be allowed to take with him the woodwork (at {Ua), the tiling (t6 xppoiov), and
the doors and windowswith their frames(t& Oupc'apcat) if he fulfillsthe terms of his lease;if he failsto fulfillthe
lease, these items are forfeited.
9 The following calculations must be recognized as highly theoretical, except for those of the weight of
the tiling. Since the weight of the pan tile averages 15 kilogramsand the weight of a cover tile can be estimated
at about 11 kilograms,the weight per square meter of the tiling can be estimatedat about 58 kilogramswhen
dry and potentiallyat about 72 or 73 kilogramswhen wet.
The nature of the roof-timbersystem cannot be establishedso firmly,however, except for the existence
of a wood epistyledirectlyon top of the columns and measuringabout 0.30 m. in width (seen most clearlyfrom
88 STEPHEN G. MILLER
the burnt area on the top of the Doric capital, A 374). I have restored a square (in section) beam for this
element. Clearly sleeper beams were needed on the top of the walls to accept rafters. I have restoredthese as
about 0.22 m. on a side because the mud brickof the wallswas of that height and twice that width. Raftershave
been restored0.50 m. apart on center and serveboth to markthe joint between pan tiles and to supportthem.
Using an average weight-bearingwood strength, rafters0.10 m. wide and 0.15 m. high would be sufficient
to support the wet weight of the tiles plus a meter of snow. I have, however, restored0.15 m. square rafters,
partlybecause it is clear that the load-bearingstrengthof wood was frequentlyunderestimatedin antiquityand
partly in order to provide a sufficientresting surface for the tiles. I thank Georgios Stephanopoulos for his
help in calculatingthe load-bearingcapabilitiesof the proposed roof system.
10 The coverage of the pan tile is about 0.45 m. sq. and the total area to be roofed in the apodyterion
is about
154 m. sq. This should mean that some 342 pan tiles were needed to cover the building. (An actual count
based on the restoreddrawing,Figure 3, resultsin a total of 340 pan tiles.) The total weight of the pan tiles on
the building would then have been about 5,130 kg. or, when wet, perhaps about 6,500 kg. The total of all
the pan tiles that have actually been recovered from the building was 3,131 kg.; they were weighed after a
rain when they were stillpartiallywet.
I See Stephen G. Miller 1988a, p. 138.
12 The total original weight of the cover tiles for the whole building can be estimated at about 3,322 kg.
The total weight of all cover-tilefragmentsrecoveredis about 25 kg.
13 Potterylots SACWAY56-59.
14 Potterylots SACWAY44, 45, 62, 53, 60-65, 69, 74-76, and 79.
0,49
0,425 0 10- .5
A A
J] L
II~~~~~~~~~~~~~~~~~~~~~~
a. AT 324
b. AT 346 + 325
c. AT 284
d. AT 258
. AT 66 + 174
f. AT 178
P.~~~~~~~~~~f AT178
0 F 05 f. M
the omegas.16 There are preserved 12 differentstamps of the first variant and 44 of the
second, or a total of 56 for this type.17
Stamp Type 2 (Fig.4:c and PI. 24:c) is much largerthan the first and bears in relief the
word E2KAEIOE, that is, the genitiveof Sokles. Despite the differencein name, we are not
dealingwith a differentperson, for Soklesis generallyacceptedto be a shortenedform or a pet
name or nicknamefor Sosikles,which mustbe the fullpropername of our man.18 The debris
ofthe apodyterionat Nemea has produced40 differentexamplesof this type (seenote 17 above).
In addition there are 2 extremelyfragmentarystampsthat cannot be identifiedby type,
for a total of 98 stamps that may representnearly a third of all the pan tiles originallyon
the roof. Not all the tiles were provided with stamps, for several complete examples lack
them. These exhibit a groove sweeping acrossthe convex undersideof the tile and executed
by a fingerwhile the clay was wet (P1.24:d). There is also one example of a cat'spaw imprint
along the side of the concave weathersurface,but this is perhapsdue to an accidentwhile the
tile was being made (PI.24:e).
All the Sosikles-Soklesstamps are located on the convex undersurfacenear the upper,
wider end. They were thus not exposed to the weather and would have been visible only
from below once they were in place.19 The same is true of modern tiles, the stampsof which
16 I have been unable to find any good parallelsfor this type of omega with the horizontalbars up on the sides
ratherthan at the base of the letter. Of course, most extant examples are in stone.
17 Some of the stamps are representedby very small fragments,some of which might have come from the
same tile. Therefore, the total number may be misleadinglyhigh, but the proportionsof the types must be
about the same, regardlessof the absolutenumbers.
18 See, for example, G. E. Benseler'srevisededition of Pape 1875, p. 1469: E(xX~q, iouq, m. = Ecsatxxiq,
Wehrenbrecht(d.i. in der Abwehr od. dem Schussenglanzend);see also Richardson 1902, p. 219: "The name
Socles, a 'Koseform'for Sosicles, is common enough, and affordsno particularinterest."
We shouldnote, however,that both formsappearfor differentmen on the same inscriptionfrom Thespiai,
SEG XXIII 271, lines 28 and 76. Presumably,the individual could chose which form he wanted used for
his name.
19 This is unlikeboth the other Nemean stampsand the normal practicethat appearon the weather surface;
cf. Felsch 1990, p. 301: ". . . tiles of Hellenistic and later date, which are always stamped on the upper face."
Since it is the upper face of the tile that is pressedinto the mold, Richardson(1902, p. 221) suggestedthat the
stamp was affixed to the mold so that the tile was shaped and impressedin a single operation. As Richardson
noted, there are problemswith such an interpretationbecause the stampsare not all in the same position on the
tiles; note, for example, the two Nemean stamps cited below, AT 66 and AT 174 (PI. 25:b), which are not
at the same distance from the upper edge of the tile. Richardson supposed that the stamp might have been
slightlyloose on the mold. One might also suppose that several molds, each with its own stamp, were used
duringproduction. In any event, and if such an interpretationis correct, the Sosikles-Soklestiles might have
alreadybeen in the processof manufacturebefore the stampswere preparedand the latterhad to be pressedon
the back side of the tile while it was still in the mold.
One should mention at this point the debate about the meaning of the phrasetintov 6xtvov xepav4bcav,
which appearsin Delian inscriptionsLD442B.172 and 457.22. The tiu5noq has been understoodas a wooden
mold for the manufactureof tiles by, among others, Richardson 1902, p. 221; Grace 1935, p. 427, note 4;
Bundgaard 1957, p. 219, note 14; and the undersigned. The -cu'noq has been understoodas a wooden stamp
used to produce impressions(or, in the present case, inscriptions)in the wet clay of tiles by, among others,
Wace 1906-1907, p. 17, note 3; Orlandos 1955, p. 113; and Hubner 1973, pp. 86-87. Perhaps the most
curious reference is that in Orlandos and Travlos 1986, p. 255, where -cntoq is defined as a stamp for roof
tiles but also as a mold for bricks.
SOSIKLES
ANDTHE4TH-CENTURY
BUILDING
PROGRAM 93
commonly have the name and telephone number of the manufactureron the underside.
Depending on the nature of the roof construction,the modern examples are frequentlynot
visible, nor were the ancient ones at Nemea.
These considerationstake us to the question of the nature of the Sosikles-Soklesstamps
and the identity of Sosikleshimself. There are at Nemea three other types of stamps that
can be associated with the late 4th-centurybuilding program; all are from the Sanctuary
of Zeus and none from the Stadium. The first of these is represented by six fragments,
all broken at some point along the right side and all on the concave or weather side of
the tile.20 The inscription reads AAMOIO-, which is to be understood as a form of
danosios (Fig. 4:d and PI. 25:a).21 Clearly these tiles are marked as public property and
stand in apparent marked contrast to the personal name attested on the Sosikles-Sokles
examples.
The second type of stamp is representedby two examples, both broken along the left
side and each found in a well south of the Temple of Zeus (Fig.4:e and PI. 25:b).22They bear
the genitiveadjectivalform NEMEIOT with a ligatureat the right end of the stamp. It takes
little imagination to restore [AtbL]Neiiudouand to see these tiles also as public or official
property of the sanctuary.23In contrast to the Sosikles-Soklesseries, these stamps are on
the concave weathersurfaceof Laconianpan tiles, as shouldbe expected (see note 19 above).
The final type of stamp is representedby a single example discovered in 1962 above
the Xenon on the south side of the Sanctuaryof Zeus (Fig.4:fand PI.25:c).24The impression
is on the weatherside of a Laconianpan tile and preservesonly the firstthree lettersNEM-.
Again, this is clearlya statementof officialpropertyof the sanctuary.
Thus, at first glance, the Sosikles-Soklesseries appearsto fail to correspondwith other
stampsat Nemea in its lack of an officialstatus. One explanationmight be that the tiles for
the apodyterionwere not officialbecause that structure,at some remove from the Sanctuaryof
Zeus, was itself of a statusdifferentfrom that of buildingsin the sanctuary.This explanation
cannot hold, however,because three examples of the Sosiklesseries have been found in the
Sanctuary of Zeus itself.25 Rather, the question of the purpose of the stamps needs to be
For a modern example of what I take to be a ut-tos (uwoq xepajEbv, see Rostokerand Gebhard 1981,
pp. 220-222.
20 AT 17, AT 195, AT 196, AT 248, AT 252, AT 258. The firstwas discoveredlong ago and no recordof its
proveniencehas been preserved,but the last five were clusteredin SectionsP 14 and P 15 near the Rectangular
Buildingeast of the Temple of Zeus. See Stella G. Miller 1984, pp. 185 and 190.
21 By analogywith the examplesfrom the ArgiveHeraion given below (see note 34 below),we shouldperhaps
understandBaV6Lo[LAtLo Neic0ou], but in the absence of the ending of the word on the Nemean tiles, such a
suggestioncannot be pressed.
22 See NemeaI, p. 303.
23
The significanceof the ligatureis not so immediatelyapparent,even though it clearlyconsistsof at least a
pi, a chi, and a rho.
24
See NemeaI, p. 285.
25 AT 157 (type lb), and AT 188 and AT 189 (type la); see Stella G. Miller 1984, p. 185, where the reading
of the inscriptionshouldbe correctedto EflEIKAEOE. These three tiles come from preciselythe same area as
do the damosios-type tiles mentioned above, and while the fragmentsof the one type cannot come from the same
tile as the fragments of the other because of the differencesin fabric, both types might have been used on
the same building. For the significanceof this possibility,see below on the tiles from the Argive Heraion.
94 STEPHENG. MILLER
center. The purpose of this groove is clearly to prevent water from running up under the
overlappingtile above.
On its concave, or weather,surface,again unlikethe Nemea examples, are two stamps.
One is near the top of the tile, 0.044 m. from the edge on the right end, 0.040 m.
on the left (PI. 27:a, b).33 It measures 0.022 by 0.170 m. and bears the relief letters:
E2KAHEAPXITEKThN. The other stamp is near the bottom of the tile, 0.065 m. from
the edge and measuring 0.022 by 0.162 m. (P1. 27:a, c). It bears in relief the letters:
AAMOIOHPA!. As pointed out long ago by Richardson,34 this stands for Ba&6aYLoL
(x~paCXoL) 'Hpcaz. We should note that the tiles are not merely public property,nor merely
the property of Hera (that is, of the sanctuary),but the public property of Hera. In other
words, the tiles belong to Argos, which controlsthe sanctuary,and they are intended to be
used specificallyfor the sanctuary. This means, in turn, that Sokles is acting as architect
in an official capacity on behalf of the city as well as the sanctuary. The letter forms of
the stamps indicated to Richardsonthat the date of the tile should be well along in the 4th
century.35A comparison of the omega of the stamp with dated Attic inscriptionssuggests
that Richardsonwas essentiallycorrect.36
Is the architectSokles of the Argive Heraion to be identifiedwith the Sosikles-Soklesof
Nemea? I believe so, even though absoluteproof is not possible. Certainly,two differentmen
stampingthe same name on tiles at two sitesso close as Nemea and the ArgiveHeraion at the
same date would be an extraordinarycoincidence. Such a coincidence becomes even more
incrediblewhen one remembersthat both sites were under the control of Argos and were
closely linked in that city's administration. For example, many theorodokoi were appointed
both for the festival of Nemean Zeus and for that of Hera Argeia.37 The same individual
frequentlyserved as the agonotletes for both festivals.38And the same men were Hellanodikai
for both festivals.39The date when this close linkagebegan is not known, but it was certainly
in place by 323 B.C.40 It would then hardly be surprisingif the same man, Sosikles-Sokles,
were the officialarchitekton of the two festivalsites.
Sosikles-Soklesis, I believe, somethinglikethe "cityarchitect"envisionedby J. J. Coulton
as a development of the Hellenisticperiod, although Coulton sees the practice beginning in
33 But note that this stampwas almost certainlycovered by the next tile above when in place on the roof. So
both at Nemea and at the Heraion the personalname was not easilyvisible, if visible at all, on the building.
34 Richardson 1902, p. 217.
35 Richardson 1902, p. 219.
36 Such evidence for dating is, of course, extremely tenuous but does give a general range. The closest
parallelseems to me to be the decree dated to 307/6 B.C. and presentedin Kirchner 1935, no. 67.
37 Fora list of examples,see Amandry 1980, p. 245, note 80; to this add Pierartand Thalmann 1980, p. 256;
see also Charneaux 1983, p. 266, who notes that only one of all the conferralsby Argos of theorodokia on an
individualwas not for both festivals;see also Stephen G. Miller 1988b, pp. 147-163.
`8 As, for example, Philip V of Macedon in 209 B.C. (Livy27.30-31).
39 As seen in a decree in honor of Alexander of Sikyon, Argos Museum inv. E 125, lines 16-18. First
published by Vollgraff 1916, pp. 64-71. For the various dates proposed for this decree, see Amandry 1980,
p. 226, note 30.
Once the Nemean Games moved to Argos, both they and the festivalsof Hera were celebrated in the
same place perhaps alreadyin the 3rd centuryB.C. but certainlyby the time of Pausanias(see 2.24.2).
from Nemea; see Stephen G. Miller 1988b, p. 162.
40 This is the date of the list of theorodokoi
96 STEPHEN G. MILLER
Athens by 337/6 B.C. at the latest.41Among the many examplescited by Coulton, we might
note Diognetos who was employed by Rhodes in the late 4th century B.C. on an annual
retainer until he lost his position to a rival, Kallias of Arados.42 Perhaps even closer to
the position of Sosiklesat Argos is the Athenian "architectin charge of the sanctuaries"of
the same period.43 Indeed, it seems likely that such an official existed within the Argive
administrativesystem even as at Athens and that it was Sosikleswho filled this position for
a time. This official architect emerges from our new knowledge of Sosikles and from a
fragment of a stamped tile discoveredin the area of the West Baths in Argos itself giving
one Menon as architect.'4 The impressionon this tile is from the same stamp as one found
more than eightyyears ago at the Sanctuaryof PythianApollo at Argos.45These two stamps
indicate that Menon, like Sosikles,was responsiblefor officialcity work and was a successor
of Sosikles-Soklesas the "cityarchitect"ofArgos. This positionapparentlyhad responsibility
for all projectsof the city or at least for all projectsconcerning sanctuaries,not only for the
sanctuariesof Zeus and of Hera. The extent to which these "cityarchitects"were concerned
with problems of design is not clear, but that they were directly responsiblefor execution
of buildings and other programs of constructionand maintenance is clear. The modern
equivalentmight be somethinglike "Directorof PublicWorks".46
Once we have so identifiedSosikles,we can understandthat the stampson the Nemean
tiles with his name are analogous to and not differentfrom those with the official stamp
of Nemea. In his official capacity as architectfor part of the building program at Nemea,
Sosiklesorderedtiles which were given identifyinglabels at the factory.47The stampsset the
41 Coulton 1977, p. 29.
42
Vitruvius 10.16.3-8. Forone of the dutiesof the Rhodian architect,see SIGC3 581.97-98, a treatybetween
Rhodes and Hierapytna dated to the period 200-197 B.C., the inscribingof which falls within the architect's
realm.
43 IG II2 840.13-14; 841.14-15; 842.2. These references correct those given by Coulton 1977, p. 165,
note 90. Another similarofficialarchitectwho was part of the city administrationis perhaps to be understood
in "the architectelected by the Demos" who was responsiblefor supervisionof the repairsto walls of Athens
and Piraeus in 307/6 B.C., as attested by IG II2 463.6. To Coulton's referencesto official architectsshould
be added those in Shear 1978, pp. 54-56 and in Bousquet 1985, p. 717 and note 6.
44 This tile, which reads MENQNAPXITE[KTQ(N)], was discoveredby Pierre Robert in an apparently
very late or even modern context. I am grateful to him and to Pierre Aupert for permission to mention it
here and to Marie-FransoiseBillot for generouslysharingthis informationwith me.
45 Vollgraff 1909, p. 448, no. 12. It reads A]PXITEKTQ(N); the final nu seems to have been left off the
stamp or the stamp was not fully impressedin the damp clay. This stamp (on both tiles mentioned) bears
some affinitieswith those from the Argive Heraion in size and form of letters and should be close to them
in date, but the broken cross-barof the alpha suggestsa later date for it and for Menon.
Although no direct connection can be establishedbetween the "city architect", the stamped tiles just
mentioned, and the Sanctuaryof Apollo, we might note that it was during the second half of the 4th century
B.C. that repairsand remodelingof the sanctuaryare epigraphicallyrecorded;Vollgraff1956, p. 110. Therefore
a "city architect"or an "architectin charge of the sanctuaries"was certainly needed by the Argives at this
period.
46 It must be rememberedthat, althoughI believe that the man is the same, the form of the stamps from the
Argive Heraion and from Argos and that of the stamps from Nemea is different. Different interpretations
are possible, such as supposing that Sosildesundertookhis work at Nemea at some point in his career when
he was not the officialarchitectof Argos, but none are capable of proof.
47 Close analysis, which has not yet been possible, might show that the tiles of Sosikleswere made in the
complex of three kilns in the Sanctuary of Zeus. Three kilns have been recovered, but there was probably
ANDTHE4TH-CENTURY
SOSIKLES BUILDING
PROGRAM 97
tiles apart, at the factory and on the constructionsite, for the work of Sosiklesand enabled
him to maintain some sort of inventory control. Even if our knowledge of Sosikles is still
quite limited, he can at least take his place in the all too short list of ancient architectsknown
to us.48 Indeed, it might even be possible to credit him with the design of the apodyterion.
Finally,we may now ask, even if I cannot provide an answer,whether every tile that
bears the name of a man in the genitive (and these become increasinglycommon in the
later Hellenistic and Roman periods49)does not give us the name of the ancient architekton
of the building from which the tile came.
BIBLIOGRAPHY
a fourththat was destroyedin the 1920'sby a well constructedat that time. Two of the kilnswere rectangularin
plan, the third circular,and one supposes that they were for the manufactureof Corinthian and Laconian
tiles, respectively,with the shape of the tile in section reflected in the plan of the kiln where it was made;
see Perreault 1990, p. 208, note 13. Although one of the rectangularkilns was put out of operation by the
constructionof the Xenon in the last thirdof the 4th centuryB.C., the circularkiln continued in operationafter
the constructionof the Xenon and apparentlyinto the earlypart of the 3rd century.It might well have been the
spot of manufactureof Sosikles'tiles and was certainlyan importantpart of the buildingprogramof the period.
For the kilns, see Hill 1966, pp. 20 and 46; Stephen G. Miller 1975, pp. 162-165; Stephen G. Miller
1976, pp. 186-189; Stephen G. Miller 1978, pp. 80-81.
48 Although I cannot prove, or even suggest,that the same man is involved, attentionshould be called to the
funeral stele of a Sosikles found at Phlious, a scant four kilometersfrom Nemea, and dated by its publisher
to the 3rd centuryB.C. See Scranton 1936, p. 246, no. 16.
49 See Orlandos 1955, pp. 115-116 for a list that Habner (1976, p. 180, note 29) has updated with more
recent discoveries. Note also the interpretationby Camp (1989, p. 51) of the ligatureAP or APX as architect
rather than as archon in associationwith names in the genitive (Dionysios and Diodoros) on roof tiles of the
Odeion in the Athenian Agora of the 2nd centuryafter Christ.
One other type of individualis documentedas puttinghis name on tiles, at least sometimesin the genitive.
who seems to be a kind of subcontractorresponsibleto the architekton
This is the ergones, for some specifictaskin
the constructionproject. See Orlandos and Travlos 1986, pp. 117-118. It seems, however, that the name is
usually accompanied by the word ergones as a specifier,perhaps to distinguishhim from the architekton. See
Orlandos 1955, p. 115, note 10.
98 STEPHENG. MILL
STEPiNm G. MILLER
UNIVERSrrY OF CALIFoRNIA AT BERxELEY
Department of Classics
Berkeley,CA 94720
PLATE 23
.*wj~. . I. _ ci
LLA'
b. Laconianpan tile, AT 316, upper surface c. Laconianpan tie, AT 316, lower surface
Bar:,,'
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Al
-21A7 ~~
ji~~i
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
0~~~~~~"'
PLATE 26
Qto
J '
~~~~~~~~~~~~~~
lip?
,,j (./ 0
j ' -~ ~~~~
A Q4.... . .... ..
PLATE 27
a. Stamps of NM 10736
L~~~~~~~~~~~~~~~k~~~~~L I
c. Detail of lower
stam7of NM 10736
DIE DACHTERRAKOTTENDES ARTEMISTEMPELS
VOM APOLLON-HEILJGTUMIN AGINA
(PLiTxs28, 29)
1\TACH PAUSANIAS (2,30,1) gab es in der Stadt Agina-im Bereich des sog. Kolon-
INTnahugels, benannt nach der einzigen heute noch aufrechtstehendenSaiule-drei nahe
beieinanderstehendeTempel des Apollon, der Artemisund des Dionysos; und bisherkannte
man nur die genaue Lage des Apollontempels. Da die alten Grabungennur einen Teil des
Heiligtums ostlich vor dem Apollontempel erfaBt hatten,I war zu hoffen, daB von den
bei Pausanias genannten Tempeln im unausgegrabenenAbschnitt noch Reste zu finden
waren. Zu Beginn der Grabungen unter H. Walter kam siid6stlich vom Apollontempel,
noch innerhalbder Temenosmauer,ein in zwei Teile zerfallenesFundamentzum Vorschein
(Fig. 1 punktiert). Spater erst wurde ostlich davon ein dritter Fundamentteil aufgedeckt
(Fig. 1 strichliert),der uiberdie Situation in diesem Bereich des Heiligtums aufklarenhalf:2
Nicht ganz so hoch erhalten wie die beiden anderen Teile, daftir wesentlich tiefer in den
Boden gesetzt, folgt das Fundamenthier dem nach Osten zunehmend abfallendenGelande.
Mit der Ecke eines in Streifenformverlegten Unterbaus wurde jetzt der GrundriBeines
kleinen, rechteckigen Gebaudes erkennbar. Es gibt gute Grunde in ihm den Tempel der
Artemis zu erkennen, liegt es doch nahe, ihr Zuhause in der Nahe des Brudertempels
zu suchen, gleichwie der Standort des Dionysostempels nahe beim Theater, das dem
Heiligtum im Osten gegenuiberlag,zu finden sein sollte. Inzwischen ist das Heiligtum
ostlich vom Apollontempel auch soweit ergraben, daB hier die Existenz eines weiteren
Tempels innerhalb der Temenosmauern mit groBer Wahrscheinlichkeitausgeschlossen
werden kann. Die Reihenfolge, in der Pausaniasdie Tempel und seine Inhaber aufzahlt,
wurde sich damit bestatigen.
Die Fundamentresteumschreibenein Rechteckvon ca. 7,6 x 14,8 m, seine Langsachse
ist um ca. 260 nach Suidenabgedreht. Es sprichtmanches dafur,daB der Tempel nicht wie
bfilichvon Osten, sondern von Westen her zu betretenwar. Auch die Tempelform ist nicht
eindeutig festzulegen. In Frage kommen ein Tempel in antis,ein Prostylos, aber auch ein
einfacherOikosbauohne Saulenstellungware denkbar,wenngleich auch unwahrscheinlich.
Doch soll weiter nicht die Rede sein vom Fundamentoder zuweisbarenArchitekturgliedern,
allein das Dach mit seinen verziertenTonziegeln steht hier zur Diskussion.
Bei der Zuweisungder einzelnen Terrakottafragmentezum Artemistempelgaben zwei
Grande den Ausschlag: mehrere Teile der Ziegel, Simen und Antefixe wurden direkt
beim Fundament oder in dessen naherem Umfeld gefunden, und, nach dem Ausweis der
Fundamentrestesowie der Untersuchungallerim HeiligtumaufgefundenenBaugliederund
Ziegel auf mogliche Zugehorigkeit,wurde dieser Tempel erst in klassischerZeit errichtet.
I
Wolters 1925b, S. 2-12; Wolters 1925a, S. 46-49; Welter 1925, S. 317-321; Welter 1938b, S. 1-33;
Welter 1938c, S. 480-538; Welter 1938a, S. 49-63.
2 Walter 1993, S. 34-53, Abb. 24.
100 KLAUS HOFFELNER
-TEMPEL
APOLLON rs
o 10 20 30 m
Es ist das einzige sicher nachweisbare Gebaude im Heiligtum zwischen Vollendung des
spatarchaischenApollontempelsund der Niederlage gegen die Athener 459/8 v. Chr.
Das Dach des Artemistempelswurdemit korinthischenZiegein eingedeckt,Strotereund
Kalyptere dabei als selbstandigeZiegelteile hergesteilt. Der verwendete Ton gleicht dem
korinthischen,ist gelbgru~nbis braunund stammtwahrscheinlichaus den in Agina und direkt
DIE DACHTERRAKOTTEN DES ARTEMISTEMPELS 101
I.- -- -193---
FIG.2. RekonstruktionAntefix
14,0
FIG.3. RekonstruktionFirstpalmette
DIEFIRSTPALMETTE(P1.28:c, d; Fig. 3)
Von den Stirnziegeln des Dachfirstes fand man bisher nur zwei kleine Bruchstucke
(Z36, Z 143). Ton- und Oberflachenbeschaffenheitsind dieselbewie bei den Traufantefixen,
mit dem Unterschied, daB das Ornament hier nicht in einer Form gepreBt, sondern nur
DIE DACHTERRAKOTTEN DES ARTElMSTEMPELS 103
aufgemaltwurde. Auch die Maltechnik ist entsprechend: das Ornament tongrundig auf
schwarzgefirniftem Grund, rot sind Blattkernund Seitenrander.
Wenigstensdie obere Stirnziegelhafte 0lt sich erganzen (Fig. 3): Uber dem Blattkern
sitzt eine neunblattrigePalmette, die, verglichen mit den Traufziegeln, von gestreckterer
Form ist, die Firstpalmetteinsgesamtdaher etwas hoher als das Traufantefixgewesen sein
wird. Vom Rankengewindeder unteren Ha1ftesind nur noch die oberen, nach auBen sich
einrollendenund spitzauslaufendenVolutenendenzu sehen. Im Zwickeldarunternoch eine
kleine, hangende Raute.
DATIERUNG
Ornamentform, Ausfiuhrungin tongrundigerTechnik und das Wellenprofilder Sima
sichern die Entstehung der Dachterrakottendes Artemistempelsvon Agina in klassischer
Zeit, genauer,der ersten Halfte des 5.Jh. v. Chr. Eine Stirnziegelserievon Delphi3 soil am
Anfang der Betrachtungstehen. Uber ihre bloBe motivische Ahnlichkeithinaus steht bei
beiden Antefixen der Dekor,verglichenmit a1terenBeispielen,in einem vol~kommenneuen
Verhaltniszum Bildgrund-ein Hauptanliegenvor allem auch des Vasenmalersder Wen-
dezeit von der Archaikzur Klassik.Nicht mehr das bloBeFullen einer vorgegebenenFlache
interessierteden Maler zuerst,sondern die Bezugnahmevon Dekor und Malgrundsteilteer
sich zur Aufgabe. Die Suche nach dem ausgewogenenVerhaltnislfBt ein neues Verstandnis
I
Le Roy 1967, S. 122-123, Taf. 43, 44 (A.42, A.49, A.39), Rek. Taf. 125.
E
I U)
U)
c a. 18,8 -
CO
[LJ C3
DIEDACHTERRAKOTTEN
DESARTEMISTEMPELS 105
in der dekorativenMalerei deutlich werden. Trotzdem gibt es auch Unterschiede: Beim
delphischenAntefixsitztdas Ornament noch unruhigund unsicherim Bildrahmen,beinahe
verstandlich,daB die Ranken durch ein Halsband zusammengehaltenwerden mussen. Die
Palmette wieder bezeugt archaischesErbe: nur die beiden untersten Blatter fallen leicht
fuber,die restlichenverharrennoch etwas steif in der Flache. Demgegenuber sind die Orna-
mentteile des aginetischenStirnziegelsbereitsfest in den Rahmen gefuigtund streng durch-
strukturiert.Das Gewicht der Palmettewird spurbarund, von den Rankenvolutenfedernd
aufgefangen,im Gleichgewichtgehalten. Die Blatterfallen gleichformigund ausgewogenin
den Facher. Kein Wunder,daBdie haingendeLotosblite, in archaischerZeit beliebtesMotiv
in der unteren Stirnziegelpartie,zuruickgedrangtund die Ranke fortan bodenstammiger
Trager der Palmette wird. Uberspitzt formuliert: Das pflanzliche Ornament gewinnt,
fuberdie gleichmaiBigpulsierendeKraftarchaischerAuffassunghinaus, an Vegetabilitatund
Verbundenheitzum Boden hinzu.
Die attischen Stirnziegel des strengen Stils von der Akropolis4 lassen sich den
aginetischen nur schwer gegentiberstellen, am besten vielleicht noch die Stirnziegeln
der Agora-Tholos,5 die trotz aller motivischen Unterschiede dasselbe strenge Formgeriust
aufweisen. Ein Stirnziegel aus Olympia,6 in dem sich Altes mit Neuem verbindet, ist
zeitlich vor dem aginetischen zu reihen. Das archaisch-Flachigehaftet ihm aber noch
stark an. Die Trennlinie zwischen spatarchaischerund fruihklassischerAuffassung von
Dekor und Malgrund zeigt sich wohl am deutlichstenin den Firstpalmettendes jiungeren
Aphaiatempelsvon Agina.7 Rein motivisch gesehen sind sie mit den Traufantefixenaus
dem Apollon-Heiligtum und von Delphi sehr ahnlich. Die Unterschiede in Ausfuhrung
und Auffassungsind groB. Ungewohnlich allein ist schon, daB Rankenpartieund Zwickel-
palmetten im tongrundigen Stil, die Blattkroneaber in schwarzfigurigerTechnik bemalt
wurden. Zwei uiberauskraiftigeRankenstrangewerden durch ein ebenso kraftigesHalsband
zusammengehalten. Die seitlichen Zwickelpalmettenfuillendie Flache nicht nur, sondern
sprengen sogar den vorgegebenen Rahmen. Die dazwischen herabhangende Palmette
allein reichte dem Maler offenbar nicht aus, zusatzliche Blattchen in den Seitenzwickeln
der Rankenvoluten schienen seiner Vorstellung angepafiter. Der abwechselnd rot und
schwarz bemalte Blattficher verrat dann seine archaische Herkunft. Die vorhandene
Flache zu fallen galt dem vorrangigen Anliegen des Malers, ein Verhaltnis von Dekor
und Malgrund wurde erst gar nicht gesucht. Fur Veranderungenoffen stand er nur dem
Motiv gegenuiberund bedingt zur neuen Maltechnik. Die Anthemien der marmornen
Traufantefixe8dagegen bezeugen fortschrittlicheresGestalten. Das Ornament wurde auf
den vorgegebenen Rahmen abgestimmtund kann sich darin entfalten. Der Weg von hier
zu den Stirnziegeln des Artemistempelsist nicht mehr allzuweit.
4 Buschor 1933, S. 45-54 bes. die StirnziegelXVII und XVIII, S. 51-52, Abb. 68, 69. Vgl. dazu einen
Stirnziegelfundaus dem Kerameikos,Hubner 1973, S. 73-77, Abb. 2 und von Delphi die Serie 66, Le Roy
1967, Taf. 52 (A.94, A.182), Rek. Taf. 125.
5 Thompson 1940, S. 65-73, Abb. 54, Farbtafel.
17 Busing 1979, S. 29-36, Taf. 4, 5. Nach ihm entstand das Anthemion in strengklassischerZeit, noch
vor Errichtungdes "Dorischen"Schatzhausesim Athena-Pronaia-Heiligtum,das allgemeinzwischen 475-460
v. Chr. angesetztwird.
18 Schuchhardt 1963, S. 797-823, Abb. 16; Schede 1909, Taf. 4:23.
19 Schede 1909, Taf. 2:13.
20 Daux 1925, S. 86-87, Abb. 89, 94, Taf. 31.
21
Thompson 1950, S. 327-329, Taf. 103:a.
22
Orlandos 1953-1954, S. 3-4, Abb. 1, 2; zum Hephaistostempel,Koch 1955, S. 65-66, 188, Abb. 67;
zur "Hephaisteionsima"von Olympia, OlForsch V, S. 112-113, Taf. 14-16.
23 Le Roy 1967, S. 145, Taf. 56.
24 LeRoy 1967, S. 141-142, Taf. 54, 103.
108 KLAUSHOFFELNER
(Fig.4). Schachbrettmaander,ionischerBlattstab25und Astragalzoneaufder Unterseitesind
jedoch wenig aufschluBreichfur die Entstehungszeitder Sima.
Zuletzt seien noch historischeGrunde angefiuhrt,die die Dachterrakottendes Artemis-
tempelsauch von dieserSicht in das 2. Vierteldes 5.Jh. v. Chr.datieren:Es ist anzunehmen,
daB die Agineten mit der Besetzung ihrer Insel durch die Athener 459/8 v. Chr. und den
ihnen auferlegtenhohen Tributzahlungen(30 Talente),bishin zu ihrer Vertreibungvon der
Insel 431 v. Chr.,kaummehr in der Lage waren, ein umfangreicheresBauvorhaben,welches
ein Tempel immer darstellte,in Auftraggeben und finanzierenzu konnen. So uiberrascht
es auch nicht, daB im Apollon-HeiligtumArchitekturfundeder hochklassischenZeit bisher
weitgehendstausbliebenund die keramischenBefunde und Kleinfunde,26im Vergleichzur
Zeit davor,quantitativwie qualitativein eher bescheidenesBild vermitteln.
DAS ERSATZDACH
Eine Gruppe von dekoriertenund unverziertenZiegeln belegt ein Dach aus dem fruhen
4.Jh. (PI1.29). Weil fur ein Gebaude dieser Zeit im Apollon-Heiligtumweder Fundamente
noch Architekturteilenachzuweisen sind, mehrere Stuickeim Bereich des Artemistempels
gefunden wurden und die Hohe der erhaltenenSima auf einen kleinen Bau schlieBenlaft,
dtirftendiese Teile von einer Neueindeckungdes Tempeldachesherruhren.
Grunde, warum das erste Dach spater ersetzt wurde, sind nicht sicher zu benennen.
Weil die beiden FirstziegelfragmenteZ36 und Z143 (P1.28:c, d) Brandspurenaufweisen,
k6nnte das Dach durch ein Feuer beschadigt worden sein. Schaidenkonnten aber auch
durchWitterungsunbildenoder aufgrundmangeInderPflegeder Bauwerke,vor allem in der
Zeit nach der Vertreibungder Agineten von der Insel 431 v. Chr. bis zu ihrer Ruckkehr
405 v. Chr., entstandensein.
Das wenige Erhaltenegibt eine VorstellunguiberForm und Bemalung der Giebelsima,
sowie von der Gestaltungder traufseitigenStroterstirnen.Das Traufantefixist nur in zwei
kleinenBruchstuckenuberliefert.Die Farbschattierungdes Tons reichtvon hellgelbbis leicht
grtinlichbzw. ocker und unterscheidetsich deutlich von den ursprunglichenDachziegeln.
Man versetzte ihn mit dunklemVulkangesteinssplittund uberzog die AuBenhautmit einer
duinnenSchlimmschicht, die gut geglattetwurde.
(P1.29:a, b; Fig. 5)
DIE GIEBELSIMA
Anhand der zwei gefundenenBruchstuickeZ 19, Z 139 laBtsich die Giebelsimaim Profil
und Dekor fast Itickenloserganzen (Fig. 5). Der Aufbau ist dreiteilig: glatte Halsfascie,
stark iuberhangendesWellenprofil,davon abgesetzt, das leicht schrag geschnittene, glatte
Kopfband. Ein Perlstabschmucktdas Kopfband, ein einfachesAnthemion aus stehenden
Lotosbluitenund Palmettendas Wellenprofilund ein Hakenkreuzmaandermit Schachbrett-
fensterndie Halszeile. Hinzu kommtein weitererStreifenmit Perlstabverzierungaufder das
Dach vorkragendenUnterseite,dem ein ungefahrgleichbreiter,flachigrot bemalterStreifen
25 Ionische Blattstabbemalungauf der Abschlussleistegalt bisher als Zeichen eines eher jungeren Entste-
hungsdatums;dazu OlForschV, S. 126-127 und Heiden 1987, S. 177-179.
26 Al-AginaII, iii.
E
U
LO
an~~~~~~~~c
?
_________________ _ 14,7 -
. .................
bd C1 N
110 KLAUSHOFFELNER
DASANTEFIX(P1.29:d, e)
Von den Stirnziegelnder Traufseitensind nur zwei kleineBruchstucke,jeweils von ihrer
rechten unteren Ecke, erhalten. Wie bei Traufantefixenublich, wurde das Ornament uber
eine Form in den Ton gepreBt,sodaB es sich leicht reliefiertabhebt. Leider ist bei beiden
Bruchsttickenvom Dekor nur noch das Ende einer sich nach auBen einrollendenRanke zu
sehen. Obwohl im Unterschied zu Z29 (P1.29:e) das Fragment Z18 (P1.29:d) zusaitzlich
noch einen roten Basisstreifenbesitzt, ist ihre Zusammengehorigkeitnicht anzuzweifeln.
Eine formale Beurteilungist schwierig.
DATIERUNG
Da im Maanderbandder Sockelzone und abschlieBendenAstragalkaum Kriterienfur
eine genauere zeitliche Einordnungdes Ersatzdacheszu finden sind, ist diese allein anhand
der Anthemionbemalungder Giebelsima zu versuchen, wird aber durch den Mangel an
Vergleichsbeispielenaus dem korinthischenRaum vom spaten 5. Jh. und der ersten Halfte
des 4. Jh. v. Chr. erschwert. Hinzu kommt die bruchsttickhafteErhaltung, wodurch der
Gesamteindrucknur in einer zeichnerischenRekonstruktion(Fig.5) beurteiltwerden kann.
Die auBerstmalerischenund dekorativenSimen XXII und XXIII von der Akropolis,27
beide etwa um 400 v. Chr.entstanden,unterscheidensich zu starkin der Auffassungdes Mo-
tivs und in der Qualitaitder Ausfhfirung,um zeitliche Unterschiede oder Gemeinsamkeiten
herauslesenzu konnen. Schon besser lassen sich die Simenfunde von Olympia daneben-
stellen: etwa die Giebelsima vom "Hephaisteion"-Typusaus dem spaten 5. Jh. v. Chr.,
die Traufsima vom Sudost-Bau aus dem fruhen zweiten Viertel und die Giebelsima des
Leonidaions aus dem dritten Viertel des 4. Jh. v. Chr.28 Allen drei gemeinsam ist die in
der zweiten Halfte des 5. Jh. v. Chr. aufkommendeUbernahme von Akanthusformenin
den Ornamentschatz,worn auch gleichzeitigder deutlichsteUnterschiedzum aginetischen
Simadekor gegeben ist. Trotz der nur bruchsttickhafterhaltenen Lotosbltite (Fig. 5) war
der Blutenkelch hier sicher nicht akanthisiert,lediglich seine Rander wurden mit rotem
27
Buschor 1929, S. 44-46, Abb. 52, Taf. 11, 12.
28
"Hephaisteionsima":OlForsch V, S. 115, Taf. 43; SUdost-Bau:Olympia II, S. 197-198, Taf. 121:4;Heiden
1987, S. 127-131, Taf. 16:2; Leonidaion: OlForsch V, S. 129-131, Abb. 42, 43; Heiden 1987, S. 131-140,
Taf. 17, 18.
DIE DACHTERRAKOTTEN DES ARTEMISTEMPELS 111
BIBLIOGRAPHIE
KLAusHOFFELNER
Berchtesgadnerstr.50
A-5020 Salzburg
Austria
PLATE 28
J~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
F.AntefixZI i:
X;.,i'>~~~~~~~~~~~~~~~1
WS:
t' 25:
7k~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
a. AntefixZi AtV
b. Antefix Z45
\. . s. ~ ~
' v| e-
c. FirstpalmetteZ36
e. GiebelsimaZ35/Z35a
KnAUS HOFFELNER: DfIEDACHTrERRAKCOTTEN
DES ARTEMISTEMPELS
PLATE 29
co~~~C
10~~~"
C~~~~~~~~~C
it
V.-4- ~ ~ ~ ~ .
r ce)~~~~~~~~~~~~~~~~
CD~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C
APXITEKTONIKEE TEPPAKOTEE
AIIO THN APXAIA YIKTQNA
(PLATEs
30-36)
3 BX. LXba8eXyeb 1926a, ceJX.46-50 xat tov xa-6Xoyo -ccv exOeV6cjv tou (&Lou ato 4Ia8exyek
1926p, ceJX. 17-23. ApXatoXoytx5 EuJXXoy5 EtXUV(5vo (Ev tc acVptvdj X"ptcj BacrwxXC) 6nou exOftet
c7UyXtVT-CLX6ntca xa-cpucae tT)v npt)CtT JILXp5auxxoyTi arcv n.
OLXL(1 McXvtxa.
4 OpXiv8oq 1936, ocX. 18.
S BX. Kpuvo6XXs i 1967, ceJX.164-166, ndv.124 avacxaxq -oraLX j?vrjy?ou cratocrje{o nou nipvaye
cFc0vm o86q KopwvOou-IH~apcv xaL
,VpXLCPJ TacoxCXvu TOu +YjLBGC06 av~pcvbxotov ayp6 I. K6XXLa.
6 Martin 1972.
7 Ee iptpUn cxpi va avaqipp 6tL j OEpVunoBoX>iXaL GuC1
apCLXG nou
tTo EXa HIpapGLcu6
touq ayanro(q auvacUXypouq Lily Ghali-Cahil,Jean Pouilloux,Yvon Garlan,Roland Martin, XaLnivcAca6
6Xc an6 tVLOL rovaoacrx-aco PierreDevambez, 4-rav6,JLxaXBX6Epo VnOpo0GE va
8V)OYEL T cYUWBEXpLXj
XaLaVOPC;;tVsaXXiTEYYT5.
8 KpuvrcXX~i-B6toi 1976, ceJX.65, ndv.58.
APXITEKTONIKEETEPPAKOTEE AHO THN APXAIA EIKTQNA 115
XaLt -rV eL8wX))V. To -ros 1975 xa0ap'tvfxe, aXE&OtTXE XOL OameUT)Xe 'to
(+yLWpL&6-ou aypo6 I. K6XXLa.9
Ooo o Baczxa~o A. OpXAvBoqCo6ae, yLa X6youq Beov'roXoytaz,nepLOpLC6OUVva
aO'O VO(ntepLfP3XXa a-tOpy5 XO(Lyppov'dk 'UOV pXaOBoXoTLx6X6PO 'qq EJxu6vocS.I0
IIaP&cxT)Xc'tou e~oeCa nPOyOPLX6( 'tO evtacypOv nou eETXYLCqT ILXUcV XaL
ftOL iLNViupa eELXc qT Xocp Voc'ov cxxoUaca va OU XieL 6'L Oca4xrav euxa'rqaX 'tou
va
ce BeEva auvex 'to Oivacro 'tou ax&,acrou auroo emtaO',uova,
'to ipyo 'rou. Me'r&
iyCve Bex'r c6 'qv ApXacoLoymXOE'rapyda i acot Vou xaL Vou Xop yPOTxe i
OmLcLavaaxaxp4 'ou ApXaLoXoyTLxoXdpou Exudavaq Ve EEXILO'a 6ca , BuCSrUX65
APXAJKOIAKPOKEPAMOIHrEMONQNKAATIITHPf2NKAI EIMEE
rPXp a-o 500 t.X. ronote-o(vUatL 'ta &o ,VAav czxpoxep6.icv xaXun-c~pcov(Pls. 30:a
XaLb). Ant6-o eva BLaaGCeraL,u6vo -6c'avOiLo Ev6aa'ro Be(repo 'to xacrdwcXpo
-Cu
Ve 'teuaIXLo'ou o6adaczoq-rouxacuXnrpa. 'Eva 6XXo'tpEroadtCe-m axipcLo (P1.30:c).
Tr Oicr -ou
t -Cs IEooELU aUt)5iXELXaLo &vtCpaq
OtT j.ou N. B6tarX nou VE ,OrjOGE OLXOVOI4LX&
yLa
to uXwx6aut6. Trv rOmx tcoucau tnapicwcasT tv
va napouaGxatC &+oya q(Ac)toypcppLx& iXj ntvtotu.
116 CALLIOPI
KRYSTALLI-VOTSI
Avyjxouv c'ov o XczL6V~os XCOTo6EiXXCOV.12
r57t0n LVOEILO H oVoL6'rvyanou eUpav[Couv
qcLT KoptvOou eLvaL evyavia-racq.13
Ve av'ULGUoLxXa Mey&Xv oioL6'xra iXOMe xaL Le
avcsXoyaBetyua.raaxpoxep.icov aio6 'roy 7t'pLvo vac6'r Map~IapL r.t 'r Aexy6V.14
O rp[kor vyeWVV Ve Vapo xcaL top'roxcrX xpc1L6Va (P1. 30:c) epaVCetL XaXTanX LX5
Xp)Vauaxcx xaL aXe~aauxGx opuOL6ra Ve aviXoyouq 'rou Ovaaupo6 'rcav Meyapicav
a'vv Oxuvnt~ca5 H opIoL6rTya auor' ?o MmV Li via aIOrca7 o'ro np6opXVa 'rr
Btax6OaV4)O -rou 8Ooaupo6 -rcavMeyapicav =6 KopLvOtou,-reXvke evcOo X6yos
cV&ytpo51 'ou vyav vtxv 'cv Meyapfcv Et 'cV KopLVO(Cv.'6O J. Heiden nap6Xi
UTVaCV'TdpcOnTOU GLorOtOLZe,Oecapel oayoupv 'Tv KopLvOmLX5 npoiXeuv 'rou t)XLvOu
L&cXOOGIOU ToU Ovoaupo6.17
H vca ex~o>X ntou VnOpE{ va BLO'UtCAGJOe, BeBopivou 6CLXaOL 7 ELXU6Va cLVayipepCaL
Gav nToxu ntcaxaL XTLpLaXr napouta GOo bsko rAv OGnaupxcav8 qq O~uun(ac e(vat
r ax&+n! 6'ULoB0&xoGoqo 'rou Onaaupo6 'rcov Meyapicav iy LeLGcq anco6ELXUCVLOUq
'reXvLE'r. 'EUOaLxep&E4oupe cr6 'v E2LXUC'VLOC yn TpeL5 IpLOaU7xoc-oaxeun nyepV~e
x nut'rrpeq yLOC 'OUq OtOCOuq LGX6OUV ano6u'a OL ea&YrepLXOEyeCpeI'rpLXOE VOVOL7tOU
B&itOUV'COapXOa" PXL'reX'rOVLXa7twaXLVC BLOXOaVI-aX&'cv o'reycv '," 19
aL.
0 5or n.X aexnpoOawnel- xaL ant6 'rtn ua<,ieyapLx 5>? otns - Xo
Xapax'pLa-
' XPxapax pwuxOx5evaXZax-cLx Bmx6aVrar
rLX&xup'r&~ervonpopRX(P1. 31:a). 4 XPpeA
enrxypuXXwv avoevtcv XaLxxetLOarcv ,nouvnouxLwv XOUXOUBL)vWX(O6 Y ou
BtaxOvoao6v
xaOL urL No Cwvec. To 7tLXpLOIa 'civ XouXouBLW've(vaL xrpLVO ev 0o nX6M elvaL
vo ,eVya&Xca ,ua6pa XczXLxa.
xurpwoVpaOLVo KacLe8 r xaCzrCpprjoVIoL6'-ra,ue -CVVa
'rj aecwvarrXm otacvrs 'rou OvGaupo06 'Cxv MeyapicAv y lveroaL avio VrL
avCX 5.20
flap&XXrjaiXoz xaL o-qv o'repy 45 'cwv \eXyp6v2'xaL oa-o S. 28 ao6 -rv K6pLvOo.22
OLyparcir otaus -rou 4ou n.X. aLwCvaelvaL rocaper. H otarj a (P1.31:b) Ve xaznt6XrB-
aCvoJnepcav(lct appovtxfi bt&&rca ev8exavyxcpu)XXcvavOiv p P
poJte rcavne7taxcv npoq
12 rla tov t-nto an6r6cav xaL trv
acx6 uav nye?6vcv xaXuxc5pcavVe -rv opyavca?ivr tapfp&Oa
auvoXxlj fis tp6aqaw PLPXLoypa(pa noXA tpOayippl B&apLp5B-zaXonouXou-OLxov6pou 1986,
r trj
Gex. 19.
13 Corinth, i xaL Roebuck 1990, aeX. 59-63, ndv. 7, 8. rLx
apXplmx& xvO4iLp PX. xxL Winter 1990,
aeX. 13-32.
14 Le Roy 1967, toit 45, aeX. 103, Jt[v.I
irxyp&n' anxtLX6vlo xaLpl. 39. 0 ChristianLe Roy Xapaxt-cLO
tlX& avaypm "argilejaune couverte creme."
15 Heiden 1987, aeX. 83 xaL ndv. 10:2 xaL Winter 1990, aeX. 13-32, apaL6-ca-qOXebaauLX5 ava-
napitapaat trjm atiyrj tou eaaupo6 mav Meyapi&v, fig. 7, 6nou xaL e86 BLaxplveCaL to &palo X4L
xaL 7 eu=OOaT)Oatou ayan)tro6 ypLouxaL exXex-to6 C&yp&pouK. HXL&xT.
16 fHauaavlaq, HMLmx& 6,19,4.
17 Heiden 1987, aeX. 84.
18 fauaavlac, HXLM& 6,19,1.
19 Pcpaiazoq1951, aeX. 108-120. KoXoxoca&q1990, aeX. 141-148, CLX.5.
20 Heiden 1987, ndv.10:1, aeX. 84.
21 Le Roy 1967,aeX. 122, ndv.43.
22 Corinth i xaL av&XoyaFA 3, FA 19, FA 430, 452 (Roebuck 1990, aeX. 46-47).
I,
APXITEKTONIKEETEPPAKOTEE AIIO THN APXAIA EIKTfQNA 117
23 Corinth
I, iv, aeX. 85-86, ndv. 20:3 (two pieces of raking sima) xaL 20:4 (two types of eaves tiles from
south half of Stoa roof [S.90]).
24 Le Roy 1967, aeX. 156, ndv.59.
25
Kax-ca&q1988, aeX. 44 xaL aXi&o IV ndv.5:8, e, 6:a.
26
Kakca<? 1988, aeX. 68-75 xaL xupLaaeX. 75.
27
@Ei va eXnsc 6tL i t6ao nOXBtly VeXi-q xaLnapouaOaa(O)ou auwo6BevOaxaOuatrp5ae
UXLXO6
yLa itoU.
28
4LabeXpe( 1926a (anox&X.u4L O.telX(v vao6 1920-1926), aeX. 46-50, ClX. 4.
29 Le Roy 1967, toit 84, aeX. 173, ndv.69, 70.
118 CALLIOPIKRYSTALLI-VOTSI
30 Le Roy 1967, toit 84, asX. 173, sdv. 69, 70. 'EXopI tn BLapopi 6tL o-ouq AeXco5 t oaitXa
BeV EEVaLtpLOV(0-ca.
3' KacX-oa&1988, EIII a ofX. 52, un. 33 6nou napa0&cLxaL nXooauxPLPXLoypLypp(x.
I, iv, asX. 84, itv. 20:1. E86 npzinL va avayppOeE6tL T) XpovoX6TyW-qq N. Eto&q V?Ct&
32 Cointih an6
via 0tOLXe1antouitpoixu4av ?ctatCe(o-CaL aoo 320 nt.X.Williams 1980, aCX.107-108, xaL 1988,
KaczXo&s
aeX. 71.
APXITEKTONIKEETEPPAKOTEE AHO THN APXAIA UEKTfNA 119
iXet a aL
Vt Ova Ltxp6avuxa-on-pta6 xs t8tavLep6-cj-oaq-q EtxucovtaO yxunctx#
Eata-c-vna-Cptaa
aCsxIv -couq.
An6 -tq np6a-eq avaxorpacpa P ,6petm nXeup&-ccv vaccv iptie -qi~tia &ptoa-
exvoupyr.lvjq a.tljq (P1. 33:a) nou Btaac5CeL ae
a~epyo av&yXupo-cs yvca-c cpu-
-cXiq exLxoeta tO aXoVI5eL tou exyp6ov-catao6 noX~veupo PXaacs6xat No Oa~ep&
p6XXca &xoavOaq.Abtmoaco Ao ato auoc6 pu-cLx6M&atio O
X aCeOcu -cLo vocVCp&ko
XeovcoxpcyaX5 6otou j Xa&nj8XoveJ)at tie CpetsGetpiq YXoyC0ro Oucaasvouq. To
au-cE-ou Cc(oue{vat utixp6xaat aLXtgip6.
Kat Y O(tiujau-cx-ono~e-eEat Oe.tiuxm&noX6 xovc& act XIII a a6-c6 'Acuuv-,41
v
oa exeev-1-ou Aecav(&tou xcl Mutaq,42 aqOxa(tieq 19565 -ou Eevtxo6 ApX-
atoXoyLxo6Mouae{ou, xaatypuLtx&xaatcaeexejv- -c N. E-o&s ujq KopivOou.43
MLxp6 AM Xapaaxc-ptPOx6v~o 4tp-Ila e{vat xcat ceIA&Xto (P1. 33:b) nou
aOITANq
8ta&ceOgt to5 xcAMa -a xpoXaPa. EV"-tvnavC)TcaLVa (amxaatXOYXoy &taYp5Xxaevda
cti lisacdca tie L&trkepo paeOoqCav- Baase-aL sxoeXceo
XLXuxLx5
yxc PXa-c6ocpeya ,us
eVcoveq pc0ocYOxOaXeOL.Yxo iva &Xpo8 aiajveCaL o
acspoyyuxa&a nou Oa
xaeunc xav
at6 xTv XeovxoxepcXaX. E7jv aOuv`eLOcsa Aa-tXev -p&nj 4C6v-1 O6Cvcat o ao5v~ecoq
evXLdiveaca To Texpiycvo ,ue -o apaxccxxc6
[AaLacvpoqxat c
x6oajA-jia.44 vIv asc6 x&rxx
oprcx etcLaveLa untpXeL accrp&yacoq yparc.6 jue &o IALxpe6xacL,IAa,ueya"X-1 Xavpa.
Ta xpdaicca nou xupLapXo5v 'vcat -co liacpo xacL-o t6er, -q KopLvOEaq.
TeXeu-adoavayeLpe-aL jtxp6 yA`jA(aaEITqpe oaveOv p6XXCzV &6xavOaq xxc)vapL XaL
exLxoeLaeLq anox #JLq, nou 6o.caq BeXvet A&Bat&p exx5 gXeXOI (P1. 33:c). To ,uLxaBXXLx6
luacpo Pf&OoqxaL Y a-ryv5 OXcrLx&xexrXOeaI xv rono0eE a-co CeXoq 4ou/ epq
3ou sT.X. oudava. lIp6xevaL yca e5prjua ano6ntaXou avacoxacp.
AKPOKEPAMOI
An6 -Tl vieq avaaxayip ioje -yna (P1.36:a) nou euxoXa j.nopel v& auyxpLOt~
Ve -q a-cpyn 81 -ccavAe)Xcpcv58cxOcO s xat &XcXc&5o anoq4ypxa Ve evaX~aa ievm
avO4LLta xat xXetLa-6a-o evva avotxc6 a-co &XXoXouXo45caXcwo6 (Pls. 36:b xat c).
TiXoq avaipfpovat &XXa&5o Tjictcaxa a-pc5pcav nyeuvxcv lie Xuyep6xopiua&vOi
(cpXoyc-rX6q puOji6q)xat avotXxa &vOTXcwo6 nou avm5ovcLat at6 nptovcoYc&i ainaxa
(Pls. 36:d xat e).59 E-co iva (P1.36:d) 6Xa -m y6X~a iXouv ypaitxci nepLyp&4lOaca.60
A~E(eLva avaocpepOe 6-Lt ca np6(pcpaaBev X4Xajie a-OotXela yta -nv 6mp~ xepa-
,lLtxc'v eyxcapkavepyacrpkav ex-6c Pipata an6 -q ,iap-upla -cj ypaon-ct'nap&loarc.
T6pa 6juar xat L8taLkcepa ,u& -o &-os 1978 ae Boxtpuax avaaxacpy5a-ov
cayp6XwcCnBdxvensivx a-co oponi8to PpiOvxe nXk5Ooq U'mepcp)vaxpoxipavcwv, ,uey&Xa
xoVV~i&-tL xaxo+Viuvou nvXo6, noXX& -yuiaxca c=6 Xeov-oxecya~Xi ubpopp6es
(P1. 36:f) x=06, xat ju~xpa ojio aOcxor (apto6.
To uXtx6 au-c6 Xpet&CL-at
xaOOptaLi6
xat auv5pca' tpEvant6-v oxoxxvpcuivv
BnuoatauaT, r
jiasa~ 6,ucr ev-ontap6
an
epyaa-qptou Oa- Ltxu6vca eLvat ,uy&Xv.
Ait6 -o uXtx6 nou napouat&Ce-atOao Ouve'pto au-c6 le -p6oo auvonLx6o ntLOec)
Otc& Vy va avaO-c&atL T au-co-ciXeLa -qr EIxuCvaVO BtXa aov jiey&Xi
6-L apXLCeLOL&
napoua(a -qq, KoptvOou.
Ano6 -q LtxuaVcaxat -qv K6ptvOo ave-Lat ntaL oXoxaOapa 6c6t Oa texLvo5amv -c
auvipcta -cc)v -XvLcuv yta oXoxXinpi -v rkXon6vvvao acXX&xat xIv B6peta EXX&Ba,
Geaatax, Maxebovoa.
Ta FioXa 4+cLpL&AYc& -cc)v avBpcV(*)V TCiVotxtLv x5 Itxu'Ox(c61 Ve Ca xexLaloiCa
avokvta xatexecL
vXwcexe{V
-vnTVucopE
cx Uvptvpv -c v n~oXc;)v acso t+cwpLco6 -rou
6a-cpou 4ou n.X. at., nou 6napXet L onoOe-qjivo ai gutL ano6 -ts aOouaes cou cont-
xo0 Mouaelou, nou oxoypmvepaavaapOcyovtp at aCXa O +YC cwcmpLn ae otxLL6v xat OaLr
ypan-cir BtaxoajiLaatL -&cpyv -cjq Maxdovkaq, Oecap6, %8 ano6 noX xp6vta, 6ot
aoMoCXo6vaVCEpp-C [Aup'tup(aYta -CV etnpxaTs nOU &CaXsacvOaTCV BLaXOOa-LX'
yX~aaa -qq MxxebovEtxcaXX& xat -qq Hnetpou.
OL,ovoypacyp(e, -ou N. Kakcak yta -tr, xepaVu6aetqa - MaxeBovlaq,,62 vex
Cc*)votXtdv qqj IHixxart63xacoda xat r ,ovoypacpLta -rj A. BXaXono6Xou-Otxov6VIou
niv 'Hs:po64 Oappc 6ct o8ebouv np6os-qv &ctopnaoucm. Htoa-5 6-L a-o
c6ao Xp5aOLo au-c6 Ouvi8pto yta -tq apXcex-ovtxic-cppax6-ce Oa xaxabeqOot0v
ao6 6Xouc,-ouq auvaBiXpouq nou VeXse-o5v-o L8toOiua -a xotv i un6yeta pe4VaLca
nou BtftpeXav -cnv yXux oa-cxIcavia, -cnv aua-cnp5 rkXon6vvvao, -5v xpua-cTcXXtvTAx-
lctx xat V pepov-av Ve ,ovonc-ta opetv& xat OaXaaotvmva-cxv D&x(&a, AxapvavLca,
'Hnetpo xacOda xat av Geaacaca, MaxeBovka. Au5c' axptpbc. -qv npayua-x~ c a
apOu
TOU 4C(V.VOo lAtco TOU eXXO&XO~5 OU
o uppiyxta4e and6 Tt5 ea6pce B&ep-
yaalec -ou vat MVetxat vacnapet
Lp tppoi sou dusts ot apXatoX6yot
yvwpLCollenoxu
xacX, (Caw Oa
0npene xat Vn
va 'n bob vaX BexC ov xat enOaxc?OVeqnou x&-cc aco6
aXXoueLBou,avayxat6L6-yec -cv {X.ou'v.
BtLXtoypayta
Amyx, D. A. 1988. ConthianVasePainingof theArchaic Period,Berkeley
Avspiou, H. 1983. <<To 1Lxp6 Ota-po tiq AVfpaxta>>, HeyLPcaXXiXpovwxi 25, aeX. 9-23
Avbp6vtxoq, M. 1984. OL
BepyTva. P3aaLXLxoE -t(ot, AO4va
Caskey,J. 1925. Catalogue andRomanSculpures,
of Greek Cambridge
CharbonneauxJ., R. Martin, xat F. Villard. 1970. Grkehelenistique, Paris
Corinth I, iv = 0. Broneer, TheSouthStoaandItsRomanSuccessors I, iv), Princeton 1954
(Corinth
Corinth IV, i = I. Thallon-Hill xat L. S. King, DecoratedArchitectural
Terracottas (Corinth IV, i), Princeton 1929
4apix)ca, N. 1971. <<ApXata Emucavoa.ApXaeq eXXnvLXi t6XL>>, AOva
iabe)Xcpe~6,A. 1926a. vAvaaxapaE Exuvoq>>, AeXt 10, 1926 [1929], aGO.46-50
. 1926g. <<ApXatoXoTyxm EuXXoyr Exu6voq>>,Ae)t 10, 1926 [1929] Hapdptrpa, aeX. 17-23
Gabelman, H. 1965. Studienzumfiiliriechischen Lwenbild,Berlin
Heiden,J. 1987. Korithische Dachziegel.ZurEntwvicklung derkorinthischen
D&her,Frankfurt/Bern/New York/Paris
Hesperia59 = Proceedings of theFirstInternationalConferenceonArchaicGreek Architectural Terracottas,Atens, December
2-4, 1988 (Hesperia 59, 1990), N. A. Winter,enty.
Haibner,G. 1973. "Dachterrakottenaus dem Kerameikosvon Athen. Ein Beitrage zur Bauornamentikdes
5 und 4Jhs. v. Chr.,"AM 88, aeX. 67-143
KaXaw&, N. 1988. llXtveq 8Laxoa iveq xepalae4Lq aco6- Maxebovia, AOva
Ko)oxoradq, K. 1990. <<Apl?ovLxaE XapdLetq ata applxd apXctextovLX6nUxLtvatou 'Apyouq xatLnq
Entba1pouo, Ev Hesperia59, aeX. 141-148
KpuacXX)n, II. 1967. <<E=6vo, Ae)t 22, 1967, B'1 [1968], aeX. 164-166
KpuaXX)B6-can, II. 1976. <<Avaaxacpfj Exuvoq>>, Ae)t 31, 1976, B' [1984], aeX. 65
AalIiptvou86xnq, B. 1975. <dep6v MaXe&tou An6XX),voq e EnEbaupovo, Hpaxtwx&1975 [1977],
aeX. 162-175
Lauter,H. 1986. DieArchitlektur desHellnismw,Darmstadt
Le Roy, Ch. 1967. Lesterrescuitesarchitecturales deDelphesM),Paris
(Foujiles
Maxap6vag, X., xat E. rto~pn. 1989. Ot OLX(Eeaptayqjg ti EX&vn xat Atov1aou tin lOX~aq,
AOfiva
Martin, R. 1972. <Rapportssur les conferences de l'ann~e scolaire 1971-1972?, Annuairede l'tcole Pratiques
desHautesEtudes,Paris, aeX. 233-237
Mobius, H. 1929. Die Ornamente desgriechischen Berlin
Grabstelen,
OpXvboq, A. 1936. <<Avaaxaq4 Ewxu6vo>>,llpaxtwx& 1936 [1937], aeX. 86-94
llavcepl?a)fj, A. 1987. <<Hxep6Vciaan tcav avaxt6pcav amn Bepytva>>,'Aln-coq. TLvnLYLX6q tc6voq TLa
tov xaOnyrvn-Mav6Xn Avbp6vLxoII, Eeaaaxovxn, aeX. 579-605
Payne, H. 1931. Necrocorinthia, Oxford
Roebuck, M. C. 1990. 'ArchaicArchitecturalTerracottasfrom Corinth,"Ev Hesperia59, aeX. 47-63
Pcoiacoq, K. 1951. Kdpapot1 cn Ka)CubCvoq,AOva
Salzmann, D. 1982. Untersuchungen zu denandkenAiesebnosaiken, vondenAnfdngen bis zumBeginnderTesseratechnik
(ArchkologischeForschungen 1O),Berlin
BBaXoto6Xou-OLxov4?ou,A. 1986. HyetV6ve XaLXOpuyatE X4paOqOL lie 8&ax6a.vqavq an6
0tqv 'HneLpO.
TUtoq <<&vOoU; Xcato6-eXtxcav>>. (&&.&aCrpLpf;),Ica)vVva
B6tan, II. 1984. 'Epyov 1984 [1985], aeX. 61-62
. 1987a. <<Avaaxaxxdj Etxu(vo>>, Hpaxxwx 1987 [1991], aeX. 66-68
. 1987P. 'Epyov 1987 [1988], ayX. 91-92
Votsis, K. 1976. "Nouvellemosaique de Sicyon,"BCH 100, aeX. 575-588
124 CALLIOPIKRYSTALLI-VOTSI
CAlioPI KRYSTALLI-VOTSI
LoukiAkrita38
Athens, Philothei 152 37
Greece
PLATE30
AI
J.-.
b. K&Trca
.lLa6axpoxcp&iou xcaXuripa500 .X
4.)
~~~~
I ~ ~ ~ O
DtKid ,
L~~~~~~~~~~~4
. t~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Z
H '
I-A
x
)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~A
A.~~~~~~~~~~~~~~~~~~~~~~~~~~~~a
I~~~~~~~~~~
>4 a ~~~~~~
?-f~~~~~~~~y 4'~~O
PLATE 32
?.0
And7',
P
-a 5~~~~~~~J l~~~~~a~
I~~~~~~~~~~~~~
too~~~~~~~~~~~~~~~~~~~~~~~~~~~~~o
>~~~~~~~~~~~~~~~~~' ---.9,; 0
o t0o
PLATE 33
.1~~~~~~~~~~~~~~~~~~~~~~~4A
pXt)V &cxaAia~
a('~XX
HICPLOX?
Fiacx
&cxxpao.tiji~ a
cays4a
v?
c. TVa e - dyXuq--
300-280 ,.X.
KRYSTALLI-VOTSI:ApXl'r?X'TOVLX& TeppczXOiec cat6 'xv ApXCI(C
CALLIoPI ESxuxvc
PIATE 34
PLY~~~~~~~~~~~~~~~~~~~~~~~~~'
-~ ~~ ~ -4pal ~ ~~~.xo~ao a ya
-p.
;ct-5___.,____,> ~ ~~~~~
5.<dAxpox~xpatog x(tog EstLaOCpou
c. Axpoxipapqo x6,op Kopyv9ou
CALuuopi
KRYSTALLIVoTsI: ApXLCX'~WVL4gTcppczxotc ciso armAp~czkzS2Lxubvc
PLATE 35
13~~~~~~~~~~~~~'
A, x~~~~~~~~~~~~~A
4 ci~~~~~~~
Oak ~ ~ ~
4~~ 8 ~ ~~~~~ t...:4
Ar.~ ~ ~ ~ ~ ~ ~ ~ ~~~~~0
:1~~ ~ X
<jA"~~~~~~~~~~~~~~~~~~~~~~~~~~~r
F>'
7 a) ~~~~~~~~~~~~~~~~~~~~~~~F b
t Y 3
l * H a
.4I
Fe-_~~~~~~~~~~~~~~_
PLATE 36
: ~~~~~~~~~~~~
'tetIt.
~ ~~ crppc-rpp
c.~ K&inso ~ ~~~~~~~~~~~~b
erntypdveta lyelia. .
c. Kd-rca
entydveta mpN. Kicc cstw&V p6van
atpcXmpc
4M--f of
'4~~~~~~Z
c. K"c
CtWVC~atcnpc yc?v
. ~ ~~~~~~~~~~~~~~ r
e. Kitw ?1tLop&VCL~
atpwnipa 7)ycvuva qp~oycro6 f. M&ca
xxxovnaj-tvou ujlo6 cat6xoy ayp6
puGpuX<Z~x
KRYSTALLI-VOTSI:APXLtCXtOVX6C Tcppaxotcc cut6 xnv Apxata
CALLIOPx ESxubva
DIE DACHTERRAKOTTENVON AIGEIRA
(PLTms37-40)
1 Zum Tempel B auf der Akropolis,siehe Alzinger 1985, S. 430-431, 446-451 sowie Abb. 23. Siehe auch
Alzinger 1988, S. 20-23, Taf. 6.
2 Alzinger 1985, S. 431, Abb. 27.
3 Roebuck 1955, S. 156, Taf. 62:e-g; Robinson 1976, S. 231-235, Abb. 9; IsthmiaI, S. 40-53.
4 Zu den Architekturfundenaus der Zisterne,siehe Alzinger 1985, S. 431-451.
5 Die Ziegellangeist uns durch kein einziges Fragmenterhalten.
126 DORISGNEISZ
22
Zur Kombination von Dachziegel und Firstpalmette,vgl. Le Roy 1967, S. 174-175, Nr. 8 und 9,
Taf. 70-72.
23
Alzinger 1985, S. 442, Abb. 38.
24 Alzinger 1985, S. 441.
25
Alzinger 1985, S. 443, Abb. 39.
26 Corit1hI, iv, S. 86 und 96, Taf. 21.
27
Alzinger 1985, S. 445, Abb. 40.
28
OlForschV,S.116-117,Taf. 17und44:2.
29 Zum attischen
Einfluss,der sich an den meistenkorinthischenDachern der Hochklassikbemerkbarmacht,
siehe Heiden 1987, S. 112-113.
VONAIGEIRA
DIEDACHTERRAKOTTEN 129
konnte.30 Die erste Bauphase erfolgte in der ersten Halfte des 3. Jh. v. Chr.,31 als das
hellenistische Zeusheiligtum mit einem Theater ausgestattetwurde. Diese Bautatigkeit
war jene Konsequenz, die auf die Gruindungdes 2. achaischen Bundes (281/280 v. Chr.)
und den etwas spateren AnschluBAigeiras folgte,32da man aus politischen Grunden mit
Reprasentationsbautenaufwartenwoilte.
Von den verzierten Teilen des Naiskosdaches wurden Fragmente der Rankensima,
welche die Traufranderschmacktegefunden,33ein Antefix sowie zahlreicheFragmentevon
L6wenkopfwasserspeiern.Trotz der geringen Uberreste laBt sich bei genauerer Betrach-
tung der Stucke, was den Aufbau der Sima, die Gestaltung der Ranken, Akanthusblatter
und Lowenkopfebetrifft, eine gewisse Verwandtschaftzum Dach der Sudstoa in Korinth
erkennen,34 die nach den neuesten Untersuchungen in den letzten Jahren des 4. Jh.
v. Chr. entstanden sein muB.3s Wie in Korinth wird auch in Aigeira jeder Simenblock
einen geschlossenen Ornamentkreis gebildet haben. Das RankensimenfragmentE 5A
(PI1.38:b) vom Naiskos D ist deshalb so zu erganzen, daB kannelierte Stengel hinter den
Lowenkopfenaus einem Akanthusblattentspringen. Aus einem geteilten Hullenblatt, das
durch einen Fruchtknotengegen den Stengel abgesetztist, wachsen unmittelbarzwei flache
Voluten. Die Sadstoa zeigt noch die frtihereGliederungdes Deckblattes,das in der gleichen
Richtung wie die Stengelkannelierungverlauftund ihr mituntersogar genau entspricht.Ab
dem 3. Jh. v. Chr. wird ein Richtungsgegensatzder Deckblattfaltungangestrebt.36 Dieses
Schema ist auch auf den Simen Aigeirasvertreten.
Die stilistischeVerwandtschaftzur Stidstoa von Korinth zeigt sich vorwiegend in der
Rankernornamentik.Allerdingsweist die Traufsimades NaiskosD schon eine nachlassigere,
auf derbe Reliefwirkungberechnete Bildung auf, was fur das Dekor ab dem 3. Jh. v. Chr.
bezeichnend ist. Ihre Beziehung zum zweigliedrigenhellenistischenSchema37spricht ffir
eine Entstehungnoch vor dem Ende 3.Jh. v. Chr.
Im AnschluBan die Sudstoa soll auf drei weitere Simenfragmenteaus Korinth hinge-
wiesen werden, die ebenfallsdem FragmentE 5A vom NaiskosD sehr ahnlich sind. FS 433,
das im Asklepieiongefundenwurde,kommtder Sima der Sudstoasehrnahe und ist aufgrund
der Stratigraphiedem 4. oder 3.Jh. v. Chr.zuzuordnen.38FS 418 und FS 419 ohne Angabe
30 Zu den Ausfuhrungendes Pausanias,der im 2.Jh. n. Chr.Aigeirabesuchte, siehe Pausanias7,26,4-5, 10.
31 Zum Naiskos D, siehe Gogos 1986, S. 32-38 mit Abb. 65, der Rekonstruktiondes gesamten Theater-
bereiches.
32 t'ber den achaischen Bund des 3.Jh. v. Chr., siehe Urban 1979.
33 Das Dach des Sudost-Bauesin Olympia ist nach dem 200Jahre alterenDach der Lowensimain Kalydon
das alteste, das wieder eine Sima an der Traufseite zeigt. Mit der Wiedereinfuhrungder Horizontalsima
und den ersten Ansatzen, ein gemaltes Ornament in die Dreidimensionalitatzu tragen, bildet das Dach des
Siidost-Baues,das in die Jahre nach 370 v. Chr. zu datieren ist, den Vorlauferder reliefiertenRankensima.
Vgl. dazu Heiden 1987, S. 127-131 mit Lit. und Abb. Taf. 16.
34 Siehe dazu CorinthI, iv, S. 83-88, 96, Taf. 19-22.
35 Siehe dazu Heiden 1987, S. 141-142.
36 Siehe dazu Schede 1909, S. 63.
37 Schede 1909, S. 80-83.
38 Herrn ProfessorCharlesK. Williams,II, dankeich sehr herzlich,der mir eine Durchsichtder zum groBten
Teil unpubliziertenDachterrakottenvon Korinth erm5glichte.
130 DORISGNEISZ
des Fundortessind m.E. wegen ihrer bereits teigigen, schlaffenFormen spater,eventuell ins
ausgehende 3.Jh. v. Chr. zu datieren.39
Im Hellenismusverlieren die korinthischenWerksttittenv6llig ihre Bedeutung fur den
auBerkorinthischenMarkt. Neue Impulse gibt es nicht, und die Handwerker zitieren
nur noch altere Dachrander. So haben sich die Bauleute von Aigeira nicht nur an den
OrnamentformenKorinths orientiert,sondern nahmen auch Einfluisseaus Argos auf. Dies
laft sich an den Antefixen ablesen, die den Simen des Naiskos D aufsaflen.40 E 24/88
38:c)ahnelt sehr dem Ziegel der Serie C 26659 aus Argos,41der sich keinem bestimmten
(PI1.
Gebaude zuweisen 1aBt,und von Mde. Billot in das erste Drittel des 4. Jh. v. Chr. datiert
wird.42 Die argivischeWerkstatthat fur den Stirnziegel das klassischeAkroterornament
stark abgewandelt.43 An die Stelle des auBerenVolutenpaarestreten lange Blutenstengel
und die Voluten besitzen akanthisierteDeckblatter. An die klassischenZiegel der Serie
C 26659 schlieBtdas Antefix C 2687 an, dessen reiche Gliederungund der zur Seite fallende
Blutenstengeleine noch fruhhellenistischeEntstehungannehmen lassen.44
Die Tatsache, daB die gleichen Stirnziegel, wie sie dem fruihenNaiskos zuzuordnen
sind, auch im Theaterbereich gefunden wurden, legt nahe, das Buhnengebaude etwa
gleichzeitig mit dem Naiskos D anzusetzen.45 Was nun die Gestaltung der Tonsima der
hellenistischenSkene betrifft, so lAiBtein Vergleich mit der Sima des Naiskos D deutliche
Parallelen erkennen. Eine Gegentiberstellungder SimenfragmenteE 5A (P1. 38:b) und
ATK/87 Theater ATK 21/72 (P1.38:d) und ATK 19/72 (P1.38:e) zeigt, daB nicht nur bei
den Antefixen,sondernauch bei der SimengestaltungdieselbeArt des Ornamentsverwendet
wurde. Die beiden zuletzt genannten Stuickesind zwar in der Ausfiuhrungder Deckblatter
verschieden, darften sich aber aufgrund ihrer sonstigen Ahnlichkeit zur selben Zeit am
gleichen Gebaude befunden haben.46 Die L6wenk6pfeschlieBlich,Kopf E 6a/88 (P1.39:a)
vom Naiskos D und ATK 3/72 vom Buhnengebaude gleichen einander so sehr, daB fur
beide das gleiche Model verwendetworden sein mag.
39 Die Knospe im Zwickel der auseinanderstrebendenVoluten lasst sich erstmalsauf der oberen Sima der
Tholos zu Delphi feststellen,die um 380 v. Chr. errichtetwurde: Roux 1952, S. 442-454 mit Abb. Ab da wird
sie zu einem haufigenDekorationsmotiv,das auch gerne in Bliltenformauftritt.Vgl. dazu die Sima der Tholos
von Epidauros:Roux 1961, S. 140-145, 160-168, Taf. 43.
40 Vgl. dazu Michaud 1973, S. 319, 321, Abb. 139.
41 Vgl. das Antefix C 19226 aus dem Heiligtum des Apollon Pytheos: Vollgraff 1956, S. 18, Taf. V sowie
den Stirnziegel vom "Dromos X": Deshayes 1966, S. 229, Taf. XLI:3, welche sich an die Serie C 26659
anschliessen.
42 Herzlichen Dank schulde ich Mde. Billot ffr ihre freundlichenAusktnfte und das mir zur Vergugung
gestelltePhotomaterial.
43 Einer Reduktiondes Akroterschemasbegegnen wir ab dem Ende des 5.Jh. v. Chr.des ofterenin Olympia,
Delphi und Epidauros. Zu Olympia, siehe OlForsch V; zu Delphi, vgl. Le Roy 1967, Serie 69 (A. 109 und
A. 95), S. 141, 146, 148, Taf. 53; zu Epidaurosvgl. Harl 1971, S. 69, 109 (Serie E 18) mit Abb.
44 Harl 1971, S. 91, 98 mitAbb.
45 Vgl. dazu die neueste Untersuchungdes Theaters von Gogos 1991, bes. S. 174-184, wo eine chronologi-
sche Abfolge der Bauphasenerstelltist. Siehe auch Gogos 1986, S. 6-31.
46 Aufgrund der sehr vagen Angabe des Fundortes zu Beginn der Grabung 1972 konnte das Fragment
ATK 21/72 auch vom Naiskos D stammen.
DIE DACHTERRAKOTITENVON AIGEIRA 131
Naiskos D erfuhrbereits kurz nach der Mitte des 2. Jh. v. Chr. einen Umbau, der mit
der Aufstellung der Statue des Eukleides zusammenhing. In diese Periode, in der auch
eine Neueindeckung des Daches erfolgte,gehort das FragmentATK 1988 (P1.39:b). Dem
Stuck fehlt jede Feinheit der Linienfuihrung. Die derben Formen haben jede Plastizitat
verloren, das Ornament ist in die Flache gebannt. Wir befinden uns in einer Zeit, in der
eine flachenhafteund linearbetonteOrnamentikim Vordringenbegriffenist. Das Schema
auf der Sima des Naiskos D zeigt eine Abzweigung von der zweiten Volute nach oben,
was an vergleichbareStucke in Delphi erinnert,die allerdingsder zweiten Hailftedes 3. Jh.
v. Chr. angehoren.47Dieser Dacherneuerungist noch das LowenkopffragmentATK 1986
zuzuweisen (P1.39:c), das die fur das ausgehende 2. Jh. v. Chr. bezeichnende Richtung der
Naturentfremdungund Tendenz zur Maskenhaftigkeitbereits andeutet.48
Zu einer letzten Umbauphase geh6rt das gut erhaltene Fragment eines Lowenkopf-
wasserspeiersATK 1987 (P1. 39:d). Fur eine chronologische Einordnung von Interesse
ist die Mahne, welche uiberder Kopfmitte gescheitelt ist. Vergleiche mit Lowenkopfen
ahnlicherHaargestaltungweisen daraufhin, daB das Dach des Naiskos D im ausgehenden
1.Jh. v. Chr. erneuertworden sein muB.49
Mit dem Theaterbereichistdas Simenfragment1/78 (P1.39:e)in Verbindungzu bringen,
das eine Leiste mit einem Reliefmaanderaufweist.50Aufgrund der Fundsituationmag zu
ihm das StirnziegelfragmentATK 1/72 (P1.39:f) gehoren, welches die in der augusteischen
Zeit so gangige Palmettengestaltungmit knopfartigenEinrollungen zeigt.51 Somit hatte
das hellenistischeBuihnengebaudein der friuhenKaiserzeiteine Dachreparaturerfahren.52
Obzwarder untereTeil vom Antefixweggebrochenist laBtsich erkennen,daBdas Ornament
das Akroterschemaeigenwillig reduziert wiedergibt. Typisch dafur sind die nach auBen
eingerolltenRanken, die ohne Verastelungaus dem Akanthuskelchaufwachsen.53
Der Naiskos F ostlich des Theaters wurde als letzter der Naiskoi wahrscheinlichgegen
Ende des 1. Jh. v. Chr. errichtet,54 als im Heiligtum ausgedehnte Sanierungsarbeiten
durchgefuhrtwurden. Von diesem Tempeldach haben sich nur geringfugige Ziegelreste
erhalten. Das AntefixfragmentATK 18/72 (P1.40:a) und Na F 16/84, deren Palmet-
tenblatterin Verdickungenenden duirftenvollig gleich gestaltetgewesen sein wie das Antefix
47 Vgl. dazu die Stucke S. 81, S. 123, S. 124 der Serie 89 bei Le Roy 1967, S. 179-182, Taf. 74.
48 Vgl. dazu OlForsch
IV, S. 70-74.
49 Vgl. dazu den Lbwenkopfvon der Sima der Nordhallein Korinth: OlForsch IV, S. 83-84, 86, Taf. 79, oder
den Kopf einer schmucklosenSima aus Olympia: OlForsch IV, S. 82-83, 86, Taf. 78.
50 Der Reliefmaanderals Dekorationauf Simen oderVerkleidungsplatten erfreutesich schon im 6.Jh. v. Chr.
an der WestkusteKleinasiensgrosserVerbreitung.Der Typus des Traufziegelsmit Reliefmaanderverzierung
an der Stirn tritt erst in hellenistischeZeit auf und wird vom griechischenMutterlandim frtihen2.Jh. v. Chr.
ubernommen. Vgl. dazu Metzger 1971, S. 78, Taf. 19, 185; Billot 1976, S. 105-106, Taf. 26:b, Abb. 3:c.
51 Vgl. dazu Alzinger 1974, S. 109-118 mit Abb.
52 Die Beobachtung stimmt mit den Ergebnissender Bauuntersuchungvon Gogos uberein. Siehe dazu
Anm. 45.
53 Vgl. dazu Amn. 42. Stirnziegelmit ahnlichenMotiven finden sich in Olympia (spates5.Jh. v. Chr.):Harl
1971, S. 88, 135, mit Abb.; Mamousia (hell.): Petsas 1971, S. 186, Taf. 168; Patras (kaiserz.):Oikonomidou
1971, S. 163, Taf. 151; Sikyon Ocaiserz.): Harl 1971, S. 88-89, 143-144 mit Abb. der Serien 6, 7, 8.
54 Zum Naiskos F, siehe Gogos 1986, S. 45-49.
132 DORISGNEISZ
ATK 1/72 (P1.39:f) vom Theater; nur die auBerenKonturenweichen voneinanderab. Die
groBeAhnlichkeitder Antefixevom Theater und vom NaiskosF lalt annehmen, daB beide
Dacher gleichzeitig eingedecktwurden, und daB Naiskos F eventuell die gleiche Reliefsima
wie das Buhnengebaudehatte.
Auch vom Naiskos E gibt es nur eine bescheideneAnzahl von gefundenenDachziegeln.
Der Baubeginn dieses Tempels erfolgte noch im 3. Jh. v. Chr., etwas spater als Naiskos D,
also die Skene bereits aufrecht stand.55 Dieser Bauperiode sind die beiden Simenranken
ATK 169/79 und E 6/83 zuzuordnen, die im Gegensatz zu den fast gleichgestaltetendes
Buhnengebtiudesund des Naiskos D eine engere Windung zeigen. Von den zugehorigen
Antefixenhat sich nur ein bescheidenesFragmentmit einer gesprengtenPalmette erhalten,
ATK 11/72 (P1.40:b).56 Ein Parallelstuckhierfiurgibt es in Delphi,57 das Le Roy in die
zweite HaIftedes 3.Jh. v. Chr.datiert. Dieser zeitlicheAnsatzwurde auch zu unserem Stuck
vom Naiskos E passen.
Das TraufziegelfragmentATK 88 mit dem breiten Astragal an der Unterseite, das
zwischen den Naiskoi D und E gefunden wurde, mochte ich am Tempel E unterbringen,
auch wenn die Sudseite zum Theater hin mit einer Rankensima versehen war. Ahnlich
der Sudstoa von Korinth58kOnntedie hintere Dachflache, die durch den knapp daneben
befindlichenNaiskos D kaum gesehen wurde, mit einer Traufsimaverkleidetgewesen sein.
Von den hellenistischenGiebelsimen wurde ein Fragment vom Naiskos E gefunden,
ATK 1985/E (P1.40:d). Auffallendbei dieserfluichtigausgefuhrtenBltitentanieist der durch
ein Karo angegebene Palmettenkern;die Verbindungsrankenzu den Lotosbtittendaneben
setzen nicht wie uiblichunter der Palmette, sondern daneben an. Ein ahnlich gestalteter
Ornamentstreifenbegegnet uns nur im benachbarten Sikyon, wo diese Schmuckzone an
der Unterseite eines Traufziegelszu finden ist.59 Diese Beobachtung wirft die Frage auf,
inwieweit eine BeeinflussungAigeiras durch Sikyon auf kuinstlerischerEbene erfolgte, das
ebenfalls Mitglied des achaischen Bundes war, und wo gleichzeitig mit Aigeira im 3. Jh.
v. Chr. eine rege Bautatigkeitnachzuweisenist. Leiderwar mir bis jetzt der Zugang zu den
Dachterrakottenvon Sikyon nicht moglich.
Die zu Beginn der KaiserzeitstattfindendenErneuerungenim Theaterbereichdehnten
sich auch auf den Naiskos E aus. Das Fragmenteines Lowenkopfwasserspeiers ATK 3/77
(P1.40:c) weist bereits typische Merkmale der r5mischen Epoche auf: die Haarbuschel,
Augenbrauenund Wimpern sind durch Furchen angedeutet, die man mit einem Stichel in
den noch weichen Ton ritzte. Kopfe mit ahnlicher Bearbeitungfinden sich fruhestensim
vorgeruickten1. Jh. v. Chr.60 StilistischeUberlegungen zu Wasserspeiernin Korinth6l
55 Zum Naiskos E, siehe Gogos 1986, S. 39-45.
56 Mbbius (1968, S. 119, Anm. zu S. 88) hWItPheidias als Erfinderder geflammten Palmette. Von den
Akroterendes Parthenonausgehendwird diese Palmetteformin die BauornamentikObernommenund findet
sich als Stirnziegelim 4.Jh. v. Chr. am Dach der Tholos von Epidauros: Roux 1961, S. 140-145, 160-168,
Taf. 43.
57 Le Roy 1967, S. 180, Taf. 74 mit der Serie 89: S. 181, 182.
58 Siehe dazu Anm. 34.
59 Mylonas 1987, S. 91, Abb. 105.
60 Vgl. dazu die Lowenkopfwasserspeier von einer Rankensimain Olympia: OlForsch IV, S. 82-83, 85-86,
Taf. 94 sowie den Wasserspeier7080 aus Ton in Munchen: OlForsch IV, S. 87, Taf. 78.
61
Vergleichendlassen sich die unpubliziertenStUckeFS 295 und FS 933 heranziehen.
DIE DACHTERRAKOITENVON AIGEIRA 133
veranlassen letztlich dazu, den Lowenkopf vom Naiskos E in die augusteische Zeit zu
datieren.
Als unter Max. Thrax (236-238 n. Chr.) das Buhnengebaude erneuert wurde,62
versuchtendie Handwerkerdie neuen Motive gleich in die Dachornamentikeinzubringen.
Nur wenige Fragmente, die sich zu einem Antefix zusammenfrigenlassen, sind uns von
der Skene erhalten, die zwar nie fertiggestelltwurde, deren Dachterrakottenman aber
offensichtlichim voraus angefertigthatte. Das StirnziegelfragmentE 1/81 (P1.40:e) hat
Parallelenin Olympia und auf Euboa, wo die Antefixe der spaten Kaiserzeitebenfallszarte,
diinne Palmettenblatterzeigen, offensichtlichein Charakteristikumdieser Epoche.63 Der
untere Teil des AntefixesATK 2/72 (P1.40:f) zeigt eine Leiste mit einem Eierstab,der sich
etwa ab der Mitte des 1.Jh. v. Chr. von Italien ausgehendim ganzen romischen Imperium
verbreitete, und mit der Staatsform auch auf die Peloponnes nach Korinth und nach
Aigeira ubergriff.64
BIBLIOGRAPHIE
Alzinger,W 1974. Die augusteischeArchitekturin Ephesos,Wien
. 1985. "Aigeira-Hyperesiaund die SiedlungPhelloe in Achaia (I).I. Die Bauten der nachmykenischen
Zeit," Klio67, S. 426-451
. 1988. "Hyperesia-Aigeira.Der Wandel eines Heiligtumsvon spatmykenischerbis in klassischeZeit,"
IlpaxcxLx61988 [1991], S. 20-23
Billot,M.-F. 1976. "Terrescuitesarchitecturalesdu Mus&etpigraphique,"AeXt 31, 1976, A' [1980], S. 87-135
Buschor,E. 1933. Die Tondicher derAkropolis,
II, Stirnziegel,
Berlin/Leipzig
Corinth I, iv = 0. Broneer, TheSouthStoaandItsRomanSuccessors (CorinthI, iv), Princeton 1954
CorintiI, i = I. Thallon-Hill und L. S. King, Decorated Architectural
Terracottas (Corinth
I, i), Cambridge,Mass.
1929
Deshayes, J. 1966. Argos.Lesfouilesdela Deiras(ttudespeloponnienneIV), Paris
Gogos, S. 1986. "Aigeira-Hyperesiaund die Siedlung Phelloe in Achaia A. I. Theater, II. Naiskoi beim
Theater,"Xlio68, S. 6-50
* 1991. Das TheatervonAigeira.EinBeitragzumantikenTheaterbau, Wien
Harl, 0. 1971. "Studienzu den Stirnziegelnder Peloponnes"(Diss. Graz-Universitat1971)
Heiden,J. 1987. Korinthische Dachziegel.ZurEntwicklung derkorint/dschen
Ddcher,Frankfurtam Main/Bern/New
York/Paris
IsthmiaI = 0. Broneer, TheTemple ofPoseidon (IstkmiaI), Princeton 1971
Le Roy, Ch. 1967. Lesterrescuitesarchitecturales
(Fouilles
deDelphesII), Paris
Mallwitz,A. 1972. Olympia undseineBauten,Munich
Metzger,I. 1971. "Piraus-Zisterne,"AeX-r26, 1971, A' [1973], S. 41-94
Michaud,J. P. 1973. "Chroniquedes fouillesen 1972. Aigeira,"BCH 97, S. 319-321
Mobius, H. 1968. Die Ornamente dergriechischenGrabstelen,Munchen
Mylonas, G. E. 1987. "Epyov 1987 [1988], S. 91
Oikonomidou,M. 1971. <<H&Hpax , AeX-r26, 1971, B1 [1974], S. 149-175
OlForsch IV = E Willemsen,Die Ltiwenkopfwasserspeier vomDachdesZeustempels IV), Berlin
(OlmpischeForschungen
1959
62 Siehe dazu Anm. 45.
63 Euboa: Sampson 1976, S. 139, Taf. 105:p. Olympia: Harl 1971, S. 94, 136, 137 mit Abb. (Antefixe
252 und 253).
64
Vgl. dazu das fruhesteAntefix dieser Gattung in Padua: Strazzulla1987, S. 295, Nr. 367, Taf. IV. In die
mittlerebis spate Kaiserzeitist das sehr ahnliche FragmentA 382 aus Korinth zuweisen. Es ist von schlechter
Qualitat mit starkdegeneriertenOrnamenten. Siehe dazu Harl 1971, S. 93, 124 mit Abb.
134 DORISGNEISZ
Dotis GNEisz
Brucknergasse23
A-7400 Oberwart
Austria
PLATE37
a. A2 b. A4
C.SI
03~~~~~~~~~~~%A A
d.A1 e. Fl
GNuIsz:
DORIuS DfIEDACHERRAKOrrEN
VONAIGEIRA
PIATE 38
II A'',
a.A14 b. E 5A
f4 f
d.ATK 21/72
_l ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I
24/88
c. E 24/88
e. ATK 19/72
DORIS GNEISZ:
DIE DACHTERRAKCOT~rEN
VON AIGEIRA
PLATE 39
*
I'
b. ATK 1988
a. E 6a/88
d. ATK 1987
c. ATK 1986
Att7 '
f ATK172
ATK 1/72
~~~~~~~~~f.
e. 1/78
GNiusz: DmEDACHTERRAKOTTEVONAIGEIRA
DORIuS
PLATE40
a.ATK18/72
b.ATK 11/72
e. _1/81 f. ATK2/72
Deckziegel notwendig waren. Die Ziegel dieses Daches lassen sich zu einem keilartigen
Dach, das ein Gebaude mit dreieckigemGrundriBdeckte, zusammenfugen.
Nun stellt sich die Frage, welches Gebaude in Olympia mit diesem Dach gedeckt
war, denn ein in seiner GrundflaichedreieckigesGebaude ist uns vorderhand unbekannt.
Einen ersten Hinweis auf die Lage des Gebaudes geben uns die Fundortangaben. Die
Ziegelfragmentedes Daches sind fast ausschlieBlichim Suidostendes Heiligtums gefunden
worden. Auf der Suche nach einem dreieckigenGebaude in Olympia hilft uns Pausanias
(5,15,5-6; 6,20,10) mit seiner sehr genauen Beschreibungdes Hippodromsweiter.
Hierbei geht es vor allem um die komplizierteStartvorrichtungfur die Pferdegespanne.4
Zusammengefasstbesteht die Startanlage aus der Agnaptoshalle und einer dreieckigen
Konstruktion, die an die ostliche Langseite der Halle anschlieBt. Pausanias betritt die
Startvorrichtungund trennt bei der Beschreibung ausdruicklichzwischen dem Teil, der
unter freiem Himmel liegt und der Spitze, dem Embolos (Keil), in den er von dem nicht
gedecktenTeil aus hineingeht.
Als Architektender Starteinrichtungnennt Pausaniasuns den Athener Kleoitas. Er ist
uns aus mehreren attischen Inschriftenbekannt und lebte und wirkte in der Zeit unseres
Daches.5 Pausanias hat dieses Dach nicht gesehen, da es zu seiner Zeit zumindest zum
Teil schon lange unter der Erde war. Der Embolos der Starteinrichtungist aber nach
dem Erdbeben wieder mit schragen Ziegel, von denen sich einige Exemplare erhalten
haben, ausgebessertworden. Das Embolosdachder Starteinrichtungduirfenwir als ersten
konkretenHinweis auf das im Mittelaltervom AlpheiosweggerisseneHippodrom werten.
Ein weiteres, sehr ungewohnliches Dach gehort zur Vorhalle des Bouleuterionsvon
Olympia. Da fast alle Ziegel des Daches im oder vor dem Bouleuteriongefunden wurden,
ist die Zuweisungso eindeutigwie in Olympia selten. Meist wurden die Ziegel eines Daches
uber das gesamte Heiligtum verstreut aufgefunden. Die Traufe des Daches besteht aus
reliefiertenTraufziegeln(P1.42:b), aufdenenje zwei frei gearbeiteteVoluten (P1.42:c) sitzen.
Uber der Fuge zweier Blockestehteine doppeltePalmette. Der WasserabfluBin der Mitte ist
sehr abwechslungsreichgestaltet.
Vermutlichwechselten sich in gleichmaBigemRhythmus Lowenkopfe(P1.43:a), Gor-
gonen (P1.43:b)und Sonnenblumenmiteinanderab. Im Gegensatzzu den Lowen waren die
Sonnenblumenund die Gorgonen keine Wasserspeier,sie verdecktennur den AusfluB.Das
Wasserlief hinter ihnen aus der Traufziegeloffnungheraus. Die stilistischeEinordnungder
Gorgo und der Lowen, aber auch die Fundumstandeeiniger Fragmente, die wohl schon
beim Versetzen des Daches in die Erde gekommen sind, legen eine Datierung des Daches
in das Jahrzehnt 460-450 v. Chr. nahe.6
Die Vorhalle ist dadurch datierbargeworden. Sie ist etwa 100 Jahre alter als bisher
angenommen. Da in ihrem Fundament ein Geisonfragmentdes Zeustempels gefunden
wurde, folgerteA. Mallwitz, daBdie Vorhallenach dem Erdbebenvon 373 v. Chr., bei dem
der Zeustempelbeschaidigtwurde, entstandenist.7 Das Geisonfragmentkann aber ebenso
8 XVII, S. 1-4.
Zuletzt:OlForsch
9 Das Beben von 402 v. Chr. im nahegelegenen Elis (RE Suppl. IV, S. 350 [Capelle]) scheint fur die
Zerstorungenin Olympia zu spat zu sein.
138 JOACHIM HEIDEN
in Achaia, beim dem Helike vom Meer verschlungenwurde und Bura in einer Erdspalte
versank, ist in der antiken Literatur mehrfach erwahnt.10 Da A. Mallwitz nachweisen
konnte, daB der SO-Bau im Jahre 364 v. Chr.wahrend einer Auseinandersetzungzwischen
Arkadernund Eleern im Heiligtum bereits fertig war, muB er also in den Jahren um 370
v. Chr. entstandensein.11
Der Dachrand des SO-Baus war an den Giebel- und den Traufseitenvon der gleichen
Sima eingefasst.An der Traufewaren die Simablockein ihrerMitte durchbrochenund mit
Wasserspeiernausgestattet.Die Ornamentikder Giebelsima(P1.44:c) ist von der Traufsima
nur durch ein kleines Detail zu unterscheiden. Das Mittelblatt des Lotosbluitenkelches
ist nicht gezackt umrandet. An der Traufe wechselten sich Blocke mit LOwen-und mit
Ttullenwasserspeiern (P1.45:a, b) ab. Die auBergewohnlichenTufllen(P1.46:a) wachsen aus
einem plastischenAkanthusbtischelund sind von einer Ranke eingefasst. Daher sind sie
eher dem pflanzlichenals dem tierischenBereich zuzuordnen.
Das Dach ist deswegen von Interesse,weil das gemalte Ornament erstmalsin die dritte
Dimension hervortritt. Man mochte diesen Dachrand entwicklungsgeschichtlichgerne fur
alter halten als die plastischenRankensimen. Das scheint sich aber zu verbieten, denn die
ersten Marmor-Rankensimenvon Delphi und Epidaurosmussen wohl doch etwas fruher
datiertwerden.12
In der Mitte des 4. Jh. v. Chr. begann man, das olympische Heiligtum mit groBen
Gebauden zu rahmen. Es entstanden das Leonidaion und die Suidhalle. Die Echohalle
im Osten wurde auch zu dieser Zeit geplant und begonnen, konnte aber erst in der fruhen
Kaiserzeitvollendet werden. Der Dachrand solcher hallenartigerBauten wurde meist von
Rankensimenbegrenzt.13Allein aus der 2. Halfte des 4.Jh. v. Chr. haben sich in Olympia
Rankensimenvon ffinfverschiedenenGebauden erhalten.
Der bekanntesteist der mit fast 1000 Fragmentenvertretene Dachrandsschmuckdes
Leonidaions (P1.46:b). Die Rankensimaist im Laufe derJahrhunderte6fters ausgebessert
worden. Auch die Rankensima der Tafel 123:4 der Olympia-Publikationzahlt zu einer
solchen Ausbesserung,denn sie wiederholt in allen Motiven und MaBen exakt die alteren
Stucke des Leonidaions. Diese Reparaturwird im 3. Jh. v. Chr. stattgefundenhaben. Fur
die Suidhalle,der R. Bormann diese Sima zuwies,14 ist sie aber zu spat zu datieren. Eher
gehort zur Sudhalle eine Traufsima(P1.47:a), deren Rankenmotiv durch kleine zwischen
den Voluten herabhangendeBluten bereichertist.
Die wegen ihrer Tonfarbe sogenannte "Rote Rankensima" (P1. 47:b) laBt sich mit
Baugliedern einer ionischen Halle verbinden, da die Dubellocher im Geison und in den
Ziegeln sich in Formund Abstandgleichen. Wo allerdingsdiese Stoa gestandenhat, ist nicht
bekannt. Keines ihrerArchitekturfragmente wurde in situgefunden.
10 Neumann und Partsch 1885, S. 324-325; RE VII, S. 2855-2856 (Btite);RE Suppl. IV, S. 350 (Capelle).
" Mallwitz 1981, S. 109-11 0.
12
Heiden 1987, S. 130-131, 181-188.
13 Heiden 1987, S. 198.
14
OIVMpiaII,S. 199.
KLASSISCHE
DACHERAUS OLYMPIA 139
BIBLIOGRAPHIE
Ebert, J. 1989. "Neues zum Hippodrom und zu den hippischen Konkurrenzenin Olympia," .Mkephoros 2,
S. 89-107
Floren, J. 1977. Studienzur TypologiedesGorgoneion,
Monster
Heiden, J. 1987. KorinliiscleDachziegeLZurEnticklug derkorinddschen Dacher,Frankfurtam Main/Bern/New
York/Paris
Hitzig, H., und H. Blrmner. 1896. DesPausanias Beschreibung vonGriechenland
I, i, Leipzig
. 1901. DesPausanias BeschreibungvonGriechenlandH, i, Leipzig
Le Roy, Ch. 1967. Lesterrescites architecturaks
(FouilksdeDeiphes ), Paris
Mallwitz,A. 1972. Olympia undseineBauten,Munchen
. 1981. "Neue Forschungenin Olympia (Theater und Hestiaheiligtumin der Altis)," Gmnasium88,
S. 97-122
Miller, Stephen G. 1988. "Circular Roofing Systems and the Athenian Tholos," in IIpaxtx& -ou XII
AteOVO6q Suvebpou KXaLX4q ApXatoXoytaq,AOva, 4-10 Eewtre#ppou 1983, I, Athens, S. 134-139
Moustaka,A. 1984. "FrdhklassischeLzwenplastikaus Olympia,"AM 99, S. 177-183
Neumann, C., und J. Partsch. 1885. Physikalische Geographie Breslau
vonGriechenland,
OUBer XI = BerichtfiberdieAusgrabungen in Olympia,
Berlin (in Vorbereitung)
OlForsch XVIII= W Schiering,Die Werkstatt desPheidiasin Olympia, ZweiterTeil(OlympischeForschungen
XVIII),
Berlin 1991
OlympiaII = W. Dorpfeld und R. Bormann, Die Baudenniler (Olympia,Ergebnisse Reich
dervondemdeutschen
veranstalteten
Ausgrabungen II), Berlin 1892
Thompson, H. A. 1940. 7he TholosandItsPredecessors (Hesperia Supplement4), Princeton
Wiegartz, H. 1984. "ZurStartanlageim Hippodromvon Olympia,"Boreas7, S. 41-71
JOACHIM HEIDEN
Deutsches ArchaologischesInstitut
Via Sardegna 79
I-00187 Roma
Italien
PLATE 41
CJJ
ad ~ ~~~~~ -t
_
U
~~~~~~~~~~~~~~~~5
3
>
1 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~i3
4w,~~~~~~~~~~~~~~~~~~~~~~~~~~4 A~~~~~~~~
weak
# # An.~~~~~~~~~~~~~~~~~~~
v -
e
PLATE43
CIO ~~ ~ ~ ~ ~ ~ I'
07.~ ~~ ~ ~ ~
0 C#2~~~~~~~~~~
Cd~~~~~~~~~~~
PLATE 44
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
c. Giebelsima 11K133
AUS OLYMPIA
JOACHIMHEnIDEN:KIASSxSCiiEDACHER
PLATE 45
:~XS ~ r
a. Traufsima 11K24
b. Traufsima 1lK50
4 ~ ~ ~ ~ <
.~~~~~~~~~~~1F
a. Tullenwasserspeier11K33
JOACHIM
HEIDEN:Ki.SSISCHE~DACHERAUSOLYMPIA
PLATE 47
And .~~~~~~~~~
THE STAMPS
The stamps that occur exclusively on Laconian-typepan and cover tiles bear additional
witness to the public and sacred character of the buildings. The most common stamp,
representedby thirty-fiveexamples,bearsthe adjectiveAAMOI0OE (= public tiling)'0with
1 Blouet 1831, I, pp. 27-33. Le Bas 1844, pp. 422-423. Sophoulis 1895, p. 27. Oikonomos 1909,
1925-1926. Orlandos 1963a, 1965, and the reportswrittenby Orlandosfor Hpaxux&, 1957-1975.
Unless otherwiseindicated,all inventorynumbersare those of the Museum at Messene.
The photographsfor Plate54:a and b come fromAnastasiosOrlandos'archives,keptby the Archaeological
Society at Athens. All the rest are due to the photographerVassiliosStamatopoulos.The topographicalplan,
Figure 1, was done by the topographerGeorge Makris,while the plan, Figure 4, was done by the architect
Dr. AthanasiosNakassis. The sima, Figure 19:a and b, was drawn by the painter EustathiosAndroutsakis;
all the rest were drawn in pencil by the author and inked by the drafterEvangelosOlympios.
2 Themelis 1989, pp. 35-36 fig. 35.
3 Themelis 1988, p. 72, fig. 15.
4 Themelis 1987, pp. 73-79; Themelis 1988, pp. 45752.
5 Themelis 1990, pp. 26-27.
6 Themelis 1987, pp. 79-90.
7 Themelis 1990, p. 35, fig. 45.
8 Themelis 1988, pp. 59-65.
9 Themelis 1987, pp. 90-96; Themelis 1988, pp. 65-72; Themelis 1989, p. 37.
10 Cf. the AAMOIOstampsat Nemea (see the articleby Stephen G. Miller in this volume, pp. 85-98 above)
and a AH MO[E10E] stamp at Pella: Kaltsas 1988, p. 101, note 314, which refers to examples from Illyria,
142 PETROSTHEMELIS
relief letters 18 mm. high in an oblong frame 2.8 x 12 cm. (Fig. 2:a; P1.48:a, b). A second
circularstamp, 6 cm. in diameter,of which only three examples exist, carriesthe ligature A
(= AA<MOE10E>)" and the initials EY underneath, most probably an archon's name
(Fig. 2:b, PI. 48:c). The third rectangularstamp, 5.4 x 13 cm., displays the words Efli
IEPEOE with relief letters 1.6 cm. high on the first line and a name in the genitive on the
second (Fig. 2:c, d). Only the name 4lAInnOE, during whose priesthood these tiles were
made and stamped, is better preserved on one of the six extant specimens; no doubt he
was the priestof Zeus Ithomatas,who also functionedas eponymousof the year accordingto
the testimony of certain inscriptionsdealing with state and cult officials as 'AywvoOxatL,
kiepooirat,rpapsaireitS,XaXetLop6pot, 'Ayopav6tiot, and 'Tnayopav6iot.L12The only
4Xutmnos we know from Messene is the fatherof the famous sculptorAcocpuv, mentioned
in seven honorarydecreesinscribedon a stone column found in the Asklepieion.'3 Phiippos
must have been active in the second half of the 3rd century B.C. and could have held the
office of eponymouspriestof the supremegod of the city duringthis period. The letterforms
on the stamp do not contradictsuch a date.
A stone standard found in 1960 at the southwest corner of the agora, north of the
Asklepieionand similarto those from the AthenianAgora and Assos,14shows the officialsize
(L. 96 cm.) and shape of terracottapan and cover tiles of Laconian type (P1.48:d).
LACONIAN ANTEFIXES
Piraeus,Athens, Tanagra, Sparta. The rarityof AAMOEIOE stampsin Macedonia, where the BAI/AIKOE
ones are the norm, has probablyto do with the differentpolitical systemthere.
" A similarligatureA occurs on rectangulartile stampsfrom Illyria:Andrea 1976, p. 348, pl. XIV: 14-16.
12
IG VI, 1467, 1468, 1469. Orlandos 1969, pp. 103-104, pl. 124:a;Themelis 1988, pp. 64-65.
13
Orlandos 1972, pp. 137-138; Themelis 1989, p. 35, fig. 34; Themelis 1990, p. 32.
14
Orlandos 1955, pp. 112-113, figs. 63, 64; Camp 1986, p. 128, fig. 106.
15 Themelis 1991, pp. 28-30. The fill to the north of the Ortheia temple where fragments of Laconian
antefixType 1 were found contained pottery fragmentsof the 3rd/2nd centuryB.C., terracottafigurines,and
two coin hoards dated to the middle of the 2nd century B.C. and related to the abandonment of the Ortheia
temple and the constructionof the Asklepieioncomplex in the firsthalf of the 2nd century B.C.
16 Cf. the terracottaArtemis heads on semicircularvotive tablets from Ithaka in the National Museum:
Kalligas 1979, pp. 59-61, pl. I, IET, especiallyIZ' with incised S-shaped spiralsaround the head of Artemis
Kyparissia(of South Italian inspiration?) and on a Laconian antefix of the 2nd century B.C. (Karapanos
Collection 1148): Winter 1903, p. 166, fig. 4. Kalligas 1980, p. 26, fig. 18. For the presence of protomes
and heads on architecturalterracottas,see p. 153 below.
z CD
I
w
I I,~~~~~~~~~~~~~~~
0 .a~-c
x U~~~~~~~~~~
C
x c..~d
X (x
>7-40 c~d
w ~~~~. -C
)1>
0.
I~~~~~~~~~~~~~~~~~~~~~
o I .~~~~~~~~~~~0 w
o 1~~~~~~~~~~~~~~~~c 1c
0. I~~~~~~~~~~~~~
0I I1
144 PETROS THEMELIS
I'm~~~~~~~I
0E-'
dX
2 5cm.
. ~ ~ ~ ~ . . .-.~~~~~. .....
. ':,,,,.,
, ,,, ~~~~~~~~~~~~~~...........
.,,,,,,,,,,,,,,,,.,................................
*.-.-*....
X . . . . . . . . . . . . . _.,.,,,. . , ,,
, ,,,,
....,. ,,...... . .,--...................
.. ...................... ........;:;
.......
A..................................... : ., ,, -,, ..,. , ,.................:........
,,,,,
the;styl of th gods'
a.;sk
well~~~~~~~, hea, th
FIG. 3. Laconian odln
antefix of her
1 with
Type fac
bust of Artemis he left and
?turnedsightl to8.z
of UArtmi Ortea Som of the Ef
.fthistyp of anei to the roo of theI.smal temp.;leB.
o~~~~~~~~~~a
the slender almost triangularform of the antefix (with the vertical sides
lightly convex) as
well as the style of the goddess'shead, the modeling of her face turned slightlyto her left, and
her hairdressbound with a stephane allow a date in the 3rd centuryB.C.and the attribution
the roof the small temple of Artemis Ortheia. Some
of this type of antefix to off Ecofe the
IsEuPE stampsmay come from the ties ofthis temple.
Laconian antefixes of Type 2 (Fig. 5; PI. 49:a) have been found on the north raised
wing of the AsklepieioncomplexI 7 where most of the AAMODO0Estamped tiles were also
18
uncovered;in theWestParodosof theTheater,probablycomingfromtheroofof theskene;
and at the east side of the Stadium, close to the stone thrones of the proedria (PI. 49:b),
deriving from the roof of a wooden shelter that protected the officialsof the contests from
sun and rain. Post holes at regular intervalshave been brought to light on the pavement
in front of and behind the stone seats.
The originalityand high quality of this Messenian mold product are obvious; from a
shaggy and voluminoustripartiteacanthusemergesa dominatingsix-petaledflowerand two
17 This north wing, composed of two symmetricallydisposed and divided large rooms on both sides of the
north propylon, the original function of which was related to Asklepios'cult, was used as the Sebasteion or
Caesareumin the Augustanperiodwhen the cult of this emperorand Roma found here an appropriateplace to
settle: IG V1, 1444B, 1462. Groag 1939, p. 115; Orlandos 1959, pp. 170-171; Orlandos 1969, p. 103. SEG
XXIII, 1968, p. 207; Migeotte 1985, pp. 597-607; Hanlein-Schafer1985, pp. 162-163, A 23; Themelis 1988,
pp. 52-58.
18
Themelis 1988, pp. 48-52, pl. 38:b.
146 PETROSTHEMELIS
ii~~~~~~r
13~~~~~~~~10
.. ~~~~~~' \
0.28
b -
-1
+ ~ ~~~
-0.114 -
greatestfame and sculpturalproductionof Damophon, who made all the marble cult statues
in the Asklepieion(Pausanias4.31.10).19 Fragmentsof Type 2 Laconian antefixes, as well
as fragmentsof Laconianpan and cover tilesbearingthe AAMOEI0E stamp,were found in
fill to the east of the Sebasteion, rather securely dated on the basis of coins, pottery, and
stampedamphorahandlesto the second half of the 2nd centuryB.C.,20 a terminusante quemfor
the dating of this type of antefix.
A fragmentof Laconianantefix(inv.no. 1716)(P1.49:c)of ratherpoor workmanshipcon-
stitutesa variantof Type 2. The linearand dry styleof its design in low reliefindicatesa later
date;it may be attributedto the repairof the horizontalcornice (ntapaeoC,)of the Caesareum
by the "Quaestorpro Praetore"Marcus Caesius Gallusin the 1st century after Christ.21
Type 3 Laconianantefixis representedby a chance findwith no provenanceand of small
size, 11 cm. high (Fig.6; PI. 49:d). Its high relief decorationis composed of a thick tripartite
acanthus from which two horizontal and two vertical spirals emerge; a broad seven-leaf
palmette springs up from a central, higher placed, acanthus leaf. It is related to Type 2
as far as the high relief and the main elements of the design are concerned.
To the Type 4 Laconianantefixis ascribeda uniquepiece (P1.49:e)foundby chance in the
fieldssouth of the city wallsin an area of cemeteries. Its.form is almostsquare(18.2 x 18 cm.)
ending in a trianglewith concave sides;it is decoratedwith the head of a Medusa (Rondanini
19 The cult statues,as theirbasesindicate,are closelyconnectedwith the architectureand contemporarywith
it. Theftoruitof Damophon is assignedto 180-160 B.C.: Bol and Eckstein 1975, pp. 83-93; Levy and Marcade
1972, p. 986; RE Supplement XV, 1978, s.v. Messene, col. 288. (E. Meyer); Despinis 1966, pp. 378-385;
Habicht 1985, p. 57. The decrees honoring Damophon (son of Philippos)mentioned above (p. 142) offer
additional evidence concerning the areas and the fioruitof his activity: Themelis 1990, pp. 31-32; Themelis
1993, pp. 34-39.
20 Themelis
1988, pp. 54 (trialtrench Z) and 74, pls. 41-44.
21 IG V1
1462. Cf. Bardani 1986, pp. 79-81.
HELLENISTIC ARCHITECTURAL TERRACOTTAS FROM MESSENE 149
:7~~~~~~~~~~~~~~I
0 5 cm.
I
FIG.7. Fragmentsof Laconian disc acroterionof the 6th centuryB.C. (inv.no. 1701)
type)22in high relief: the fleshy and rounded face of the female daemon is surroundedby
snakes mingled with her hair, while two short wings sprout from her forehead. Despite
weathering, the plastic virtues of the work are undeniable and comparable to classicizing
sculpturalworks of the 2nd century B.C. and in particularto the Tritonesseson the throne
of the Damophontian Lykosouradivinities.23The form of this Laconian antefix as well as
the structureand motif of its decoration with a frontalMedusa head occupying the entire
front plaque without vegetal decoration is unusual for the Greek mainland, where figural
representations(heads,busts,whole figures,or animals)normallyoccupy the centerand lower
part of the pattern, being rathersubordinatedto the dominant vegetal design composed of
spiralsand palmettes. South Italy and Sicily seem to be the source of inspirationfor this type
of antefix;it will sufficeto refer to the extensive series of Medusa-headantefixesin baroque
style from Morgantina.24Some examplesfromAsia Minor and the BlackSea seem to derive
from Italianprototypesas well.25 The Medusa medallionon a bowl from Phaistos26dated to
the period of Antiochos III (223-187 B.C.), of similar structurebut more "pathetic"than
ours, and the applied Medusa head in the middle of a vegetal relief frieze on polychrome
22 Buschor 1958, pp. 26-27.
23 Tritons and Tritonessesof Lykosoura:Karouzou 1967, p. 170, nos. 2171-2175, pl. 58.
24 Uhlenbrock 1988, pp. 139-140; Kenfield 1994; cf. Brandes-Druba1994.
25 Cf. Stella G. Miller 1994; Zimmermann 1994; a peculiar type of Hellenistic antefix from Samos (frontal
The Corinthianantefixescan be tentativelydivided into ten main types on the basis of their
design and ornamental structure;seven of them have been associatedwith corresponding
simas. These main types are distinguishableon the evidence of their common stylistictraits
as four groups. Their relativechronologicalsequence is based partlyon contextualevidence
and partlyon their attributionto the roofs of securelydated buildings.
Group A or "ClassicalStyle," of the 4th century B.C. comprising one type of antefix
(Fig.8; P1.50:a).
Group B or "Plain Style," dating to the 3rd century B.C. and consisting of five types
(Types2-6) of antefixesand four simas (Figs.9-17; Pls. 50:b-52).
Group C or "Rich Style," which dates to the first half of the 2nd century B.C. and
comprisestwo types (Types 7 and 8) of antefixesand four simas (Figs. 19-23; Pls. 53, 55:a, b,
56:a, b).
Group D or "Late Style," comprisingtwo types (Types 9 and 10) of antefixes(Figs. 24,
25; PI. 56:c, d).
New finds will certainly enrich the material and will complete or modify the typology
and the grouping.
A. CLASSICAL STYLE
To GroupA belongsthe antefixType 1 (Fig.8; PI.50:a)representedby a few fragmentary
examplesuncoveredto the northwestof the Asklepieionnear the retainingwall of a sanctuary
adjacentto the Ortheia temenos; this sanctuary,not mentioned by Pausanias,is not related
to Athena Kyparissia,as we first suggested,30but to a hero cult, according to more recent
27 The date of these vases is disputed,but a late 2nd-centurydate seems probable: Pollitt 1986, pp. 194-195,
fig. 208, bib. no. 9.
28 Themelis 1965, p. 207, pl. 213; Yalouris1968, pp. 57-65 and note 1 with bibliography.
29 Themelis 1987, pp. 87-88, fig. 12, pl. 68:b. E. Meyer's rather dogmatic statement that "irgendwelche
alterenSiedlungsfundeliegen von der Stelle nicht vor"in RE, SupplementXV, 1978, sxv.Messene, p. 137 is not
correct.
30 Themelis 1988, p. 63; SEGXXIII, 1968, pp. 209-210.
HELLENISTIC ARCHITECTURAL TERRACOTTAS FROM MESSENE 151
0185 - 0.126-
(1 ~ ~ ~ ~ ~
0.16 -0-5 13
finds. The antefix is decorated with a pair of broad-eyed volutes joined by a channel to
an acanthus; a quite accuratelymodeled and fine palmette of eleven leaves emerges from
a tongue-shapedheart. Two modest stems spread horizontallyto both sides of the heart
below the palmette; the volute zone is low and wider than the anthemion one. The light
ocher-painted relief contrasts with the dark-paintedbackground. This type of antefix is
attestedmainly in the Argolid,where it is dated earlyin the 4th centuryB.C.31 Our example
cannot be earlier than 369 B.C., and a date in the second half of the 4th century seems
rather secure; it is the earliest architecturalterracottafound at Messene thus far, after the
foundationof the new city.
B. PLAIN STYLE
3' Waldstein 1902, p. 130, pl. 23; Deshayes 1966, p. 230, pl. 51.
32 Billot 1976, p. 123, note 93 with bibliographicreferences. Kaltsas 1988, pp. 69-71, notes 200-205,
colorplate VII, pis. 8, 9, 15, 20, 21; Vlachopoulou-Oikonomou 1986, pp. 114-187 and 279, pis. 10-15;
Andreou 1983, pp. 20-21, pis. 6, 7, 8:a.
33 Budina 1994.
34 Bauer 1973, pl. 30:2; Ehrhardt1985, p. 275, fig. 48.
35 Hoepfner 1990, p. 7, fig. 10.
HELLENISTIC ARCHITECTURAL TERRACOTTAS FROM MESSENE 153
and with a painted astragalon the underside. The terracottaantefixesof the small temple of
Pamisosat Aghios Floros in Messenia are similarto the ones from Messene (Fig. 18).3
The slightlylarger antefix Type 3 of the same group (Fig. 10; PI. 51 :a) is similar to the
previous one as far as composition and motifs are concerned. Different are the depressed
lyre pattern, the appearanceof acanthushalf-leavesalong the ground line and on both sides
of the upper spiralson the antefix, as well as the single twist of the spiralson the sima and
the emergence of acanthus and cauliculifrom the angle between the fascia and waterspout
(Fig. 15; PI. 51:b, c).
Type 4 antefix and sima of Group B (Figs. 11, 16: Pls. 51:d, 52:a) consists only of one
antefix and two fragments of simas, one of which was found at the sanctuary of Artemis
Limnatis on the south slope of Ithome (Fig. 1, no. 7),37 dated mainly on the basis of its
architecturalstyle to the late 3rd centuryB.C. The sima is similarto the previousone except
for the smaller dimensions, but the antefix demands special attention: the combination of
a female bust portraying a divinity,presumablyLimnatis, with an abstract vegetal motif
composed of a seven-leafedpalmette that emerges from her head like a high crown and by
pairs of spirals on both sides of her shoulders are closely related to the Laconian antefix
Type 1 from the Artemis Ortheia temple as far as style and motifs are concerned. The
bust of the goddess on both types occupies the centrallower area replacingthe heart of the
palmette and the lyre pattern. This structuralfeature, with the palmette still dominating
as the main ornamentalelement in contrastto the South Italian parallels,where the frontal
face of the figure is the dominant decorative element occupying most of the plaque of
the antefix,38 suggests a rather mainland Greek and especially Peloponnesian source of
inspiration.39Corinthianworkshopsof the late 4th or early 3rd century B.C. seem to have
introduced figural subjects into the vegetal repertory of architecturalterracottasattested
since the Early Archaic period; the innovation spread quickly and was adopted by local
workshopsnot only in the Corinthiaand the Argolidbut also in Arkadia,Messenia, Elis (and
Olympia), Akarnania,Phokis, Epiros, Illyria, Kerkyra,and elsewhere.40 The repertoryis
relativelyrich, comprisingdivine or daemonic heads, busts,whole figure, and masks,as well
as animalsthat are not confined to the antefixesbut occasionallyappear on the simas as well
among vegetal motifs and continue to be reproduceddown to the Roman Imperialperiod.
Even though most of the Hellenisticarchitecturalterracottas(includingthe figuralones)
from the above-mentioned regions still remain unpublished, the material known so far,
althoughscanty,allowsthe followingtentativeconclusion: except for subjectswith a general
apotropaicand chthonic Dionysian character,such as Gorgon-Medusas,silens, satyrs, and
36 Valmin 1938; Papachatzis 1979, p. 112, fig. 38.
37 Le Bas 1844, pp. 422-423. IG V', 1442, 1458;Themelis 1988, p. 72, fig. 15.
38 See notes 24 and 25 above.
39 Cf. Billot 1976, pp. 131-132, notes 135-138; Hubner 1976, pp. 175-183; Pliny,NatHist35.152 says that
Butadesof Sikyon "primuspersonastegularumextremisimbricibusinposuit."
40 Billot 1976, p. 423, note 93, p. 131, note 135, p. 133, note 146. Hubner 1976, pp. 175-183; Vlachopoulou-
Oikonomou 1986, pp. 165-183, 192-210, pis. 17-19. Corinthianinfluence in Illyriathrough the Corinthian-
Corcyraeancolonies of Epidamnosand Apollonia dates from the late 7th and 6th centuryB.C. From the early
4th centuryB.C., however,South Italianelementswere introducedin the region afterthe extended colonization
under the leadershipof the tyrantof SyracuseDionysios the Elderwho founded Issa (islandof Vis) in 397-390
B.C. and Pharos (islandof Hvar) in 385/384 B.C.: Stipcevic 1977, pp. 38-39.
154 PETROSTHEMELIS
1 _.
A _ ''j~9
0.67-
0.16 -
C. RICH STYLE
The two types (7 and 8) of antefixes and the four simas attributed to Group C are
outstanding representativesof this style of the first half of the 2nd century B.C., a period
during which Messenian arts and crafts, including architecturalterracottas,reached their
zenith (Figs. 19-23; Pls. 53, 55:a, b, 56:a, b).
Type 7 of this group is the best preserved,consistingof one antefix (Fig. 19:c)and many
complete simas not only lateralbut also raking(Fig. 19:a,b; PI. 53:a-d), found mainly in the
area of the agora north of the Asklepieion. The lateralsima of this type, although adorned
with the typical motifs (for example, waterspoutsin the form of a lion's head separated
by double acanthus scrolls)is distinguishedby the dense arrangement of motifs and the
especiallyhigh relief;the stem of the spiralsis partlytwisted,while flowers emerge not only
from the spiralcontacts but also from the acanthusthat crowns the cauliculus.44The lion's
heads are freelymodeled (notcast in a mold as is the case with the whole sima decoration)and
appliedseparatelyto the spoutopeningbeforefiring.45The deep depressionson both sidesof
the lion's foreheadservedfor saferattachment. The turbulentmanes cover most of the basic
acanthus foliage and part of the start of the spirals. The free modeling presumablymade
by differenthands produced differentresultsapparenton the two better-preservedexamples
(PI.53:a), which would undoubtedlyhave led to falsely differentiatedstylisticevaluation of
the two animals'heads had they happened to be found detached from the sima.
The special features of the style are even better illustratedby the antefix (Fig. 19:c;
PI. 53:a) and especiallythe rakingsima of the same Group C (Fig. 19:a, b; PI. 53:b-d). No
descriptioncould be considered satisfactoryfor such sculpturalworks in clay: two winged
female figuresvery much like vegetal Nikai nakedabove the thighs are flyingfrom both sides
toward a floral centerpiece that forms the point of attractionor object of veneration; the
lower part of each body is shaped into an elaborateand upward-curvedacanthusleaf from
which emerges a thorny spiral with double curl; similarlyshaped are also their acanthus
42 Forerunnersof this type of sima and flower see in Kaltsas 1988, p. 75. Cf. also Dyggve, Poulsen, and
Rhomaios 1934, fig. 47.
- 43 See notes 17 and 21 above.
4 See note 41 above. Cf. Billot 1976, nos. 14-21, pl. 25. Dyggve, Poulsen, and Rhomaios 1934, fig. 47;
Kourouniotes 1911, p. 68, fig. 69; Hoepfner 1990, pp. 8-9, fig. 16 (= Artemisionin Magnesia),p. 10, fig. 38
(= Smintheon in Troas).
45 Cf. the lion's heads of similarmodeling in Billot 1976, pl. 25. Gneisz 1994.
0
-9 k'o -0
C
Q
-k9XO- C
4"0
Cd
Cd
Cl) ~~ ~ ~ ~ C
: m;
.....i....g W~~~~~
I , i 4arm 1
B y A o
We~~~~~~~~~~~~~~~~~~~ n O~~~~~~~~~
\
t L i A a~~~~~~~~~~
HELLENISTIC ARCHITECTURAL TERRACOTTAS FROM MESSENE 157
wings. Both vegetal women touch or take hold with their outstretchedhands (the right one
with her right hand) the strongly curved and twisted sheaths of the cauliculi, which rise
symmetricallyon both sides of the central floral pattern and end in flowers. Horrorvacui
seems to have guided the artist'shand to fill all gaps between the vegetal Nikai and the
central pattern with all sorts of large flowers that hang from long flexible stems springing
up from the acanthi in all directions. The central floral pattern is repeated in larger scale
and slightly variegated on the antefix (PI. 53:a) of the lateral sima: two twisted S-shaped
stems ending in spirals, acanthus, and flower buds between them spring up boldly from a
tripartiteasymmetricacanthus;a large cornucopia-likeflowerwith a thorny calyx emerging
from the central acanthusleaf occupies the space between the S-shaped stems.
The problemof originand sourceor sourcesof inspirationof ourvegetaltheme combined
with "plant figures"demands a more detailed analysis. Vegetal ornamentation in Greece
experienced an essential structuraltransformationfrom the late 5th century B.C. onwards.
Vigorous naturalisticand sometimes complex vegetal representations,including acanthus
leaves, the lily, and other flowers, appear on the side or gradually take the place of the
conservative and paratactic groups of ornaments in architecture and minor arts. The
invention of the "Corinthian"capital (attributedto Kallimachos)probably had a share in
its adaption of the new ornamental taste, while coroplastsand metalsmithscontributedto
its diffusion. New vegetalthemes with distinctiveregionaltraitsbecame especiallypopularin
South Italy already in the early 4th century B.C., indicated by numerous examples mainly
on the neck of red-figuredApulian vases. Thin volutes and spirals in perspective illusion
and all sortsof large flowersin groupsor in tiers("Dreiblutengruppen","Stockwerkbltitten")
and in various phases of growth as well as thin stems emerging from a central acanthus
calyx often crowned with a human, usually female, head are some of the characteristics
of South Italian subjects. South Italian ornamental taste seems to have had a certain
influence (throughimports, travelingartists,or pattern books) on Macedonian, Thracian,
and especiallycentral Italian architectureand minor arts (tomb painting, mosaics, textiles,
jeweler, etc.) of the late 4th and early 3rd centuryB.C.46 The influencewas especiallystrong
on Etrusco-Italicterracottafriezeswith vegetal ornamentsamong which human heads also
occasionallyappear.47
The long strong tradition of vegetal themes on the Greek mainland did not allow the
intrusion of South Italian or Italic vegetal systems in its ornamental repertory48and con-
tinued to inspire the peripheralworkshops. Even in Macedonia where Italic influence is
generally detectable, imitations or even imports of late 4th/early 3rd-centuryB.C. archi-
tectural terracottasfrom Athenian and Peloponnesian (especially Corinthian) workshops
46 Klumbach 1937, pp. 29-30; Andren, 1940, pp. 53-55; Toynbee and Ward Perkins1950, pp. 1-4; Kraus
1953, pp. 31-32; Jucker 1961, p. 164; Lullies 1962, pp. 12-13; Zervoudaki1968, pp. 1-88; Trumpf-Lyritzaki
1969, introduction;Borker 1973, pp. 287-288; Bauer 1973, pp. 9-13; Robertson 1975, pp. 440, 486-487, who
suggestsan origin of the vegetal motifs from the painting of the Sicyonian painter Pausiaswho, according to
Pliny,NatHist35.123, "ad numerossimamforum varietatemperduxitartem 111am." Themelis 1979, pp. 136-
137; Hamdorf 1979, pp. 31-41; Pfrommer 1982, pp. 119-130; Lauter 1986, p. 250; Lauter-Buf61987, p. 65;
Sauron 1988, pp. 3-4, 10-19; Schauenburg 1989, pp. 36-37; Kanel 1991, pp. 172-174.
47 Kanel 1991, pp. 173-174, note 28, 175.
48 Pfrommer 1982, p. 167.
158 PETROS THEMELIS
W~
~~~~~FG 15 SiaoGru
V
retd to
'4~~~~~~~~~~
atfi Typ 3t
FIG. 15. Sima of Group B related to antefixType 3
la ~ ~ ~ ~ ~ ~ ~ ~ o
IT~~~Y
are well attested.49 The diffusion of vegetal motifs during the Hellenistic period and the
development of many regional styles makes detection of the possible sources of inspiration
complicatedand often fruitless.The lack of artistictraditionin Messene before 369 B.C. and
its geographical situationwould justify an Italic influence, as is probably the case with the
Laconian antefixType 2 describedabove. Certainfeaturesof the architecturalterracottasin
Group B, Type 7, as, for instance,the large size and form of flowersor flowerbuds hanging
from long, thin, flexible stems on the rakingsima, could be compared with Italic examples,
but not exclusively because most of the motifs seem to derive from a Hellenistic vegetal
"koine".50 Consequently,a formal analysis of each motif separated from its ornamental
context would be meaningless and probably misleading. The general design and effect,
the composition of motifs on the antefix, and especiallythe raking sima displayingthe two
flyingvegetal Nikai that flankthe centralfloralsystemare unparalleledamong architectural
terracottas,earlieror contemporary,as far as I know. Thus it is reasonableto accept that we
are dealingwith the productof a local Hellenistic(2nd centuryB.C.) workshopcloselyrelated
to workshopsin western Greece comprisingElis, Arkadia,Achaia, and Illyria.
Semihuman, semivegetal figures ("plant divinities")winged or not, male or female,
appear on various works of art from the second half of the 4th century B.C. on.51 Most
of them, however, are represented in rigid "archaistic"frontal postures not having any
apparent morphological, iconographic, or stylistic affinity to these flying vegetal Nikai.
I mention the gold diadem of the 3rd century B.C. with vegetal decoration from Kyme
4' Kaltsas 1988, pp. 69-71. Eclecticismseems to characterizenot only Macedonian architecture(Grobel-
Miller 1971, pp. 228-229) but also mosaic floralpatternsand architecturalterracottas.
50 Similarlong, thin and flexiblestemson a sofa capitalin Rhodes: Konstantinopoulos1969, p. 483, pl.481 :c
and on western Greektombstones:Fraserand Ronne 1957, pp. 52, 55-59; Fraserand Ronne 1971, pp. 53-54;
Papapostolou 1975, pp. 291-292, fig. 1; cf. Dyggve, Poulsen,and Rhomaios 1934, p. 350, fig. 55.
51 (1) Mosaic of the Vergina Palace: Petsas 1965, pp. 41-42, fig. 7; Andronikos,Makronas,Moutsopoulos,
and Bakalakis1961, pp. 21-22, pis. XVI, XVII; Salzmann 1982, p. 114, no. 130.
(2) The so-called Dionysos-Psilax winged figure on a marble throne found near the pronaos of the
Parthenon (AkropolisMuseum 1366) and a similarone in the Berlin Museum: Mcbius 1926, pp. 117-124,
pls. 18-20; Richter 1954, no.4, pp. 271-276, pls. 47-50; Picard 1935, pp. 317-337; Picard 1963, pp. 961-962,
969, figs. 385, 386.
(3) A marble acroterionfrom South Russia in Leningrad:Mobius 1926, pl. 19; Picard 1963, fig. 386.
(4)A kraterfrom Canicattiniin the SyracuseMuseum: Libertini1950, pp. 97-98, figs. 2, 5; Schauenburg
1957,pp. 198-221.
(5) Stone helmet model from Mid-Rahineh(Egypt)in Amsterdam(AllardPiersonMuseum 7866): Ponger
1942, p. 86, no. 177, fig. 4, pl. 39; Mobius 1964, p. 32; Neumann 1965, p. 149, note 28, pl. 59:1.
(6) Gold diadem from Irakleion,Crete (Coll. Metaxas no. 442): Alexiou 1963, p. 310, pl. 359; Neumann
1965, pp. 144-15-1,pls. 56:1, 57:3.
(7) Wall painting of Tomb No. 69 at Myra in Lycia: Fedak 1990, pp. 98-99, fig. 123:a,b.
(8) Sundial in Rhodes: Konstantinopoulos1969, p. 466, pl. 471.
(9) Capital from Salamis, Cyprus: Smith 1900, pp. 263-266, no. 1510, pl. 27.
(10) Capital fragmentin Cumae: Gabrici 1913, pp. 723-724, fig. 257.
(11) Pilastercapital in Kos: Neumann 1965, p. 149, note 27, pl. 58:1-3.
(12)Erosor Nike on two reliefpillarsand on moldmadebowlsof the early2nd centuryB.C. from Pergamon:
de Luca 1990, pp. 595-597, pl. 93:2.
(13) Winged vegetal figure (Hekate, Enodoia?) on a pillar capital from Didyma: Voigtlander 1990,
pp. 415-416, pl. 63:2.
160 PETROS THEMELIS
Gld
Cn
Cld
162 PETROS THEMELIS
no~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~c
56
Cawadias 1893, p. 11, pl. 4; Dickins and Kourouniotes 1906-1907, pl. 12; Dickins 1911, p. 311, fig. 2;
Karouzou 1954, p. 170, nos. 2171-2175, pl. 58; Karouzou 1974, p. 33; Lattimore 1976, p. 56.
57 Orlandos 1976, pp. 26, 38, figs. 13 and 37; Lauter 1986, p. 268, pl. 36:a, b.
58 Goethert 1936, pp. 76-78; Bertoldi 1962, pp. 18-20; Borbein 1968, pp. 97-98, 102, note 506; Zanker
1970, p. 513, figs. 22, 23.
HELLENISTIC ARCHITECTURAL TERRACOTTAS FROM MESSENE 163
so i~~~/
PAW~~~~~~W
: - -: .
S -i
dT
~;;;k -2m
0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.1
Parallelsfor the vegetal decorationof these simas are to be found on the sofa capitalsof
the Hellenisticgrave stelai from western Greece59attestedthus far in Elis, Achaia, Arcadia,
and Kefallonia, in an area characterizedby a certain cultural unity to which Messenia
also belongs. The vegetal frieze on a sofa capital from Dyme in Achaia (PatrasMuseum,
inv. no. 273 1)dated in the 2nd centuryB.C.60 providesthe closestparallelto our rakingsima
despitethe absence of the vegetal figures:two double cauliculicrownedwith acanthusleaves
spring antitheticallyfrom a central tripartiteacanthuswhile thin flexible scrolled-upstems
ending in large flowers emerge from the cauliculi.
The scene on the raking sima (Fig. 19:a)cannot be merely decorative; it should have
a symbolic meaning as well; I would dare to call it "The Glorification of the Floral
Symbol" and relate it to a divinity of the vegetal world. The temple of Aphrodite seen
by Pausanias (4.31.6) in the agora (?) of Messene near the sanctuary of Poseidon would
be an appropriatebuilding to bear these architecturalterracottason its roof. A terracotta
figurine of a winged Nike and two terracottaheads of Aphrodite with a flower stephane
and flowersaround the hair61are the only evidence we have concerningthe iconographyof
this goddessat Messene (P1.54:c). The smallerhead wears a thorny crown that shows formal
similaritiesto the thorny ring of the central flower on the antefix and the raking sima.62
The figuralvases of Attic and Olynthianworkshopsfound in Athens, Megara, the Corinthia,
Chalkidike,Boiotia, CimmerianBosporus,Euboia, Phokis,Lokris,and Akarnaniashow an
overwhelmingpreference for scenes of the circle of Aphrodite;63the goddess, Erotes, and
winged female figuresor bustsare representedamong acanthusspirals,leaves, and flowersor
emergingfrom acanthus calyxes or seashells,thus testifyingto the connection of Aphrodite,
Eros, and winged women with acanthus motifs and flowers. The Apulian and Lucanian
vases of the second half of the 4th centuryB.C. are also conclusivein this respect.64
Type 8 (Group C) of architecturalterracottasis rather conservative and classicistic
compared to the previous one. A special feature-of the sima decoration is the very long
two-storied cauliculus as well as the bell flowers (Fig. 21; PI. 55:a), while on the antefix
are the fleshy, sinewed palmette leaves, the second small palmette around the heart, the
broad horizontal acanthus leaves along the bottom line, the modest spiralsemerging from
the acanthus, and the S-shaped thin stems ending in dotted flowers on both sides of the
central palmette leaf (Fig. 20; PI. 55:b). Similar n-shaped openings across the upper
59 Papapostolou1975, p. 292, fig. 1. Cf. the acanthus-spiralfriezewith a winged Eros on a marble throne in
the Castle of Kos: Neumann 1965, p. 151, pl. 59:2. Cf. McPhee 1979, pp. 159-162 (on a group of squat
lekythoi in Agrinion and Ioannina decorated with figures surroundedby vegetal motifs also found on the
Adriaticcoast, attributedto local workshopsof northwesternGreece bearingadditionalwitnessto the "cultural
unity"of the region).
60 Papapostolou 1975, p. 292, fig. 1.
61
Messene Museum, Mv.nos. 952, 955, 956, unpublishedchance finds.
62 Foran earlierparallelto the thornyspiralson the rakingsima, see the gold capitalpin from the Peloponnese
in the Museum of Fine Arts, Boston (inv.no. 96.718): Depper-Lippitz1985, pp. 153-154, fig. 106.
63 Trumpf-Lyritzaki1969, pp. 126-131. In South Italy there are imitationsof these vases: Trumpf-Lyritzaki
1969, p. 112: Plato, Symposium 196: o5 8' & avO~ 5 cet xal 6rj~ t6noq 5,Ivcauoa b xce tit xalcivet.
64 Trendall 1967, p. 159; Robertson 1975, p. 440; Sichtermann 1966, 54K-79, pl. 138. Cf. also the Nikai
and Erotes on the gold wreath from Crumentum in Munich (inv. no. 2334): Hamdorf 1979, pp. 31-33,
figs. 3, 6.
HELLENISTIC ARCHITECTURAL TERRACOTTAS FROM MESSENE 165
/~
~~~~~~~~~~~~~~~~~ &~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~i
I6~~~~~~~~
Go_ ~~~~~~~~~I 0
l
0.16 0.o6
edge of the acanthus occur on the antefixes of the Stoa of Attalos (159-138 B.c.) as well
as on the Corinthian capitals of the Olympieion in Athens from the period of Antiochos
IV Epiphanes(176-165 B.C.).65The stone antefixesand simasfrom the stoas and the temple
of the Asklepieion,dated to the 2nd centuryB.C.66are similarto these Messenian terracotta
antefixesand simas (P1.55:c). Two more simaswith no correspondingantefixes(Figs.22, 23;
P1.56:a, b) ofthe same GroupC presentslightdifferentiationsas far as form and arrangement
of flowersare concerned.
D. LATESTYLE(TYPES
9 ANDIO)
Two antefixeswithout correspondingsimas of rather poor workmanship(Figs. 24, 25;
P1.56:c, d), Type 9 a chance find from the Stadium area, Type 10 of unknownprovenance,
are at this time the only representativesof GroupD. Their motifsare relatedto earlierpieces.
They may derive from buildings, especiallystoas, which had been repaired during the 1st
century after Christ.67
65
Mobius 1926, p. 121, pl. 19:a, b. Cf. also the acanthi on the finial of the stele from Samikon dated to
the 2nd centuryB.C.: Yalouris1965, p.210, pI. 234:b;Fraserand Ronne 1971, pp. 75-76, fig. 38; Papapostolou
1975, p. 294, no. 5.
66
Orlandos 1969, p. 100, fig. 12, pI. 132:a,b.
67
See notes 17 and 21 above.
166 PETROSTHEMELIS
BIBLIOGRAPHY
Alexiou, S. 1963. v.ApXaL6vtlj-cxaL Vvipcca Kp4-cyrq,Ack? 18, 1963, B'2 [1965], pp. 309-316
Andrea, Z. 1976. "GermimetArkeologjike1974/5," Iliria4, 1971, pp. 347-349
Andren, A. 1940. Architectural from Etrusco-Ita&Templks(Acti Insititi RomaniRegni SueciaeVi),
Terracottas
Lund/Leipzig
Andreou, E. 1983. <<Topluxp6Oed&tpotiq A4P3paxcac>>, HtcLpcati~x&Xpovix&25, pp. 9-23
Andronikos,M., C. Makaronas,N. Moutsopoulos, and G. Bakalakes. 1961. To acvxtopo tvq Bepylvaq,
Athens
Bardani,V. 1986. <<ELt IG V', 1462?, HOROS6, pp. 79-81
Bauer, H. 1973. Korinithische Kapiteledes 4. und3. Jhs. v. Chr. (MitteiwgendesDeutuchen Instituts,
Archiologischen
Athenitche Abteilung.
Beileft3), Berlin
Bertoldi,M. 1962. Rieerche sula decorazione nelForoTraiano,
architettonica Rome
Billot, M.-F. 1976. "Terrescuites architecturalesdu Mus&etpigraphique," AXct 31, 1976, A' [1980], pp. 87-
135
Blouet, A. 1831. ExpIitionscientifiquedeMorkeI, Paris
Borker,C. 1973. "Neuattischesund Pergaminischesan den Ara Pacis-Ranken,"JdI88, pp. 287-317
Bol, P., and F. Eckstein. 1975. "Das PolybiosStele in Kleitor,"AntP15, pp. 83-93
Borbein, A. H. 1968. Campanareliefs. undstiliische Untersuchungen
Typologische desDeutschen
(Miteilungen Archaolo-
gischenInstituts,RmischeAbteilung. Erginzungsheft14),Mainz am Rhein
Brandes-Druba,B. 1994. "EinigetarentinerArchitecturterrakotten," in HesperiaSupplement 27, pp. 309-325
Budina, D. 1994. 'ArchitecturalTerracottasfrom the Towns of Kaonia: Antigonea and Buthrot,"in Hesperia
Supplement 27, pp. 217-219
Buschor,E. 1958. MedusaRondanini, Stuttgart
1961-1962. "Die Grabung im Heraion vom Samos. Frahjahr 1961," AcXt 17, 1961-1962, B'l
[1963], pp. 279-280
Callaghan,P.J. 1983. "StylisticProgressionin HellenisticCrete,"BICS 30, pp. 31-39
Camp, J. M. 1986. TheAthenian Agora,London
Cavvadias,P. 1893. FouillesdeLycosoura, Athens
Dakaris,S. 1952. <<AvaaxacqKaaadans>>,IIpaxrtLx 1952 [1955], pp. 326-362
1954. <<AvaaxacpKaaa'6in>>, lIpaxixlx&1954 [1957], pp. 201-205
. 1963. "Das Taubenorakelvon Dodona und das Totenorakelbei Ephyra,"AntK-BH1, pp. 35-55
- 1989. Kaaa6xn, NedaprLpac avaaxapca 1977-1983, Ioannina
De Luca, G. 1990. "HellenisticheKunst in Pergamon im Spiegel der megarischenBecher,"in AktendesXIII
internationalen klassische
Kongressesflir Berlin1988, Mainz am Rhein, pp. 595-597
Archaologie,
Depper-Lippitz,D. 1985. Griechisches Goldschmuck,Mainz am Rhein
Deshayes, J. 1966. Argos.Lesfouilesdela Deiras(tdespeloponnesiennes M), Paris
Despinis, G. 1966. "Ein neues Werkdes Damophon,"AA [JdI 81], pp. 378-385
Dickins, G. 1911. "The Sandale in the Palazzo dei Conservatori,"YHS 31, pp. 308-314
Dickins, G., and K. Kourouniotes. 1906-1907. "Damophon of Messene II," BSA 13, pp. 357-404
Dyggve, E., F. Poulsen, and K. Rhomaios. 1934. Das HeroonvonKalydon, Copenhagen
Ehrhardt,H. 1985. Samothrake, Stuttgart
A
Evangelides,D., and S. Dakaris. 1959. <<Toltp6 t-q ctdvrjq, ApXEp, pp. 1-176
Fedak,J. 1990. Monumental Tombs oftheHellnisticAge:A StudyofSelected
Tombsfrom totheEarlyImperial
Pre-Classical
Era,Toronto
Fraser,P., and T. Ronne. 1957. Boeotian andWestGreekTombstones, Lund
. 1971. "Some More Boeotian and West GreekTombstones,"OpAth10, pp. 53-89
Gabrici,E. 1913. "Cuma,"MonAnt22, pp. 1-871
Gneisz, D. 1994. "Die Dachterrakottenvon Aigeira,"in HesperiaSupplement27, pp. 125-134
Goethert, F.-W.1936. "TrajanischeFriese,"JdI 51, pp. 72-81
Groag, E. 1939. Die rimischen RJeihsbeamtenvonAchaiabisaufDiokletian,Vienna
HELLENISTIC
ARCHITECTURAL
TERRACOTTASFROMMESSENE 167
Grobel-Miller, S. 1971. "Hellenistic Macedonian Architecture: Its Style and Painted Ornamentation"
(diss.Bryn Mawr College 1971)
Habicht, C. 1985. Pausaniasundseine"Beschreibung Griechenlands"Munich
Hanlein-Schafer,H. 1985. Veneratio Augusti.EineStudiezu denTempein deserstenrdmischen Kaisers,Rome
Hamdorf, F.-W.1979. "Ein Goldkranzdes 4.Jhs. v. Chr.,"Muijb30, pp. 31-41
HesperiaSupplement 27 = Proceedings of theIntrnationalConference
on Greek Architectural
Terracottas of theClassical
andHellnisticPeriods,Athents,December 12-15, 1991 (Hesperia Supplement 27), N. A. Winter, ed., Princeton
1994
Hoepfner,W. 1990. "Bautenund Bedeutungdes Hermogenes,"in Hermogenes unddiehochhellenistischeArchitektur,
InternationalKoUoquiumin Berlin1988, W Hoepfner and E.-L. Schwandner,eds., Mainz am Rhein, pp. 1-34
Hubner, G. 1976. 'AntefixaDeorum Athenarum,"AM 91, pp. 175-183
IG VI = InscriPtionesLaconiae
etMesseniae,W. Kolbe, ed., Berlin 1913
Jucker, H. 1961. Das Bildnisin Blitterkelch: undBedeutugeinerromischen
Geschichte Portradform
(Biblotheca Helvetica
Romana3), Lausanne/Freiburg
Kanel, R. 1991. "Zwei etruskisch-italischeTerrakottaplattenmit vegetabilem Dekor in Genf," AntK 34,
pp. 170-177
Kalligas, P. 1969. <<ApXat6-tYteqxat Vvneta IcavicovN~acjvo, AXtE 24, 1969, B'2 [1970], pp. 258-279
.1979. <<ApXatoXoyTxL eupijVxa ac6 tvnvIOMxno, KeaXXvLdtax& Xpov.x& 3, 1978-1979, pp.45-69
. 1980. <To tep6 tou An6XXjvoq TneptreXe4taatv Aaxjvca>>,in Hpaxcztx&tou A' Aax&vtxo6
uvebp0ou, En prTreO?Lo 7-11 Oxt., 1977, Athens, pp. 10-30
Kaltsas,N. 1988. HIXtveq&caxoa native xpaV cctGa6tv6 -Maxebovvca,Athens
Karouzou, S. 1954. EuXXoypy)unrdtv, Athens
.1967. uUXoy yXutrn6v,Athens
. 1974. <H VtXp~ Ccj(6poq t&)v 8eplionuwXv xat n ivvota tou OacXiatou OtLaouo, ApXEP,
pp. 26-44
Kenfield,J. R 1994. "High Classicaland High Baroque in the ArchitecturalTerracottasof Morgantina,"in
HesperiaSupplement27, pp. 275-281
Klumbach, H. 1937. Tarentiner Grabkunst (7tibinger
Forschungen undKunstgeschichte
zurArchaologie 13), Reutlingen
Konstantinopoulos, G. 1969. vApXat6tYteq xat jivnta AcabexavVaouo, AeXt 24, 1969, B'2 [1970],
pp. 457-485
Kourouniotes,K. 1911. Ka-czXoyoqtou MouaE(ou Auxoao~paq, Athens
Kraus, T. 1953. Die RankenderAraPacis:EinBeitragzumEnhiicklungsgeschichte deraugustdischenOrnamentik, Berlin
Kyrieleis, H. 1978. "Ausgrabungen im Heraion von Samos 1978,"AA [JdI 93], pp. 250-259
La Rosa, V 1990. "Ceramicheellenisticheda Festos: per il problema della distruzionefinale della cittA,"in
B' EntatTrnjovxL Euvivtnan yta tnv EXXnvtatLXrm Kepajtxie, Athens, pp. 160-166
Lattimore,S. 1976. TheMarineThiasosin Greek Sculpture
(MonumentaArchaeologica3; Monograph9, Archaeological
Instituteof America),Los Angeles
Lauter,H. 1986. DieArchitktur desHellenimmus, Darmstadt
Lauter-Buf6,H. 1987. Die Geschichte dessiketiosch-korinthischen Dersogennante
Kapitells. italisch-republikanische
7ypus,
Mainz am Rhein
Le Bas, P. 1844. "FouillesAMessene,"RA 1, pp. 422-426.
Levy,E., and J. Marcade. 1972. "AuMusee de Lycosoura,"BCH 96, pp. 968-1004
Libertini,G. 1950. "Grandecratereda Canicattinidel Museo di Siracusa,"BdA35, pp. 97-107
Lullies,R. 1962. Vergoldete ausTarent,
Terrakotta-appliken Heidelberg
Makaronas,C., and E. Your. 1989. Ot oLx(Ee apnayfjqvrq EX&vn xat Atov~aou tvnqH&cag, Athens
Marshall,E H. 1911. Catalogue ofjewdlleryin theBritishMuseum,London
McPhee, I. 1979. "The Agrinion Group,"BSA74, pp. 159-162
Migeotte, L. 1985. "Reparationdes monumentspublics a Messene,"BCH 190, pp. 597-607
Miller,Stella G. 1994. "ArchitecturalTerracottasfrom Ilion," in Hesperia Supplement27, pp. 269-273
M6bius, H. 1926. "Eine dreiseitigeBasis in Athen,"AM 51, pp. 117-124
. 1964. Alexandria
undRom,Munich
168 PETROSTHEMELIS
*1990.<<Mewctvr>>,
'Epyov 1990[1991],pp. 26-35
*1991.<<Mecacav>>,
'Epyov 1991[1992],pp. 28-35
*1993. "Damophonvon Messene. Sein Werkim Lichteder neuen Ausgrabungen,"AntK36, pp. 24-40
Toynbee, J. M. C., and J. B. Ward Perkins. 1950. "Peopled Scrolls: A Hellenistic Motif in Imperial Art,"
BSR 18, pp. 1-43
Trendall,A. D. 1967. TheRed-fgured VasesofLucania,Campania andSicily,Oxford
Trumpf-Lyritzaki,M. 1969. GriechiischeFvgrenvasen desreichen
StilsundderspitenI(assik,Bonn
Uhlenbrock,J. P. 1988. TheTerracotta ProtoWaefrom Gela,Rome
Valmin, M. N. 1938. The SwedishMesseniaExpedition (ShifterUgima av Kungl.Humanistiska Vetenskapssamfrndet
26), Lund
Vlachopoulou-Oikonomou,A. 1986. HyreVcer xaL xopuyaptOLxpa4ot Ve btaxoavan 1o7 6 tv 'HneLpo,
Ioannina (unpublisheddissertation)
Voigdtander,W. 1990. "VorklassischeBaugedankenin spat- und nachklassicherArchitekturKleinasiens,"Akten
desXIII internationalen
Kongressesfiirklassische Berlin1988, Mainz am Rhein, pp. 414-416
Archdologie,
Waldstein,C. 1902. TheArgiveHeraeum I, Boston
Winter,F. 1903. Die 7ypendefigiirliehen
TerrakotnH, Berlin
Yalouris,N. 1965. vApXat6-re xaL VvtVea HMtac>>,AekX 20, 1965, B'2 [1967], pp. 209-213
-1968. <<ToaxpCiY5ptoov-ou Hpa(ou OMuitna>>,in XapLapt5pLovA. OrlandosI, Athens, pp. 57-65
Zanker,P. 1970. "Trajansforumin Rom," AA [JdT85], pp. 499-544
Zervoudaki,I. 1968. "AttischePolychromeReliefkeramikdes spaten 5. und 4.Jhs. v. Chr.,"AM 83, pp. 1-88
Zimmermann,K. 1994. "Traufziegelmit Reliefmaanderaus dem Schwarzmeergebeit,"in Hesperia Supplement
27, pp. 221-251
PETRosTHEMELIS
UNIVERsrrYOF CRETE
Department of History and Archaeology
74100 Rethymnon
Greece
PLATE 48
,~~~~ ~ ~
t4vp'r. ~~~~~~~~~~~~~~~~~~~~~~
(2,t-~ N 4
ZIN ~ ~ ~ ~ -
k ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
c. Fragmentof Laconian
antefix. Variantof Type 2 I
(inv.no. 1716) -i
'~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-7
f. of Archaiclaconian
acroteria(inv.nos. 1708 [top],
Fragcaents 1707 [bottom])
e. Laconian antefix Type 4 with Medusa head (inv. no. 96)
HELLENISTIC
PEmRos THEMELIS: TERRACOTTAS FROMMESSENE
ARCHITECTURAL
PILATE50
C'4~~~~~~~~~~~~~~~~
0 X~~~~~~~~~~
0O
~cn
(U
4 I t..
4-J ~ ~~ 4
'' '~~~~~~4
co~
co~~~~~~~~~~~
6~~~~~~~~~~~
0o
PLATE 51
A I
co0
4$ W[ I I te i
, ...0~v
Y. - 4
N LO
oi
C-
PLATE 52
=i~~~~~~~~~iilli~~~~~~~~~;.;T
[~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4'Id
'AD
a. Two fragmentsof sima relatedto antefixType 4 (inv.nos. 1207, 1832)
t;<> S';' . 4;
At__
A.,
..Jj.. t,. ~ ~ ~ ~ ~ ~ ~ .
Be _ ~~~~~~~~~~~b.
CorinthianantefixType 8 (inv.no. 1594)
9Williams 1988, S. 227; Heiden 1987, S. 17; zur Terminologies. Winter 1990, S. 13 mit Anm. 1.
10 Felsch 1987, S. 14-15.
Hubner 1990.
2 Felsch 1987, S. 17, 25; vgl. S. 11 und die Abb. 2-3; 12; 26-27.
13 Felsch und Schuler 1980, S. 98.
14 Felschund Schuler 1980, S. 88; Felsch 1988, S. 63.
5 Felsch 1991, S. 88 Anm. 30.
16 Habner 1990, S. 170.
17
Sie beginnt mit den rf. Simen-Beispielen, die keine Fruchte aufweisen, Buschor 1929, S. 20-46,
Taf. 1-4, 7-10; Vlassopoulou 1990, S. xxiii, no. 41-43; vgl. auch die Simen 56 und 65, Taf. B. Hinzu
kommen Traufziegelmit diesem Ornament, Buschor 1933, S. 17-18; Vlassopoulou 1990, S. xxviii, no. 58.
Vorlaufig singular ist das Beispiel mit auf das Blatt gesetzter Frucht aus dem Kerameikos, Hubner 1973,
S. 73-77, Abb. 2, Taf. 57:2.
DIE KLASSISCHEN TEMPELDACHER VON KALAPODI (PHOKIS) 173
Muse vom Helikon umkranzt,18so wird man eine solche Koinzidenz auch als ikonographis-
ches Indiz werten duirfen.19
In dem hier interessierendenZusammenhangregen die gerade aufgezahlten Grunde
dazu an, diesen Traufziegeltypusmit dem provisorischenKultbau zu verbinden. Getragen
wird die Annahme nicht zuletzt durch den Umstand, daB die stilistischeDatierung des
Ornamentes,wie der Vergleichmit der erwahntenSchale zeigt, eine Zeitstufeimpliziert,die
mit dem Munzfund aus dem FuBbodendes durch die AdytonmauerverkleinertenKultrau-
mes korrespondiert.20 Tatsachlich ergabe das oben postulierte BreitenmaBvon 0.58 m
pro Ziegel fuinfzehnfachvervielfachtdie Gesamthingedes provisorischenKultraumes,wobei
ein Spielraum von je 0.012 m zwischen den StoBfugen der jeweiligen Ziegellangsseiten
hinzugerechnetwerden muB. Die Schlussigkeitder Beweisfuhrungwird jedoch dadurch
abgeschwaicht,daB entsprechendeAntefixformulierungen,die, jedenfalls attischerTypen-
traditionfolgend,21fur diese Art von Traufziegelnzu fordernwaren, nicht vorhanden sind,
ja daB Antefixe uiberhauptzu fehlen scheinen. Andrerseitssoilten sie aber zu erwarten
sein sogar dann, wenn lediglich ein hypaithralesUmfassungsgeviertmit einer tOnernen
Ziegelleiste abgeschlossengewesen war.22 Dieser Aporie ist nur durch die Vermutung zu
begegnen, daB entweder die noch intaktenarchaischenAntefixe als kronendesSchluBglied
verwendet oder daB die schon fertiggesteiltenAntefixe des ersten klassischenTempels23
bereits vorlaufig hier versetzt waren. Wenn hier beide Losungen erwogen werden, so
geschieht dies deshalb, um die Aufmerksamkeitallgemein auf die Frage zu lenken, ob in
dem vorgegebenen Rahmen, wie dies ein solcher Kultort darsteilt, sichtbare Wiederver-
wendung resp. Kombination von dekoriertenDachterrakottena priori moglich ist. Eine
derartigeMaBgabe setzt selbstverstandlicheine inharentemetrologischeAbhangigkeitvon
in situ aufeinanderfolgendenBauten voraus.24 Fur die archaische Zeit hatte zumindest
in Kalapodi gegolten, daB einem einzigen Kultgeschehenzwei Tempelgebaudezugewandt
waren, deren Dacher in sich wechselweise,je nach GroBe des Baus, verschiebbareresp.
austauschbateElemente besessen hatten.25 Eine cum grano salis vergleichbareHypothese
fur die direkte Nachfolgearchitekturin Betracht zu ziehen, mag daher vor diesem Forum
legitim sein.
Der Unsicherheitsfaktorder Uberlegungen wird freilich durch den fragmentarischen
Zustand der Lorbeerzweig-Ziegel, deren Tiefenausdehnung in jedem Fall verloren ist,
unterstrichen. Vollig spekulativbleibt so die Frage nach einem (theoretischmoglichen)
Giebel. Jedenfalls existieren keine Simenfragmente, deren AuBeres in Konsistenz und
aber die Moglichkeit dazu voraus, man vgl. etwa auch die Interdependenzenzwischen Vorparthenon und
Parthenon,s. demnachstKorres (in Vorbereitung).
25 Vgl. z.B. die Antefixe der archaischenTempel, Hubner 1990, S. 170-171.
174 GERHILD HUBNER
28 Le Roy 1967, Dach 56 (S. 128-132), S. 128-129, Taf. 46; Heiden 1987, S. 175, Abb. 6:4, Taf. 12:1.
Dachglieder dieses Typus aus Kalapodi sind abgebildetin: Felsch und Kienast 1975, S. 9, Abb. 12; Hubner
1990, Taf. 15:a.
29 Muller-Wiener 1988, S. 104, Abb. 55. Die in Kalapodi zahlreich zutage gekommenen Eisennagel
gehoren wahrscheinlich zum hcilzernen Dachstuhl. Ihre Bearbeitung durch H.-O. Schmitt, Heidelberg,
ist abgeschlossen. Gerade in der zweiten Halfte des 5. Jhs. v. Chr. tritt Nagelung bei Terrakottaziegeln
verhaltnismassighaufig auf, vgl. z.B. auch die attischenDacher, Hubner 1973, S. 89.
30 Le Roy 1967, S. 131-132.
31 Kienast 1980, S. 102; Felsch 1988, S. 63.
32 Kienast 1988, S. 102.
176 GERHILDH(BNER
a b
JC~~~~~~ _
l t Rl ! ~it C
'4!1 A
\C/ ANJ l
X. '0 I' 1l
9LIZ./.
H
; -
- A~~~~h 'a~
kW))gX~~
AS
(~AI~~ I
Ne s
- Wa(- e X ,~~-
178 GERHILD HUBNER
BIBLIOGRAPHIE
34 Vgl. die Antefixe von Samos, Ohnesorg 1990, S. 188-189, Taf. 20:d ("Rhoikos"Antefix); S. 189 mit
Anm. 12, Taf. 21:a (SUdhalle),die archaischenAntefixe der Akropolis,vgl. Vlassopoulou 1990, S. xvi, no. 19,
xx, no. 32, Taf. A. Zum Parthenonantefixzuletzt, Korresund Bouras 1983, S. 41.
35 Die Voluten der Akropolisantefixe,s. Anm. 34, rollen sich bis zu einem gewissen Grad vergleichbar
nach oben auf, bilden allerdings mit einem gegenlaufigen Pendant eine geschlossene Schlaufe. DaB, in
Analogie zu den Samosbeispielenund den Parthenonantefixen,die Voluten des Kalapodi-Antefixessich aus
der horizontalenStandleisteherausnach oben entwickelthatten, dann also um 900 gedrehtvorgestelltwerden
mussten,verbieten die Breite und Laufrichtungder erhaltenenVolutenfragmente.
36 Zur sog. Fruhen Rankensima,s. Heiden, Kongress-Beitragdieses Bandes;zur Telesterionsima,s. Schede
Felsch, R. C. S. 1987. "Berichttiber die Grabungenim Heiligtum der Artemis Elaphebolosund des Apollon
von Hyampolis 1978-1982," AA JdI 102], S. 1-99
. 1988. "Das Heiligtum bei Kalapodi in vor- und fruklassischerZeit," in IIpaxxtxd tou XII AteOvous
Euve~ptou KX=Lx#s ApXatoXOoy(aq, AO5va, 4-10 Eewce4pptou 1983, IV, Athen, S. 61-64
. 1991. "Tempel und Altare im Heiligtum der Artemis Elaphebolos von Hyampolis bei Kalapodi,"
del'antiquit,Lyon, S. 85-91
in L'ispacesacsficieldansks cilisaions mediterraneennes
Felsch,R. C. S., und H.J. Kienast. 1975. "Ein Heiligtum in Phokis,"AAA8, S. 1-24
Felsch, R. C. S., und H. Schuler. 1980. "Apollonund Artemis oder Artemis und Apollon? Bericht von den
Grabungenim neu entdecktenHeiligtum bei Kalapodi 1973-1977," AA UdI 95], S. 38-112
Felsch, R. C. S., und P. Siewert. 1987. "Inschriftenaus dem Heiligtum von Hyampolis bei Kalapodi," AA
UdI 102], S. 681-687
Heiden,J. 1987. Korinthische Dackziegel.ZurEntwicklung derkorinthischen
D&Ier,Frankfurtam Main/Bern/New
York/Paris
Herrmann, K. 1974. "Die Giebelrekonstruktiondes Schatzhausesvon Megara,"AM 89, S. 75-83
Hesperia 59 = Proceedings oftheFirstInternational onArchaic
Conference GreekArchitecturalTerracottas,
Atns, December2-4,
1988 (Hesperia 59, 1990), N. A. Winter,ed.
Hubner, G. 1973. "Dachterrakottenaus dem Kerameikosvon Athen,"AM 88, S. 67-143
. 1980. "Die Dachterrakotten,"AA [IdI 95], S. 112-115
- 1990. "Die Dachterrakotten der archaischen Tempel von Kalapodi (Phokis)," in Hesperia59,
S. 167-174
Kienast, H.J. 1980. "Die klassischenTempel,"AA [dI 95], S. 99-108
. 1988. "Die klassischen Tempel im Heiligtum bei Kalapodi," in Hpaxtx& tou XII AteOvobq
Euve8ptou KXaatx#J ApXatoXoyUaq, AO5va, 4-10 EeJxtepppou 1983, IV, Athen, S. 100-103
Knigge, U. 1988. DerKerameikos vonAtzen,Athen
Korres, M. (in Vorbereitung).Eu3poX~ atrjv otxo8O.LXn VLeXirY tCav ap~cCav xtov(~v (Arbeitstitel)
Korres, M., und Ch. Bouras. 1983. MeX&iqanoxa-cata-ccA tou HIapOcsvoq, Athen, S. 37-40
Lauffer, S. 1989. Griechenlad,Lexikonderhistorischen StattenvondenAnfdngen S. Lauffer,ed.,
bis zur Gegenwart,
Munchen
Le Roy, Ch. 1967. Lesterrescuitesarchitecturales deDelphesII), Paris
(Fouilles
Leekley,D., und N. Efstratiou.1980. Archaeological Excavations
in CentralandNorthern ParkRidge, NJ
Greece,
Muller-Wiener,W. 1988. Griechisches Bauwesenin derAntike,
Monchen
Ohnesorg, A. 1990. "ArchaicRoof Tiles from the Heraion on Samos,"in Hesperia59, S. 181-192
Philippart,H. 1936. Lescoupesattiques fond blanc,Bruxelles
Schede, M. 1909. AntikesTraufleistenornament, Strassburg
Vlassopoulou, C. 1990. "DecoratedArchitecturalTerracottasfrom the Athenian Acropolis,"in Hesperia59,
S. Vii-xxxi
Williams, C. K., II. 1988. "CorinthianTrade in Roof Tiles," in llpaxwxx CouXII ALEOVo(6EuvEBpEou
KkaXaLX4ApXatoXoytaq,AOva, 4-10 Een-celpptou, 1983, IV,Athen, S. 227-230
Winter, N. A. 1990. "Defining Regional Styles in Archaic Greek ArchitecturalTerracottas,"in Hesperia59,
S. 13-32
Wrede, W. 1933. Attische Mauern,Athen
GERHILDHBNER
INsTrrrr FiJR BAUGESCHICHTE DER UNIVERSITXT
Englerstr.7
D-76128 Karlsruhe
BundesrepublikDeutschland
PLATE 57
a)~~~~~~O W
X"i a T4?~
;X_
' ;F~~~~~~~~~~~~~~~~~~~~~~~~~~~-
i adr
* 4, .
,44
1 EuXXPLCt6 0:pjiA tovy xaOpytc x. E. A&XapC YLa tT)V napaXdp~ a tTx 68EVs ,VTs s XaL
8nioOtEuOTs tou OXE-tLXo0 uXtxou, xAcO') XaL YLa tLc notxUtsLjI uno8Wt~Et tou. EuXaPLGtC( EnttaTx-tTV
xaeOyprtptV x. B. Kovtoptlv YLa tT)VcXV&yV&OT)tOU XELPOYP&(POU XaCL
tL XP~OLVEs 8tOpOUXoEtL,xa0ck XaL
tous npototajiuvous tr5 IB' Ecpoprta ApXaLotoC5tv x. I. JUXOYLOU-Av8p6ouXat Lx. H. Av8p~ou ra TLs
8LEUXOXU'VGEL IOU XpEt>XvXa-CattY)8L&pxEtL tr5 6pEuvos oto MouGEEo I(avvtv(v.
tLOU ,o Ot tp(oypccptaE
t6)V scpp(XYLCII&v ocprtXovtXt oto cp&toyp(cpotou Mouo{EouIcxavvtv&)v x. B. Hlltv6nouXo.
Eto xEtE~vo, ExtO6 c6 tto PtLXLtoYpWcpx6s GuVToVOipaypEEt, XPa LLOUVaL Xat 'a napaxat&
14 0
avavcxayia avaVipet 6ct < OupiOTaav ... xat xepatLibe4tic extnoug sntypad&~,&It NOpViCaaxoU,
AL6s Ndou ... tLouo>,(A&xaprj 1971, ceX. 126) xat aloX <<AL6c Ndou, &it NoptIAaaxou,-LtoU>>
(A&xaprS 1969P, ceX. 26 A&xapnq 1969a, acX. 19). Epay~aat-ca 6Vcaq lie tra yp&VLaIra-tBou Bev
iXOUvfpee(, nap&tTv entaCa LVT 4peuva, aOtqLaxoOfxeq tou MouacJouIoavXLvwvv.
Ext6q 6Vca ac6 ra tapandwt acpppaytapa-a,acv&Vcaaaro touxdo uXLx6ac6 ti Ac&bvisv PpMOixe
xat iva ivttno ayppytLOa as
aXcL a -rptycSvou,lie eyyeypaVpvo xuxXo tou btaxoacetratLacwcpmX&
Vs e&cpuXuo p6&axa. Ectq yxCv)q tou rptySivou(povtrat etEtra XaL&>> tTaypVI4a-ca -tqq etypaypq
AHI[MOIEI]A. Ta yp&VLaI4a xat Btax6aV as dvsat ixtuta (PI. 60:a). To an&V6tVaq xepaVEBaq
X
Bev i(epe xaVL&avaaxaqpx VMLet{! xat aptO46MouacEou. Kavov x& Xom6vtco napant&o aqpdytaOa
itpete vaevcaXvtO ati B' otif&a, av xaczt ti bt&pxetatrj 4peuvaqbe btanta6Vvace 6tL (aca tp6xerczt
yta to a(ppytaL4a iou avayipet o Adxapcp (Adxaprj 1952, ceX. 384) 6Lt iXet fPpeeEatqv KaatpE-tac
Icxavvsvcav xat csavacaV3&vcl acv-tola o Franke (Franke 1961, csX.312 [6]), Xapt 6ticaq qca-toypaypa. H
entypay( AHMOEIA auvavt&trLat noX aUX>v ac aypay(aJaara xepaEcLCav xat a~fa(vtc 6tCL6T XepdvCacf
av)xetL aqv L8toxqTaca Ltou Brgoa(ou. BX. a>Xcr.x&OpX&vboq1955, ceX. 114 Hubner 1976, acX. 180,
ajni. 29. E-q MaxeBovwaaasv- auv-ropoypaqrisvi (AHMO) as xepaVEBaac6 tiv Hflxa (Ka~xc&a
1988, ceX. 101, arjV. 314, 315), cv6 ocSCetatsitngr as oapaybraoata ac6 tvqvAXfavsa xat CtLX6tcpa
ac6 Tsv neptOX4acv&Vea atov A6o xat aro revo5ao (Ceka 1982, asX. 104, 105, 114, 116-118,
6nou xaL Xa~6XapT f3tLfXtoypaqpa. lIp,. Andrea 1976,acX. 348, ndv. X1V:2,4). BX. sitnga tq xitn
AaV6aLoq as n4XtXv aypay(&a tou 3ou at. (Albanien,acX. 278, ap. 150). Etiv 'Hnstpo e(vat yvcaac
=a6 aypaypayaoLwa tv A,4paxcax tie o a6VPoXotq nt6X, BrX.opeXiaxo (Patcuxo)xat acn6 rtq No
nXeupi tou ti Xitf [A]AMO[E]IA (Franke 1961, aeX. 312 [8], 315, 316, Iw. 61:5 TCoupdpa-Eo6Xn
1984, aeX. 432 T~oupcpa-Eo6Xv 1987-1988, acX. 101, a"V. 29). Ent(oaroev Ap3ppaxcaPpFOvrxc -Va
n'tXvou utpoacaX4vaVs tyxoLXata yp&V a-ta AAMOUION (Boxotonto6Xou 1976, acX. 194), vcvoSevr
Kaaactnv PpiOvxav 2 X&XxLvaoawOVi& (oxtst 14) tou qpouv to acv&yXuo Vov6ypavva AA[MOEIA]
(Axxapnq1989, csX.57, it(v.17:a).
15 TaEzta npoPX4Vafra
auvacvt&xacve( oaqv tpoan&OCLa a XpraOLonotiast6,t tLOX5Ct
yta rtL sCnta7V4C
entypaLTpc xpovo~oyrx6xpcrpto yta ra tcxtovLx& anVeia xat ctq
entypaLq iou Xapdaaowvtat ac6
xruq'rXcvE-t.Ifnitza-AcX?o6Zou1967-1968, aeX.383.
xat yta orvaxi
16 rta aX?tewx P3L3Xtoypaqpa a bLctacpoppq nmv
oavq Vopq4 ypajd&msv 6Xt V6vo
acv&VLcoaca ec~nyaVeqxat Xcatxiq cntypayk, axXX xat asvdtcaa c ta yp&VLa4ra ntzypay6v tiq &ctaq
ntpL68ou tou tpoipXovtcat ac6 ttaCop?TX?q pLoX? j ax6pV xat av&coaacL a yp&VLaIra q taqLa
epntpfajy fX. BXaXonodXou-OLxov6Vou 1986, aso. 257, 264, aoV. 28-30, aoX. 265, aoV. 31-36.
17 KaXcXa&d 1988, aoX. 104, 6nou xat oXerLx4PtPXtoypay(a.
186 AMALIAVLACHOPOULOU-OIKONOMOU
A' OMAAA
I. Mov6ypaxpA
j
To Vov6ypaVuca > (Pls. 60:b, 61, 62:a) epnepLeXe-aL e Te-pa6.ywvo vcTuno
xaL nepLtipdcver -a ypa'Vra-a A, P, T, N, K, entctv evwuna. H ecOJcepmXA
cOppdcyLOcLa
xcou A e{vatL OXeacxivv, evw' crvv opLC6vtLaxepaCa, nou -o etLwrcepeL,cs6Covmt
xepaCa
axpe?6ver nou xa- uO6vovcxaL npor -anarv. Iap6Xo nou r ,uopqp xa -coero ye~og
rcv ypa ua-cv e{vatLEBa, caxrr6oo eOL T)v ivWcunwv-pe-c~ayw)vwvntOxcxouv,
ooL8Lac-XaO
npayvua nou mtOav&vOIOaivm 6-cL -a cupppaytcY~axca npoepXoV-atLa6 8Lbapope-rxex6
cypayE8~e. 'EXouv Ppe0e[ -ca papaxa-w oyppay'Icvraca:
ayyec(V30 xxn.
Ka-6 -ov Aoxapr31 a-o Vov6ypaVqa e IpTspLXovatL ra yp Vacaa A, II, P, T, N,
K xaL eLvu =tOav5v.mvayvwaoK[OI]N[O]N AHI[EI]P[O]TAN.3 H &ao+n rou Franke33
6-t np6mx-tX yta -cn auvtoioypacpLa rou ovorro esv6r,ebvu~ou &pXov-a,34aCXX& XOL
n LtOav6-qra35va np6xraL yLa -coovoua
?cou -Cxvvc36,uaqobyyotv va avaoyrCcoouve
-Tv epjivvda rou ,uovoypa ua-ogr ce [LLa Ompa' HnteLpwcLXw'v ovoyiaxwv nou ev-
-a
nepLexouv napanivw ypa ?[a-ca, 6nwc. X X a ovo6ia-r 'Axpa caxX-v, 'Avbpoxa8a&,
'Av8p xxacx, 'AvBpoxxec8rp, 'Av8poxxe!wv, 'AvBpoxos, 'Av8p6vLxo, 'Avrtpxaq,
'ApLacoxp&r xXn.37 As OaeICwOe otav8koa ac*'6rov Franke OXecLxa 6OVoLou
[Aovoyp&V?axoqro e -co 6vo~ua'AvBp6vLxos.38
26 rta
vovoyovva-a a appay avaa xepavwa?&v at6 tn MaxBovca jX. Kat-a6q 1988, aeX. 97-105
xat at6 tvnvAX),avca PX. Andrea 1976, aeX. 348, idv. 14:14-16 Ugolini 1942, aeX. 232 (4), LX. 244.
Ee axian VEto a5voxo s(av acppaytav6L-cvac xtpaVEteq-a VIovoyp6V4a-La auvavc6v-at anavt6trpa.
Hubner 1973, aeX. 86. rL( au X&kylia-ca ypa tVtccv at6 tvnv AMttxLXaL trnv EXeua~vaPX. Httca-
Ae?XpoiCou1967-1968, aeX. 384, anv.. 28-30.
27 rLa Vovoyp6V.a-ca ac Xap43 appopicv at6 to Nexpopavcto Iou AXipov-ca PX. BXaXoto5Xou-
B' OMAAA
I. IAPA, OarepyvLPm3mvX8VLtxaLo
eOo( qou oara Ge xepauv6 Ve wxxnepo.
Yxo xzvrpo rc6)v O-rpoyyuxcr)v ivWuncav OcppayLaOarCXv -rxj B'I xaomyoplaq (Pls. 62:b,
63) taptLOtive-tcL ivaq act6q ntpoq -ca Be& JtOU oatci cov xevtpCxd6Beo?6 esv6 opLt6v-
'rLOu xepauvo6. 0 xepauv6q ano-ceXelt-L cot6 8o av'rLx6pucpya OUv4utiupLXXa 6?opcppa
Au c acp8
aay b5Xwov tcoavoeae&Xiv. K&'rw aco6 toy xcvtpLXo Beoa.6 Lxovt4e-tL iva
wxxOn-cpo.50 'O~r) r) Jaacspaal nepLfa3XXet-L cot6 meycpL Xa~v8Lq Ve Beoa6 acto
x&i'r Vipoq. ApLatp&i xaL Be& kot6 tov ae-ocp6pov'rm ra yp&Vua-ra IA xaL appLarepi
xaOLBe~Ldct6 'ro wox6xrrpo ra ypau'vara PA. (p'peratLeno~ivca T entLypacp5 IAPA,
8cOpLxo6 t(cton trj ?XRvq .5 ' 'Exouv Ppe0ot -a ~apaxdcx-c acppyaylav'raa:
49 rLX tCL OLXOBOVLXq aiLq tou BoutXu-qptout xa1L-tav-at-OOLXEq OtyEq tOu PX. BXcxXono6Xou-
OLxov6iiou 1986, asX. 222, a"V. 41, aeX. 239, 244, 247, 248 BXaXonoXou-OLxov6Iou 1988-1989,
ae). 84.
50 rLa TatCOX6TEpa pX. Aixap% 1980p, aeX. 27-34.
5' BX. Frisk 1960, ato X4t.a. IIpp. Boisacq 1950, a-o XAida, 6nou XaL OXVCLX5PLPXLOTypE(pE YLa
-! X (oui
) X(PO XOELtotu ftitov.
52
Ahxapv% 1980p, aeX. 30, idv. 4:2.
190 AMALIAVLACHOPOULOU-OIKONOMOU
H
HnLypocyp p ct
auvoVtctcL oxv cxvOV OE GYppaytc-rac xO(Oo-
xepa4Itl8xcv53 XOCL
ptLL 6'rLI xepaticaal a VXaL a'CrV LB&OX'rjca'rou Iepo6. TnovotrCatL v) ?kv <<xep4LtL5>,
BX. otepa'xepautiL>, yL'axur6xaOLtep q yoprs aucvavra&CaL tier oov'rtno p, X.
d<trpoLxgpa4I0>.54 H Xij & oatvrX&xOL Co'qv e6oveiv xacrvyopECaGcppOCyLad&'r(av
'
(B'II),55x~xXVpLO 'rou'rvEXouq -rou 3ou rcv
raV a
-rou2ou ac. t6 -rr Ac8avrl.56
rLa-rovae-r6, 'rov<<-XeL6-ra-rov
nerelvcv-O,57X(L 'rovxepauv6, -co<<Zlv6q&yptnvov
eLvou yvco'r6 6'rLauvOov'raL &itecna tie -rc Xa-
A6yac>>,58
PfXo3, xa-racprcLrjq dxnvicav
'rpeca'rouAL6q.59Enolikvcs ce auvdcp-rjajtie -rj ?vite &uvovorau: <<'Iepa& xepatIs
OpXivBoq 1955, aeX. 114, anV. 4, 6nou XaL aXtLXi PLPXLOTypapta. EultXTps C4Va-Lxi pX. Daux
53
1959, aeX. 628 Lehmann 1960, aeX. 37, 38, 109-118 HMnna-AeXpouZou 1967-1968, Gex. 384, GiV. 28,
31, 6tou pX. XaLYLaTo a6Vnsey a tAv ypaVV.itcwvHP, itotawq BTXCSVELTo eitWdto
hLrpoo 4 h
54 OpXivBoq 1955, aeX. 114.
55 BX.napax&-ccaX. 194-195.
56
A~xap% 1986p, aeX. 107.
57
'O.Vp. E 247.
58 AtaX. Ilpoi?. 358.
59 0
ae-6q, cAeL~p6 xtv6 tou AL6q, (vaL WoMb yvcaat6q c6 ivav VeyiXo apLOV6
ipycov -q nxaawxf,
tTq VLxpo-reXvto(q XEL 6tou GUv5Ocqnapnt&VE-atLoZ0uq va etaxovrtECL VE
tT% VOVLGV0EaCOXOstdaq, to
TA E(PArIEMATA KEPAMILQNAHO TO IEPO THE AnAnNHE 191
BEtt XEPLtOU tOV xepauv6, npop3XXovtroitpoq tv<vxat Ouva" OUt~5to pLataep6 XiPL,6itou auXv&
X&0CtELato oviatpoyo tvq Xa6~.t o aet6q. BX. Oder, RE I, at. 371-375.
60 rPL TO
auVoXlo -cwvovoV6-rcovfX. Brommer 1988, aLX.69, 70, 6nou Xa1 aeLtXVLX5PLXLOypayta.
61 Ext6q Hntapou ac*6 XaL LMtypay OIQNOE eLxovEeVtaLaG aGyppyLtapa t
xepaVEBaq ApXatxfj
nEpL68ou a=6 to Ka~X=68&BOLCOt-Oa (Felsch 1990, aLX. 303, eLx. 1:a, aLx. 316, B2, ndv.51) XaL aex6
tOUapCptELivcx yE8& XacL MtpayLN P.ANILI.PFaE GypiyLaa pP.atx5q xepaLiaq a6 tvv ItaBcx (Del
Chiaro 1989, GEX. 59).
62 rLa napata-C6G toaEC xaxapauvou atv-V 'HntLpO PX.B~aXonou'ou-Otxov6Vou 1986, aOX. 192-
201, 203, ajji. 13, aLX. 204, ajji. 14, OLX. 254-258.
63 H napiataaj avcot GExepauv6 ENv=LyvCOat atvv 'HntLpO xuptca a=6 tCL JXLLMX~ anotxeq,
MLtL5 aonotXobae to auXv6tcpo Xvipkjtia at vojiajta-a tcv HMecav(Schrader 1937, aOX. 213, 214
Franke 1961, GEX. 94). Ait6 tE trXj t)ou 5ou aL., Ltd oEap~ita,XOELcti6 to a' VLa6t)ou 4ou aL., Lid
AWx-rt, OLMo)oaaoE UtLOOEtCOv toovact6 XacLtov xepauv6 VaCEVe .qv npotoni~ tOU AL6q Ca ?OvLxi
tout a443poXa.Ta No tLXetu-ta XPI t)Lono n XV6 Caq? CXftOOV aOtLO06ttunltotiv VOJ4LOI46-t6)V
tout (Franke 1961, aOX. 89, idv. 8:R3-R12, 9:R13, R14, 11). Eto P'VLa6 t)ou 4ou M. o VoXoaa6q
PaGLXL6qAX cavBpoq A' (342-331/30), xa=t tv B&ipXELa-Cj exatpactac tot)ouatvv Kitc IwtMa,
ixo?e VEa ELtp6 XpuCaV, aaTj.LVLWV XaCLX~iXLVWV VOVLGV.it(A)VVLE tap6iataar aCLtOt xaL xEpaUvo6,
tou avtuyp6(yOuV tca atp,4oxa tG)v VOXOGGLXC'OV
VOVLaLit)v (Franke 1961, aOX. 89, 90, idv. 55:12).
To capflmyV6 tou axoXouOvae i HMtpCXtLXfEtuiLpcXt (Franke 1961, idv. 55:12, 14-16), o IIHppoq
(297-272) xaCc-to Kowv6 uav HntCLpC0tCV(Franke 1961, idv. 20-31). rPL ti.q TC0CpCat-ardGE totU aCc-oCas
xepatuv6, lTOUXPGLVOntCOLdxtccL Wq 4.43X t Ea 6Xc aXeL6v t voVtapa t tav HItoXe~aLccvtv Aqixtou
P3X.BXxXonotXou-OLxov6Vou 1986, aOX.198, 206, ajV. 23, OG. 207, aov. 24, 25 XCXL
YataLq T OpCXatq
GElcaxeBovxi voiAaliact ((D.bXEnou E' xaL Hepahq) PfX. BXcXonoOXou-OLxov64?ou 1986, aOX. 199, 207,
auV. 28, aOX. 208, alli. 36.
64 r.a Tn XpovoX6yTa)n tT) t u)
abataarj jt)-aXtaX PX. Aixapcp 1986a, aeX. 118 x0cc aIV. 32, 33.
65 rLac-co Kowv6 trv HstELpC0tC6V,PX. capacitv OLX. 188, auV. 44.
66 A^xapr 1980a, aOX.26.
67 Franke 1961, idv. 20-31.
68 rta -ov xepauv6 a-qv apXaca avCtoXLXf xac eXXsvtxfi tcXvr PX.Jacobsthal 1906, aeX. 39 xeL.
69 rca c Tv -tu)oXoyca tou x
tpauvo xac yia tCL capaccaycc tOU aX4ccat6qtou xaxt& tvnv EXX)nvAtLXt
neptLoo PX. BXaXono6Xou-OLxov6jiou 1986, GEX. 205, a"V?.17, 6nou xac Xetux5 PftXioypayca.
192 AMALIAVLACHOPOULOU-OIKONOMOU
nep~poXo, nou iXouv anox6+m xxti tTv xaxaaxcuv touq, xa a6yXpovTle tTv xaxtaxeuK tccv nup-
yoeL86v avanX 6&rcav tou Oe6tpouxatL tv env8u&a tou x-ptou M. EneL85 to M XpovoXoyetaL atov
4o mL.xaL t-a avaXf4jiiata tou XOLXOU nLOav6ta-t atouq Xp6vouq roufl6ppou (297-272), i xataaxeuK
tou E2 XpovoXoyetatrOtL apXiq tou 3ou aL. EndaEs,i Xpnaonosoa tou ajuuoXEOou yta touq xloveq,
7ou elvaL yvxcrtn ac6 tcL apXiq tou 3ou a. xaL 1 opOLqtTXtatou E2 xatLtv ato&q, ca npoq tCvv
.le
tOXoo8?ojia t axiv5 tou Oe&tpou xaL &g npos ti jopyn le ti atoi tvj axTvfs, evLaX6ouvtY
XpovoX6TyaTtou E2 xatLti atoiq auq apXt q tou 3ou a. ALxagpq 1966, aex. 80, 81.
94 BX. napanivc aeX. 188, avji. 46.
tvq Iepdq OLXaqxaxt to 8' 'd-tapto tou 4ou a., 6-av 8T)uLOupyTktd
95
lIpp. emL8LpO6)aO tvq atcys
o LaotojI x6q eEp(poXo0 aE txaxaa" -ou xuxXLxo6nepLp6XoUpe touq tpbntoe. BXaXono6Xou-
OLxov64uou 1986, aeX. 100, 101, 110, a"ji. 51.
96 ruzx-ra-rjptir xtqqugpda-cu ariyrS r onotmcEXcn\Xd?So30,20 ji., exr6q =6 uqSmvqtptbe xm-r
jijxoq 6Xwvtcwvto(Xwv tou E2, ato eawctpmx6tou unipXav 86o amtpiqan6 3 wVLXO(Mx(oveq, OLOnO(OL
xat& -qv avomxob6,iTaTjie tTV at6)Xmx xawtatpoy4 (218) aut5Ovxasv te tvv npoaO5xTduXcv 86o.
A~xapvs 1986p, asX. 58-61.
97
'Olji%.II 233-235.
194 AMALIAVLACHOPOULOU-OIKONOMOU
II. IEPA
H ntLypocpc IEPA (P1. 64) ,uiepLtXeotacL opOoyCvLo ivruno Ocpp&yLOa4a nou
OpE~TOCL c=6Xe-rs -rocvLO
XOCLe(O-pMLX&acGTpOCyaCXW-rl r o07otCa
BLCx6a'iirGi, OaOTeXeUrOL
c=6 evoXacsoo6tivv BtX)r x&O'ri ToLvtaxaL xot4pto. H Btax6tivuao xaL ra ypa&itara
etvaL 6x-runa. 'EXouv ppe0et ra napaxdctc oqpaytioa-ra:
Komr&-Tv &o1tO(J TOU o axap? To OcpP&YLaOt l Tie X-q tepM a~V)Xeto lrLEx
5
Tou 4ou TLt apXer -ou 3ou aL. Q?Zo6oo,ave ap-qt-a aco6 Tls 8! BlTUatGIEVE
mLupu&etqytcL -r Buva-r6-rxaXpovoX6ynorq-G)v OcppayLaVI&-r)v Ve pac&l-q ,uopyp)
YGVypTtA-,O&vl ra ypavv4ra TOU OUyXEXp4li6VOU ocppaytolacrocBev tpoocEppouv
xatiq&acycpxcXpOVOXOylXt 6V8eLi7. To &Xpai.X. tie opt6vrta T-qVeaoGreplX1xEpaca
vaL tlev npounapxeL TOU Xcxyati TzOxacqlOvr Trv eaOG)pXtXgxepala, AMX ecaxoxouoet
va e ovat ae TacuT6Xpov7Xp~a 6Xt ti6vo oe entypaycS TO)u3ou at.,102 AM~Xxat xaCx T7
8t&pxeLa-rou2ou xat TOu lou aL.103
E~4Uou 7) EXXl(e Oap4pcaproypacpLx6v eVklte)V, Oe OauV&pTpal ti Tls napanitwv
BuaxoOMe,ano1p1tsL -qv acpacr xpovoX6ylalrircv OapayLO4&tA)cv. M6vov 6vaterminus
antequemintopet va BoOetxaL auCt6elvaLIj pwtpclxn xacorarpoq5 Tou 168/7,104 (a(;
o awvaxacqic avayepeL1 6-L To O(ayPp&YLca ,Ti- X6in pp6oTXe OTO "ouVexe(
OT(ptav TTl pCa4LocLX' xaTraaTpoqpq, tie p0ova Ouv p4tLaTa Xep&4ca)VOTar7yq>>nOU
tanltLOT-Or(xeoe 6XI TTrv ExTcaal Tou v6TLoU T-y4LTos Trou E2. 0 tlxp65 apLOti6q
106 To Ypp&yLa4Cca ap. 1 npoipXeraL ac6 rv avaaxapq5tou v6tLou qn5pa-roqtou E2 (A&xaprs 1969p,
reX.26) xaL To aCpp&yLCO4a ap. 2 npopXpetaLac6 tvx avaaxazp5 tou 1965, nou Jyve aro Xepo avipeva
atrv NA ycavLa -rouE2 xaL Tou vataxou A npoq A (A&xapnq 1965, aeX. 53).
107
atdEq Tou ,8X.napaniwa aeX. 189, anji. 49.
rLa tLq OLXOBOIILXiq YpaELq tou E2 xatL tL aCv-tLOtOLXE
108 BX. apan&vcaaeX. 189-194.
109BX.napaniwa aeX. 193.
196 AMALIAVLACHOPOULOU-OIKONOMOU
12. 'T+. 0,02 ji. M75x.0,06 ji. eial: Eiapo* avaaxap5 1959.
llvX6S uitpuOpo? xLtpLvcan6q (5YR 7/6). 410E[NAOY]
8iar: Ava-ro?,xdtou E2 avaaxaqpf1965.
22. Ap.M.: 5725.
AIOE[NAOY]
'T+. 0,04 ji. M5x. 0,07 ji.
13. Ap.M.: 5745 (PI.67:a). llj65 unipuOpoqxvrptvca6q (7.5YR 6/6).
'T+. 0,01 ji. M5x. 0,04 ji. leia: E2- avaaxaq5 1971.
qX65uinpuOpoqxvrpLvct6q (5YR 7/6). AIOE[NAOY]
Avaaxaqpf1971. 23. Ap.M.: 4095."'
[AIor]NAOM
'T+. 0,05 ji. Mfx. 0,07 ji.
14. 'T+. 0,02 ji. M5x. 0,04 ji. llj6s unipuOpoqxvrptvcIa6q (5YR 6/8).
ljX6s unipuOpoqxvrpLvwt6q (5YR 7/6). eiaF: E2- avaaxaqp 1966.
[AIOENA]OY AIO[ENAOY]
15. 'T+. 0,02 ji. Mfx. 0,05 ji. 24. 'T+. 0,05 ji. Mfx. 0,07 ji. (PI. 68:a, apt-
jX6si untpuOpoq xvrpLvwt6q (5YR 7/8). ateep&).
Avaaxapq 1959. lljX65 unpuOpoq xvrptvcan6q (5YR 7/8).
[AlOE]NAOM tou E2- avaxacq 1972.
Geir: EJto&
16. 'T+. 0,083 ji. M5x. 0,21 ji.
AIO[ENAOY]
jX6si untpuOpoq xrp Lvct6q (5YR 6/8). 25. Ap.M.: 3487 (PI.69:b, Be~t&).1 12
Geia: 0 avaaxaqpf1990. 'T+. 0,03 ji. M5x. 0,08 ji.
[AI]QENAOY IIrX6i unpuOpoq xvrptvcan6q (5YR 6/8).
eMGs:eia-rpo- avaaxaqp 1959.
17. 'T+. 0,011 V. M~x. 0,11 ji.
A1O4[NAOY]
xurpLvcIa6q(5YR 6/8).
jX6si untpuOpoq
Geia: 0 avaaxazp 1990. 26. Ap.M.: 5736.
[AIOEN]AOY 'T+. 0,04 ji. M5x. 0,05 ji.
IIrX6i unpuOpoq xvrptvcan6q (5YR 6/6).
18. Ap.M.: 4086 (PI.67:b, apLatGp&).
e3aF: E2- avaaxaqp 1968.
'T+. 0,02 ji. M5x. 0,04 ji.
f qX6sunipuOpoqxLtrpwLVCa6q AIO[ENAOY]
(5YR 6/4).
Geia: eia-rpo- avaaxaqp 1959. 27. 'T+. 0,05 ji. Mfx. 0,05 ji. (PI.68:a, iAcrov).
A[IOENAOY] fMX6qunipuOpoq (2.5YR 6/6).
c
eicr: 01, bpcitlAto avaaxcxazq 1982.
19. Ap.M.: 5726 (PI.67:b, BEL&).
AIOE[NAOY]
'T+. 0,02 j. M~x. 0,11 j.
HrX6i unpuOpoq xvrpLvcan6q(5YR 6/8). 28. 'T+. 0,04 ji. M5x. 0,05 V.
Avaaxaqp 1971. jX65iuntpuOpoq xvrptvcan6q(5YR 7/6).
[AIOE]NAOM 8eia: E2- avaaxaqp 1971.
A[IOENAOY]
20. Ap.M.: 3484 (PI.68:b, En&VCO
apLatGp&).'10
'T+. 0,05 V. Mfx. 0,09 V. 29. 'T+. 0,05 ji. M5x. 0,06 ji.
HMX6unipuOpoq (2.5YR 6/8). fvX6s unipuOpoqxvrptvcan6q (SYR 7/6).
EM": e@a-po*avaaxaqp 1959. eiao: eia-po- avaaxaqp 1959.
AIO:[NAOY] AIOE[NAOY]
21. Ap.M.: 3486 (P1.69:a, apLat-p&). 30. Ap.M.: 5744 (P1.68:b, En&VCOBL&).
'T+. 0,05 V. Mrx. 0,06 V. 'T". 0,047 ti. Mrx. 0,087 ti.
HrX6i unpuOpoq xLtpLvcan6q(5YR 6/8). IIrX6i unpuOpoqxutp Lvcan6q(5YR 6/8).
110
Axxap% 1960, aEX. 32, ndv.12:y Franke 1961, aFE. 312 (5), 315, rdv. 61:1a.
"' A&xxap%1966, aFE. 82, ndv.82:x.
112
A&xap-5 1960, aEX. 32, afv. 12:y.
TA E(PAriPMATA KEPAMIASIN AHO TO IEPO THE AQAQNHE 197
eAia: eia-po* avaaxayp 1959. 35. Ap.M.: 5742 (PI. 69:b, apLatGp&).
[AI]OE[NAOY] 'T+. 0,048 ji. Mfx. 0,078 i.
7I-X6qunipuOpoq(2.5YR 6/8).
31. Ap.M.: 5735. 8imn: eMrrpo*avaaxapy51959.
'T. 0,03 V. M5x. 0,06 V. [A!O]NAOM
II-qX6qunipuOpor,xvrpwLvca6q(5YR 6/8). 36. Ap.M.: 3670 (P1.68:b, x&TCOBL&).
eima: E2- avaaxaqp 1968.
'T+. 0,034 ji. M5x. 0,07 i.
[AI]IOE[NAOY]
IIfqX6qunipuOpoqxvrpLvcan6q(5YR 6/8).
[AIOEl]NA[OY]
32. Ap.M.: 5741 (PI. 69:a, betL&).
'T+. 0,049 ji. Mfx. 0,077 i. 37. Ap.M.: 3917 (P1.68:b, x&tCCappLaGp&).113
II-qX6qunipuOpoqxvrpLvcwa6q(5YR 6/6). 'T+. 0,048 ji. M5x. 0,13 ji.
Eima: E@a-po* avaaxaqp 1959. fMX6qunipuOpoq xaca-rv6q(2.5YR 5/4).
[A]IOEN[AOY] Oena: eia-po* avaaxaqp 1959.
[AIOQ]NAOY
33. Ap.M.: 5738.
38. 'T+. 0,049 V. M5x. 0,068 V.
'T. 0,02 V. M5x. 0,06 V.
(2.5YR 5/6).
lliX6q epuOp6q
HIIX6iunpuOpoq(2.5YR 6/8).
Omae:01 cavaaxaqp1981.
[AIO]ENA[OY]
[AIOEN]AOY
34. 'T+. 0,049 V. Mfx. 0,07 V. (P1.68:a, betL&). 39. 'T+. 0,048 ji. M5x. 0,07 ji.
HIIX6iunpuOpoqxurpLvcIa6q(5YR 6/6). HvX6qunipuOpoq (2.5YR 6/8).
(esa: 01- 84IAtLO 8- avaxcaGpa51982. (fao: 01, BC4LtLO Anavacxazp 1981.
[AIO]ENA[OY] [AIOEN]AOY
VE yCOVL6c;8 TrV ?CaXCpLXfi xEpaa XaLt CxpeV6vcxnpoq tg x6-tca atjV apXaLEp5 xpEaLa Tnq EaCCtpLX)5q
ycavLa. To 64LxpovdvaL at poyyuX6 XaLLaOoliye0e. M6vo Oao Oapp&yTLOF4c ap. 10 (PI. 66:a) 8VXXpEVCtaL
a8C&yVC)OatOTpLycOVx6aX~5ia (nt0av6v UXta) oto xev6 av&i?eaaaiv nta&YL( betL& xEpaLa tou UXta
XtL to yt;LCtX, VC tO X&tC) acXXoq ntap&XTn)O npOq Tn Btr L xpEzaXEcou UOXtc. ME EMTOXCcu n EstEana
exqpp6Cvat n m0av6trntra va np6xerat yta tr auvtrolioypayqAa trj Xrnq 8(a(loaa), n onoda aUVxVtatE
auXV6oa-a aypaytaaa TCOV xgpaVE8C0V (pX. napaniwo aeX. 185, aniA.14 XaL napax&CC aeX. 206).
116 llpp. 8UO &c1XOao4YrLX6tpty(,)VCz XV64V4aCz
Oat yp644iVtX tEnq entyp( OypiyL
P4IVCOOU, OE
Ceka 1982, aeX. 109, ndv.HII:31.
cu6 trjv AXpcavLa:
xepaVEBLa
198 AMALIAVLACHOPOULOU-OIKONOMOU
At6q xat Ncou. Eto 6Xo GyppcytcG~La elvct atcLaOY',aautLO,u~ vri BtaxoaVLujtLX5-caij.
E-qjv y' unoxaxti-yopla (P1. 67:b) ev&cOcGov-at ,u6vo No ano-oVnua-ca (ap. 18, 19), ac6
'c ostolca aco npcato 8mac6exLat ,u6vo to ypaqCuua Wkcxa. To aypyaytacxo elvoLte'vtuto
OpOOyCOvtove ixtcuna ypa&[1La-a,17 4ouq ?0,0118 ,u. EX-cv 8' unoxwxrqyopla nept-
XaxpL3vov-atxLa Opacioactcx ap. 20-39 (Pls. 68, 69). Tot yp&,uqoacol8 elvat ivtuna,
6+ouq ?0,024 ,u., xat opLCovXat xax&a1nIxoq a=6 86o napa&XXvXeq,stnlraq ivtuneq
tawlveq.
H entxXvjoj tou AL6cqtrlq ACOOvrq caq Na'Lo o Nzoq 5'Naoq etvat n topa XoM
yvscota. H ntaXaL6t~pv iPVveta o-T ypaoc'T nap6ooo ava'yetaL oTov 40 aL., AM&
enLypacpyxa anoetOVLxvuet aPXaL6PTepr. 19 Ta nL[Oexux lv 6atofr tou Euayye-
X[8r xat tou Acxapn,120 npeneL va ucxeX-tCLOOUv Ovu tq Xe'{etq v&a-vcx6q (= Mp&
otxta) Iou B&IXsvouvtCjV xacXotxcaX ou OeO6, I oota e6XetBtantoaoel YLXOXOyLX
xcXL avacwxcxpL4',l2' axcsx~otLXc spos tvj X (j IyOvOCxoc (= acuT6qnou xaxtotxel a-vj
cpvy6122). H ypcpyi NOCO5123 eXeL nO8eLXOe e l LXOElXEOCT pj. ECOCppCxyt-
CMTOC Ve TjV 7ELYPOC(pAL65Naou nLOaV6vU7EOVOelCaL v XE'v t (8vx. (kIepc xepavcL
Al65 Nocou>>). 124 HoX* XOLV, E(V0l ) xP5a7 ,u6vo 1ouCtlOErou M125 5 ,u6vo tr*
yeVLX'q X-TCtLX'$ tOU 5OO tvjq Oe'cq Tcov Oedav
rLa -o 6votia tou AL6q tie avtaotoLXeq cmtLxX)am 6nwtq AL6q 'Oxu nlou act6 tjv OMutnda xaL
tiepLx& yp&titiaw ato6 tv cmttypap)
Ze5 'EXX&vLoq act6 tcv AtyLvaPX. Franke 1961, aoX. 315, aorV.14,
15. rLa -o 6vota AL6q cc aypp&yLtaOxep&Viouact6 trv AXpav(a PX. Ceka 1982, aoX. 105, ap. 12,
aoX. 120, ndv.I:12.
127 rLa trc Xp~ai t-ou A
te yvW68Y tY<v caw)LcpmXxepata PX. napai&vw aeX. 188, aori. 40-43.
128 rLa t-o
Vnvoc8E otytia PX. Kiaffenbach 1957, aoX. 43. llpp. Hllats 1961, aoX. 40. Epay~atlat
Iaxcv8OvLxvxVpaXlp8WV tie tlDVOCL8Ei GtyvOaXpovo~oyo~vxtL aota ictrtr paaoXesaq-cou XbubztouE'
(MntaxaX&xrq 1934, aIX. 104-113, cLx. 5-8), evC mttypap5 act6 tr< A8wvr, aotrvontodaauvavt&6IatL V-
vocLq acytia av ieca cc &XXatctppaaxeX), XpovoXoyTEatLant6 tov EuayyextB a-a 370 (EuayyeXEM&q
1956, aeX. 4). Et-L apyupiq 8paXty~q6V?g tou KoLvo6tcv HncLpwt-v tr5 ntepL68ou215-195 Xpna-
iontomatL to tDnvocL8E oatira. Franke 1961, aoX. 142-145.
129 KaXczat&1988, aeX. 105.
130 rLa -rNXp~ao tav axpeti6vwv a Xpovo~oyzo6 Xppt5po PX. napai&vw aoX. 188, art?. 39.
131
criv aucw6XpovrXp5aj tou &Xpa tie opLC6v-tLatYv cawpXcp)5 xepaCa xatLtou &Xpa tie yCavu'08)
rLa
PX. napantwvaoeX. 194, aorV.102, 103.
132
&&xaplq 1967-1968, aeX. 388. rLato aytia tie opLC6v-tLexpateq ntpp.Woodhead 1967, aeX. 64.
BX. cEdarq yLa to cytia xatLto vL BXaxonto6Xou-OLxov6Vou 1986, aoX. 257.
133 rLa to KoLv6tcwvHntepCotcv
PX. napantvo) c,. 188, aoV. 44.
134 A&xap-qpv 1960, aoX. 32- A&xap-qpv 1969a, acX. 19-
1966, aeX. 82- Axap-qv 1969p, aoX. 26- A&xap-qpv
Axcxapr 1971, aoX. 126.
135
xpxagq 1969p, aoX. 26.
136
BM.napa~tivo) ocX. 194, arli. 104.
137 Fraser
xaL R~nne 1957, aoX. 89.
200 AMALIAVIACHOPOULOU-OIKONOMOU
rF OMAAA
I. ArEETPATOE
E P'I
xrmia-yopta (P1.70) eLVepxrxCL I eTELyp(payArEXTPATOX ae opOoyc'vLo
iv-cufo tou
acpp&iyLOLcZ opECe-caL an6 B&iX65papaXX-Xe(Xexctk -LaLvEe46 OLoneotre
138 At6 tca evvL&
OpautiacLat(ap. 1-9) tqq a' utoxa-ctyop~qtL -a ecpt&(ap. 1-4, 6-8) 2tpoipXovtcLaut6
to E2, Iva (ap. 5) act6 avacxazp tou 1967, r onto(a endoal iyLve aotl acoo tou E2 xaL atrv ntepLoX)5
ava-roXwxtAXpL-co vataxo A (A xaps 1967, oeX. 33) xaL Iva (ap. 9) act6 -o 0 ntouPppaxetCLv6rLCa
-ou E2. Enolivw tac acppypayatia-cia t5 a' untoxa-ctyoppaq npopXovatCLXwptPaCLpLoxCa ant6 to E2.
H E8LaaventLp45axxt at68ocY Bev LaXUiL -Ca
YLCa Oyppayoatiatca tcv un6Xoutcwv uoxa-ryopLv. At6 -a
i.L Opaactoaw (ap. 10-15) trcs 3' untoxa-ctyoptaq&6o (ap. 10, 14) {vaL Xwp'qa-roLXaC,iva (ap. 11)
npoipXtcaLatn6to Oia-po, Iva (ap. 12) act6 to E2, Iva (ap. 13) act6 avacxaZp tou 1971, xa-C& tr<vonto(a
8LcpcuvOrxcto 8uLx6 tqA~Va -ou E2 (A&xapp 1971, ceX.124) xaLiva (ap. 15) ?(vaLaCL.apL-Ta o6eCvo,
?2tLB5 tpoipXctCaL acu6tvv avacxaczqtou 1959, xa-rdtr<voto(a 8LcpcUV5O7Xe t6co to OiaCrpo 6co xaL o
xwpoqv6-tLat)v avafLOiaux(v P&Opwv, av&ILeaactvv Ikp&OLxdaxaCL to vaC6tv% E4?LBoq (EuayycW8vs
1959, aeX. 114). AJt6 tca No Opa?ctaLta(ap. 16, 17) tvq 3' untoxa-cyop(aq to Iva (ap. 16) tpoipXeatcL
acu6to Oa-cpo xaCL-o &XXo(ap. 17) ant6 -rv avacxacqp tou 1971, xaCt&tYCvo0t0oa 8epeuv)OYxe to
Bu&tLx6tq~ica tou E2 (PX. napanztv). Ant6 -a etxoaL Opa~aiawat(ap. 18-37) tr% y' unoxactyopPaq
oxTa (ap. 18, 19, 23, 27, 28, 30, 33, 35) npoipXov-tcLacn6to Oia-cpo,i~t (ap. 20-22, 24, 26, 29) ant6to
E2, -raaepa (ap. 25, 32, 36, 37) ant6-o 01 xaCLNo (ap. 31, 34) e{vaL Xxpc atoLXXeka.
139 A&xaplq 1966, aeX. 82 A&xapp 1969p, aeX.-26 A&xaps 1960, ocX. 32.
l4 BX. 2tapat&v&)ceX. 188, cIV. 46, ocX. 189, cIV. 49. rLa xrs a-iyeq -ou E2 xaCLtcl cxjv~q tou
Oe&tpou[IC-&tjV arctcXL5 xatatcpocpq PX. BXaxoto5Xou-OLxov4?iou1988-1989, ceX. 84, 87.,
141 OpX&vBoq 1955, ocX. 115. Av-tOcrsav Kap68&OcapeE6-t cuv)Ocaqa-CaIp& Bev iouvie LB&aELepa
epyac-)pLa. Walter-Karydi1981, ocX. 45.
c
142 revLx6tcpa yLa tO-tLx&epyaacpa xaxpaceiLx5 v
'HitcLpo oxaL v
XCLL apiaatp A-3ppaxta, 6tou
ixouv ppeOC{ nt5xweVtllnpeq yLa tCjvxawctxeun axpoxep&Vicv(BXaxonto6Xou-OLxov4?iou 1986, ocX. 30-
33, 184-186) xat eLBwX)Mv(A&xaplp 1964y, ceX. 311, 312 Daux 1967, ceX. 681), xacowqXaLaotIv
Kacactnj, 62touiXouv actoxacXupOE eyxa-catc1&cLq Lt8jpoupyeTouxat XaXxoupycTxv(A&xaplp 1989,
aeX. 35, 51, 57) f3X.rpap43,&1988-1989, aeX. 119, 120.
143
A&xaplp 1967, ceX. 40-42, CLx. 5 A&xaplp 1986p, ocX. 60.
144 Aixap 1986p, ocX. 101 KouXet vrq-Boxotonto6)Xou 1975, cX,. 165.
145 Franke 1961, ocX. 123, 124, 150 Hammond 1967, ocX. 638- A&xapcp 1986p, ocX. 23- A&xapcp
1987, ceX. 14-16. llpp.rpapi&v1988-1989, ceX. 121, 62tou,exr6q -rwv&XXcv,PX.xaL aXecrLx& Ve Trv
axxBxoe&pxvnoa -wv Cntayyx4&t)v nou tU ObeXV6CL -r) XELCOUpy(axepavoupye{cv oE ouyxpotCiVat
CpyaGapt1v (axixoupyecwv, vo4tG4a-CoxotErA)vxXnt.).
146 rLa opOoyTvLo ntXatcLontou opE~ctc- ant6 &to -cawca PX. aypp&yLtiaant6 -r)v AXfpavta: Dautaj
1972, ocX. 154, ndv.IV.
TA E?PAriUMATAKEPAMIAflNAHOTO IEPOTHE AQAQNHE 201
auvaiovcL eaGYepLX&c te vwlXpif papa&XXXeX XBesxr -LavteS. H BLczx6atival xa-cot
yp&itioxraetvaL Excnuse. 'EXouv Pp~eOe
da napax&i-cco
apytpyaxaa:
1. 'TY. 0,025 V?. Mix. 0,10 V?. (PI. 70:a).147 Avaaxaqc 1935.
IIHX6M puOp6~(2.5YR 5/8). [ArE]IETPATOE
Deia: 01 avaaxaip 1981.
OX6xXrpoayp&yTLaia.
ArEETPATOE 3. Ap.M.: 3557 (PI. 70:b).149
2. Ap.M.: 271.'48 'T. 0,026 t?. M~x. 0,032 t?.
'TY. 0,026 t?. Mix. 0,069 t?. HjX6q usnpuOpoqxapmn6k (5YR 7/6).
HIjX6qeXapp&untop6B&voxaa-cav6 (5YR Avaaxayq 1958-1959.
6/4). [ArEETPAT]OE
alpophp)v nou ?(VaL ?EtnEoaxtOL:BaxXono5XoU-OLxov6Vou 1979, aoX. 279-298) px. rpap&vv 1988-1989,
aeX. 123, 124.
156 Gravani 1986, aoX. 126, ndv.126:6.
15 Gravani 1986, aoX. 129, ndv.133:3.
158 T~oupipa-Eo6Xn 1987-1988, aoX. 118, "V. 167, ?LX. 16.
159 Aixapvq 1981p, aeX. 30, 31, eLX. 50.
160 BX. napax6x-t aoX. 207, o"V. 208, 209.
xEpaVtEc(v i8~ an6 tov So aL. Felsch 1990,
161 H nap&aotav"asvOeVouE{VaLyVot5G oGppcytopatatC
oE
oeX. 305, ntv. 52. llpp. Ocpp(PPLaaanc6to Iep6 tvq Apt iLBOq atouq Aouooq. Reichel XaLWilhelm
1901, oEX. 64, ELX. 142. BX. ME",t TOaVOiLOGECYP&YLOLa ac popiaanc6tv- E&o. Garlan 1986,
aeX. 239:d.
162 rLa aV-taEpocpa
yp6V.a-Ta XaLEnLypacpy PX. Tapaotcvw aEX. 197, aV. 114 XaL apax&tCOaEX. 208.
TA EIPArIIMATA KEPAMIAQNAHO TO IEPO THE ASIASNHE 203
III.AHIOAAOAQPOTATEANIA, avOe4?o
E-a opOoycvLaivurusnopaytoiya.toc -qv T'III xa-qyoptca (P1. 72:b) xa-cd un'xoc
rTcBe~ ; tIocxp&c nxeXup&c T etypocpy5 AHIOAAOL\{POT, nopetipaf&XXeoL
YpepeCatL i
OXntaLonoleTvo eVVLayuBxo avLtLo xaL xa-XOi
pOxoc ;XO apwp~ t 7axpacTcupdc
epeparLr eypay ATXEANIA.To avO4tLo u-p cveLacvocta c=t6 &o avuOtux&
XLVOtieveV OmyLostieeb&kxxecL63 nou XOOGo~V-co xevtpLx6 -nutip oao xaxr tifpoq rou
cppcyLotlacoq. Lo
41)u)XXpto164 v opcpY iVpv(oxou,165 acxoutin oTa xopuyata &xpa
rcov extx6)v, e) Xenr6gTpo cuXX pLo Y u cpwe t6 -co to eGov 7tepbLoulJ )V e&vO
OneLpcaV.Ol66 Ta yp&utioc xoa r Bmx6atua elvatL ix-una. 'EXouv PpeeO-ta napaxdr&a
ocppaytoatica:
163
rLa !t7 yivEaG} XaL EiXL{Y tCV EiXEXV otn Pf3& V avVWEv c'oVVetrtnv Xep6Vwv PX.BX~Xono6Xou-
OLxov64ou 1986, ocX. 15-21.
164
rLa YpUXXBpLa W bLaXOOGLtLX& OtOLXEtaeXtxcv PX. BXaXonotXou-OLxov6vou 1986, omopao&x&.
165 BaxXono6Xou-OLxov64ou 1986, ocX. 168, 175, oVi. 18.
Hpp.
166 o
Aq CY)VELWOEO6-CLOt GXi&O tOU OGpay(Oca-co ap. 4 aco6rov Kapai&vo (Carapanos 1878, oce. 111,
ap. 6, ,dv. 61:11) to avOUiLo aTot taL Wq EZTt&qU>XO,?VC')?LVaL?VVL&CUXXO XaL BE 8 XVOVXEL OL
XCTO~?PCL? OtL iXLX?5, OLo0coto 6V.AcqBLaXpEVOVtCaL oay6p GOt Opa6Gva ap. 1.
167 Franke 1961, ocX. 312
(1p), ndv. 61:2.
168 Carapanos 1878, ocX. 111 (6), ndv. 61:11 Franke 1961, ocX. 312 (la). Hap&tvv CnLtOta iVT ipEuva
to Opa6Goa BEV XELPpeOCEOtLq cutOO4X?5 t0o MoUCEou I~xavvEW&V.
204 AMALIAVLACHOPOULOU-OIKONOMOU
To 6volia Auoavtac lvaL toXMyvCOr6 xaL oa &Xacz ipips rvx EX)&8ac,169 aczxx
xaL Oav nePLoX5 rrc Hntlpou, 6tou elvaL BLaBeBotevoI70 XaL tiap-cupec&u emtypa(pxi.
Exr6c -c)v &XX&)vBmaOaser Oa e7cLypa(p5 rou r6iXouc; rou 4ou aL., 7' XaL o- Upe
mypaye axc6 -cn Acadvr (SGDI, ap. 1339, 1350 xaL 1360). An6 aure oL Nto npcreg
(SGDI, ap. 1339 xaL 1350) XpovoXoyo~vcaL a=6 rov Hammond172 X(yo tpLv-o 170.
O Auoavtac, o ozTotoc avayeOpeaL; T7CLypaCpq auTrxe (SGDI, ap. 1339 xaL 1350)
Vusopel va LvatL -o ?&to np6aoano. Ac oasVIuaftct eUXXou 6&L -o 6vosua Auoavtac
OuvavtaxaL oG BpaXtin rcouKotvo6 rccv HsteLpwrcw'v,73 xc0zc XaLoL GeOXMPBLVO VXcavia
as6o Tx A6)C8vT V xaxoaxoyo noxcuv OL onototL ecxav npoOapm xpntlxi ntooN
yta -uv avoLXo~iToa rou Ikpot tie-ca' LrvauLccx) xomaca-poyp (219).174 Gae(Xe
L&~l-Uepoevtacppov av 4rav Buva-o6 va uaurctooutie rov Auoaava -ccv OpayLpatiO"&rv
re-ov AuoaavLa rcv eLypacpdav SGDI, ap. 1339, 1350, -uBpaXt -ou HneLp&YrLxo6
KoLvool75 Xal rOupOX~p&V0U 5aVVoCL7t va BLa-aouve
r x&noLa OuyyevLxL
a~rl ,usaib xO,177
To 6votia 'AnoXX68 pos, ouv uv -a-pXe >Lte ro 6votia Auo~avLaoa-Ea Gpypaytca ca
c T'III xa-uryoplaq, OuVaVca-aL L&lamhpa OUxVa rTXV apxaXa XX1VlXn ovotioaroxo-
yl78oC avrtOuovaVe -uv HnmLpcarLxxnpoo&ctoypapLa, 6OtoueLvCL 0aCvLo.179 H oXe>a
-&)v tO0 ovotlaEv reouai rouc Bev npesmL va eTv(L O>en nOru xaLOU,
eXOpa xaeac; Bev
eX?utle Bety~aTcaavaypap naUpOvrUtLXo o6 oppaytoyatiac xepatlzt&v.'80 Kca rov
OpXMv8o'8' Ta ov6tioma aurca eLvaL ra ov6tia-ua -&)v &pyOv&)v 4 r&)v epyoOGaOLapXCv
169 Pape XaLBenseler 1911, acto XBpja. IIpp. Fraser XaLMatthews 1987, acto X?pa? IHitacaq1961,
aCX.27, aoV. 3.
170 Hammond 1967, Index I, aCX.808 Cabanes 1976, aeX. 621. llpp. rpapsvv 1988-1989, aeX. 123,
aCV. 205.
171 EuayyeXtE8v1935, aCX.245 Hammond 1967, aeX. 564, 565.
172
Hammond 1967, aeX. 649, 650. llpp. Franke 1961, aeX. 145 XaL Cabanes 1976, aeX. 364, 456.
173 Franke 1961, aeX. 314. rLa to KoLv6 xtv HnELp(XCV3X.napan6tv) GEX.188, acv. 44.
174 Avtcvwou 1991, aeX. 131.
175 GUVasVttaL to 6vova tou
-tqv Kipxupa, otvv AnoXXv~xa XaL GTOAUpp&XLO BLOu Zt)VZ4vOU
GE
oGpayEtoaCtaxCpaVt&Bv XaLcc voV4oaxt. Franke 1961, oeX. 314.
176 Avtavwou 1991, oeX. 131.
177 O
XaL eyyovo6 Vvr46 tou AvtLv6ou, o onotoq avaippEtaL G
HIpp. tvC oXiGv nao=t mypaqi tOU
trXouq tou 4ou aL. (SGDI, ap. 1338) XaL tou AvtLv6ou, o onotoq awaqpC- ?CtlYPGCnay
O (SGDI,
ap. 1339) tOUXPOVOXOYEE-XL Y6PpCaTO170 (Hammond 1967, aeX. 648). rPvat6 E&XXOU EEaL 6-tL 0
X&PO+, YLOqtOU MaXat& (IoX~p. 20.3), O o0tnoEoPO5OvaCE tOy I~aXIVLVEOatLq nL XELPijeL XaXt& tCoU
llhpaia (A&xap% 1986|3, aeX. 22) eEXc vavveyyov6 Ve t-o ?Lo 6vova. Aixap 1968, aeX. 48-51.
178 Pape XaL Benseler 1911, oto XBijpp. IIpp. Fraser XXL Matthews 1987, oao XBgipa? Fraser XaL
Ronne 1957, aeX. 16, ap. 25. Hpp. t-qv eLypaqpfi 'Ent 'AnoXXo8Wpou GENUOocppcy~aptctaxepapt( sv
?XX7VLGCtLXCV Xp6vcavaco6-v Kipxupa. Aovt6.q 1966, aeX. 87, ttv. 83:a.
179 BX. cc cnLypap an6 -qv 'Apta. Hammond 1967, Index I, oeX. 798.
180 Franke 1961, oeX. 314.
181 aL Ent nXEVOOU
MVVNOVEO0UGLV S XEp&6OU to 6voya tou d&pXltrXtOvOq,
<<EnaVL6tEPOV EtlypaqCaE XaL
aUXV&Be totOU 4ycavou, Ev6S ,5XaL860 ... C
tiXoq TOtOU EpPYOCatGL&PXOU tCV>>.
T5 Xa-taGXEaq2
OpX&v80o1955, aeX. 115, 116, aCJi. 1.
TA E(PPArIEMATA
KEPAMIASNAIIO TO IEPOTHE AQAQNHE 205
Trv xepapttl&v.182 AvtO6eTa o FrankeI83 Oecpet 6TL np6xeutci yLa ra ov6Vaia &5o
apX6v&Tv, Ouv'O6ta yvc)aOd xcaL a6 Ta Hntep&yrX&i voptotatia-, o&ou avaypa&povtaL
Ta ov6cuarc 5oL oauvToioypacpt~erv &5o T
ovo~d&rucv rpLcv apX6vtcwvTauTx6Xpova.
To ?(&oyptv6tievo OXerxcx OuXv> XaL OG OGppypaylOtia4ca
Oavav-uVdtr xepctxABwv a=6 Trv
AXpavta.184 H &dtocn rou Franke evLOX~xarL a6 -o epeX6tevo Tc)VtiXPL oaiepa
6yv&a)orv OcppayLOt4&urGvTq B'II xa-uvyopta'. Ta yp itia-a TA nou O6CovtaL orunv
xopucp5To avrotuv ou paOveat mtOav6vupo 6tL Bev elvaL Ta apXLx xtptouLov6laioro,
acxxoaXaaXPLX T( Xn yvcaao6 6tL oraCoGpcyc.loc.ca xepcziacav
xa(jaS).'86 EwvaL
auvTeO(-tCCL r avacypac" rou stcavutlou &pXovxcz,ntou Oeutl(eL TL; tevonoTl(x
senBia,
"nXLVc; r v cavcxx6pcv,
OrLc ono(te ypacpovrcavTa ov6tiacca rcv tiovcxpXyv.
TLc TlO nOXXic pope OUVO G tOVt(aL tie TnV p6poeav &nt(87 Xcxapl np6Oeoa te
anX4 avaypacp rou x5pLou ov6iogro, cae yevLx5 -caa 188 H avacypap rou rkrXou
UUVavrxriL aXtrLX&
rnex at& ,189 6ncx asaq ppypLa an6 Tvv Kepxupa: 'AcruvoyiouvToq
Euvcan~ccvo(90-5ae GyppdyL.La 6 -qv OMulta: [&L]ti[eXrToo]'AXetiaXou.'9
Eu2ypcavo Xoma6vVe Ta napao"tv& o AucavL'ac Bev eLvaL o oeXvLxMc q ? ep-
YOa dc7L&PXT; T~cv XepaCL&o),?cAxoifva 8nr,6aco tp60&flo toua-a cypaycava-ca TFS r'II
xapaxrTp(tcZLat an"6rv L86rT-t& Tou (Loc
xoctxflOPocpa , GvppaycLavara
MO) a
Txa
F'III xa-yopaoc auvou6ecat e -co 6vo.ca ev6 a"XXounpocacnou, Tou 'AnoXxo8dapou,
L 6O.tc( ae ayypayava.La xepavx.L(&v
t vav evacVL 8&.L6ato np6acano. EneNt
7ou vnopecnt tn(
eLvaL8uva-c6v -o 6vo.c Tu poe7tvVou PpXov'a va cuvo8~e6arL Ve To 6voca Tou
xepav.oupyo6,'92 8ev untopou4t va anoxxetcouve rvv oatOv6rra o 'AnoXX68&pos va
vT'vat'a' pXovaX acc o xcaxauxeuzrj Tcavxepac.t(&v.
Ev8tacpe~pov e(vat -co 8taxoacVTtx6 04eIaTcv avTeOLX&a Xwou6IewVv opL46VTLCOV
ametpcv xoTca a=6 To avOe4Lo rr( Fii xoc a-yopLaoc, tou Be auvavTapsat GrL( 8t-
axoav~aeuS S f3?aav a-a av.elcacota Ve" a nyel.6vxcv xat xopupatcov xaXutrc pcov
r
VLXp6-epl a=68ooaj tou 61ixpov ae aXial Ve -a un6outa yp&.ija-a anav&i oe 6X1t-E B&ipXa
202 H
tTN E?% TqVLOTLX XcaL TV,N npc6In; Pipatx75; nepL68ou.BXaXono6Xou-OLxov6Vou1986, aoX. 257,
o"i. 35.
203 BX.napanivc oCX.197, o"y. 115.
204
Carapanos 1878, aeX. 111 (ap. 6), t(v. 61:11.
205 rPL -qtuopyp tou adya xczt -rqvEXX7vLvtLxx pEPCo8o Ronne 1957, aeX. 86, 87.
fX. FraserxoXL
206
rLat-qxpPa~rcavV
arVXPEt6v@V cXpovo'oyx6 XpLtcpLO,3X.napanov& aeX. 188, alli. 39.
TA E(PArIEMATAKEPAMIASIN
AHOTO IEPOTHEAOAONHE 207
alytic ti tOXup5mni6xX~t aULt5xepated r0ou e(vat XapaxP pta txf yta rov 3o act.,207
vc xara -cov 2o at.,)av xat Bev eyxa caxebt'cxat ev-cX(,2u8 ca'c6ao PAOVp86vano-
Vaxp~vr-cat xat av-txaEtcrca-cat ct6 -rv a=68ocnr rou atytia tie napd7cz
xvxe xepcztr.209
H napana"Wv& TcpaxT)pj Oeavnopouop vaCMoTrkEssXiL 4ta iVBLT)~ 6-tL np6xeVCcL yLCa
GYpayEGpa-Ca Ta Oot7a 7tsptL va orCOMO8r00o6v OGOVreCZL>XVLO avpircza ocrov 3o xaL
-o 2o aL., Ve 6Xeq 6Vc, -Cq UtYUXO"C&Mq MOU Ouvo~e6ouv -r XpovoX6Pyjaj -rcv MU-
6rCV
ypaczcw acppayL0aVCT()v V fc arvsopp rcv ypavvdc-ccv.210 H XPOVOXOYLXt
au-5 xoso
rCore sVOX6etacs'at ano -jv ntOav6-ca -rrs u
L Otansrou Auocavtauav ocpa-
yLOt&)cv Ve rov Auoavca tou tiap-pupdratL entypaypaL XaL XpovoXoytcZaL tpLvc=6
To 170.211
QoT6oo r tiop'p cwrv av cOAtv Oa-c ocpayloyatca xaOLTcv &No xacTlyop'OLv xaL
L&OCt{keparvj 1'II xonrcyoptaq (Ta Auaocvtcz),6'ou ano&tO-Tm ev-ceXc ypacztLX&,
OuvavTanLaouvi5Owqaco6 xaVoeaaTcou2ou at. xaL ie-ca. Ta avOe4tiLa XapaxncpCov-a
ao6 vIvotoL6T-a tie y-tcp6 4 xXcBE,212 Ve cp6Xxcz XOXL
XecT6sVLXOXC e7tELVIXI, Ta onola
xcaZxTIyouv OnCaxpa Touq OGe LOXuP ouoTp4cpy5. Ta p6XXc Be PXasa-Ovouv c=6 ivav
xevTpLx6itup'va, aczaXacoi6 Tov tqnyxj ptiO> Totuxopucapcou XoyX6o>Xjiou p6XXou.
EX-aOcpayaptOa4ca-q T'III xa-qyoptaq ('AnoXXo8pou Auoacva) -a &0o xat-
X6oTepa c6xxa cTou avtutvou npoq -a Be XLxac apLOT-p& LvaLyxoyoeL5,2L3 tie T6oO
LOXupe6, OuaTpo$eqtou ovowcouv tie oeLsOsL8e(e6XLxeq. Ev' -EaOTe)XeXj Tc)VTpLW)v
npoq -ca ea'vw cupXX)w<v Ltva I
un)va xaL xLo6vTcaL X
napa'XXj<a po xTO
-co cvTpLx6 p6XXo,
To xatjx)6nTepo c6xxo npoq -a Bet xOaLapLOT-pa eXeLOxe86v opLCOVTLa TnooeTTOal.
Me Pa&v-l OUYXpLTLX 5 av-mapa"Oe tie -l ipopy Tc)V vO i(&cwv XaZL TcA)VMxcxv,
O
tou xoatuOa v atlepuwn)a jyeti6vwv xaL xopucLawv xa)XutnTpcl'W cw 6noTV 'HteLpo,
B&Omta-oWe-VaTL
ooti6TXya ti6vo tie -cl Lax6aoijao LtxoOL iL ano-ctjpaOI&cG)v
veI-T6)7t
lyei,6v ov xaxuntcTpwv,214Ta oTo(a 7tpopXovTaCL c6 TTlOT-cyjTTl B&)poyplqOxlviq
rou 0&tcpou OT7CA\AJV1, nou xpovoXoyTa L Vti a ti6OcaTtou2ou aL.215
1. 'T+. 0,052 V. M5x. 0,073 ji. (PI. 73:a). Eial: 01, 8COtd'rLO V 1982.
cxvcxaxcpxq
l)X65 untpuOpos xutpLvcan6o(5YR 7/8). NIKOIMAXOY
EAar: 01, MB&POIIOsV cxvcaxczp41982.
4. 'T+. 0,047 V. M5x. 0,065 i.
NIKOIMAXOY
IIoX untpuOposx'rpwvcn6o (5YR 6/8).
2. 'Tr. 0,051 V. M5x. 0,055 ji. eOaf: 01, 8C4&ALO Vd avaaxaczq1982.
HIIX6cuntpuOpos x'rpLcva6s (5YR 6/8). N[IKO]IM[AXOY]
eAat: 01, OIAdtCLO 81 avaaxacq 1982.
5. 'Tr. 0,048 V. Mfx. 0,069 ji.
[NIKOIMAXO]Y IIHvX6untpuOpos xL'rpwvn6o (5YR 6/8).
3. 'T+. 0,052 V. M5x. 0,078 V. (Pl. 73:b).216 eOaf: 01, V
8COI&ALO cxvcxaxcpx5
1982.
llos6 untpuOposx'rpwvcn6o (5YR 6/8). NIKOIMAXOY
To BeXpvL eLvaL e'va tica LB~a-epa ayacs ac6 v apXata eXXTvLxq ceXvv.217
Ev6 xa-capX5v Xp1onLtmuOL5xewq oa6VPoXorl Oa&Xacaao,
yplyopa exaae auTo
-o Xapax-cpa xaL xacat rvlv EXXT)vLaO-xtneptopoxox epcl)xe cq au-co&vaLo B&-
axoOj-cLX6 Oeica. Ee apayta~a-ca xepacz&v OuvavTaaT(L58v
T ac'6 rov 6o aL. orcv
A~Lc218XOLr
Atytva~l c~zt&219
xa. -co Ka~aoBas8.2
To 6vo~ua NLx6tIaXoXOuvavc&TatLOuxV& Oa7cv apxada exxTVLx5 ovoracoxoy(a220
xaL 8ev eLvaL&cyvwclaoorvcv HnteLpcArLxnpoocrnoypa(a.22 Ta OTOLXLc tO6oXOu4e
OTE B&&Oeoa5 ,uaX Bev encLpe67ouv Va 8LeuxpLvcaoule Ve pepatc6l-a rv L86T-y-a rou
NLxo~ciXou. Ka-cz rov OpXavbo222 -o 6voua-cou apXLeTxTOovaf rou ipywvou excpepe-CaL
Oa yetVLX, 5 Oa ovoIaczTLX5 nrcwiao (oi6ce unovoetlU= 'co &76m), evCw rou eitwvuvou
apXovTIX Ouvok6eT-L c=6 rvlvnp6o'oea&it. E-ca ocpayLo -cca6jiwq-co rIII xa-yo-
plaq223ano8eXOTxe6-cLEouv? 6voiuaBnvI6ooua&vpaiou exypepe-caLoaeyevLX' ntcwao.
Evctaipfpoucza Oa ntav T =tOavorTxa oa6vkoaq rou NLxovuaxou crv OcpaypyLOucw)v V,
'o NLx6,uaxo nou ,uapTupektL Oa ateXeuOeppcmrLX6 Icou
cpYva e'Xouq rou 3ou xaL
rcwv CpxGv rou 2ou au. acoi6T-clAAcWvj224 XCLOa VOX6P&VO
6xscoavaaci6n vq AGc)8G)v
216
Adxczpj5 1982, aeX. 87, ntv. 62:ac.
217 BX. Stubbing1929,atopczBx& Rumpf1969,aeX.90 xe{., 97 xEg.,6tou XatLdpOovv PLPXLOypaEp(a
YLaT'L MpaEM s BEXpLVLO6 a-Tv apXata eXXT)VwX5
-rXv.
218
Welter 1938, a'r. 486, ELX. 9.
219 Felsch 1979, aCX. 26, 32, 33, EX. 5:2, 3 10:1-3, ntv. 2:7, 8 5:6.
220
Pape XZLBenseler 1911, a-co Xktiicx.lIpp. Fraser XcLMatthews 1987, a-co XB5iilcz.
221 Hammond 1967, Index I, aCX. 810. Hpp. -co 6voa aE appdyLa~occt6 'rvv AX~P(vt(a. Ceka 1982,
aCX. 112, ap. 58.
222
OpXcivbo 1955, aCX.115. llIpp. nopanvw ae). 204, acyi. 181.
223 BX. napaivca aCX. 205. xepapt8xcv t6 -rv pcapctx5 oLxca TOv LEP&OV(<<?aXvX>>)
Hpp. apaypalia-ra
a-o IEp6 -rou An&XcXvos McxeX&czx -rvq EntLB&apOU e -co 6voacx -rou GUyXXT)'xo Avmvtvou (205 CaL.
a.X.) ae yeVlXw tr6av. AaCpLvou86x%5 1990, aeX. 14, ELX. 16.
224 SGDI,ap. 1349.
Hpp. Adxaps 1969p, aeX. 35 Cabanes 1976, aeX. 472, 554, 589, ap. 75.
TA E(PArIEMATA KEPAMIASINAHO TO IEPO THE AnAANHE 209
lou 0e-wi
XpovoXoy-riat -o 219.225 oa,6oo I &ypOovI juxa itou PunXptLXet o tIrX6q
6Xcov-ccv ano-cuiiaccv -cxv xepavE8csv-c5 I"Iv xar-cyoptaq oouv&prcIaIv
Ce -cIv
nap&oceo7aGTou BeXpLVLo6,iou [G(o1 BL=TIpec rC ouv4poXcxft
o6v8eove ,s Ma
&Xacaoca
xaL -TL OOa&OeO 8paacO pL6treq,226 Oa ,unopo6oe va Vaq oBnyp5Oa -csv unt6Oeco
6-Tc np6xectcL yTaOxepajil8e otL onote P
PpoXpXovTatL c6 Iap&XLo xepacioupyeto xaL
enojuvcaq -co 6vo~ia NLx6iaXoq jiuopet va avxt5x OcTov ipycoVl) 5 OCTovepyoGcZaL&pXI
TTS xa-caaxeuv5 -couq.227
KaT& Tov avacoxacqa228ra yp&vLctta-cr.spv OyppyLOv&t(I)v-cIqIv xa-ctyoptaq
avfxouv a-cov 4o 5 OcTLqappxq -rou 3ou aL. fcoTc6oo, -o vL [VLeLao0uDxI xaL cap&XXvXa
-ca &5oox4Xv, -co &Xqxa
V opLC6v-cLa -cv eOacwcplX xepaca, -co 6VLxpov [LXp6Tcpo ci6
-a &Xxayp&v4LcZa XaL -co IL tie eXacpp& aVOqXTq XepacZe( aOVaVT6)vT(aLeicaou oUX>&
xaL apy6-cpa.229
Ta Oyppay(ii-c(Xa npoipXovLat a=6 -cv rctrpuya 01 tou ano-ceXel tpooOj5xI orvI
P6peta ieup& -couIIpu-caveou (0) -cvcq
I MA&Gc
vvs.230 poc ( caytxa
231o[udv
T xa(aaxeuOX Tou omLxoBoVu?a-coq 0 etvaL o6yXpovl
Ve -cou BouXeu-clptou (E2)23' OaL-
apX~q Tou 3ou aoa -ce XI -ou 4ou aL. Y-ca -e'XI -cou 3ou aL., xa-C& -clv neptobo
-cou Kotvo6 (234/3-168/7), XpoVoXoye aTLI vca 7nt-puya 01232 soU Xpa[Veue yta
TTlv ecOr(aca xaXL-l &avuxTcpeuOY Tc)v apX6v-cw)v -ou KoLvo6. Ta 0cpayt(vcaT
npofpXovTa(L c6 TY OTe'yI auT#q Tjq =Tpuyaq233 xaL [la XpovoX6yjaY') Touq OTov
6aOTpo 30 aL. Oa Ouv#intLe ,e -co Xp6vo xa-caaxeu'; -rou x-rnplou. H neptoboc auTc iXeL
oyX~o Ve -cI ,uey&XI a&vOvao -rou Ikpo6 xaL TTv ava&n-cu~I Tou KoLvo6 TcAwHntepcaTcwV,
TO onoto e[Xe -cv ibpa -rou a-co BouXeuTcjpLo TTjr,ABelqevv. TIv senXT(xOa ienpa)xav
V. EIII (DOPMISEKOT
Kaczt&Vnxoq -ccv opOoy6c vLv ivruncaiv OCPocyLOVdt(
I)V rv xocrvcyopczq (PIs.74-
ixcutnj l etLypocp'EIIIfOPMISEKOT. 'EXouv PpeO-ctr pax&pcca
76:a) yp?pe-coL
ocppyuavocCC:
12. Ap.M.: 4100 (P1.74:a, entV&). 22. Ap.M.: 5746 (P1.74:a,x6t& 8et&).
'TY. 0,02 V. Mfx. 0,11 V. 'T+. 0,02 V. Mfx. 0,04 V.
lIIXM unipuOpoqxvtptvcan6q (SYR 7/8). lIIXM p6btvo (7.5YR 7/4).
eiac: Eto& 'tou E2 avaacxacp 1966. eGan:E2- avacxayp 1969.
[E]l1l4'OPMIC[CKOY] [Enli0O]iMICC[KOY]
13. Ap.M.: 5728 (P1.74:a, xtco aptoatep&). 23. 'T+. 0,02 ti. M~x. 0,07 V.
'T+. 0,02 V. Mfx. 0,05 V. HIIX6 unipuOpoqxvtpLvwn6q (5YR 7/8).
HrjX6unipuOpoqxvtpwLvca6q(5YR 7/8). eias: E2- avaaxaypi1967.
E@af:E2 avaaxapyj1971. [Enl4OPM]ICCKOY
[Enfl]OPM[ICCKOY] 24. Ap.M.: 3695 (P1.76:a, x6t& aptatep&).
14. Ap.M.: 3698 (P1.75:b, entV& Bet6L). 'T+. 0,01 14.M5x. 0,05 14.
'T+. 0,02 V. Mfx. 0,09 pt. HrjX6p68voq (7.5YR 7/4).
HrjX6p6tvoq (7.5YR 7/4). eiaf: Dto& tou E2- avaaxap5 1968.
[Enfl]OPMI[CCKOY] [Enl4OP]MIC[CKOY]
15. 'Y+. 0,02 V. M~x. 0,15 V. (PI. 75:a,Be~t6). 25. Ap.M.: 4103.
HrjX6unipuOpoqxuLpmVn6 (7.5YR 7/6). 'T . 0,02 .t. Mfx. 0,08 .t.
eiaf: E2 avaaxap5 1967. HrjX6 p6btvo (5YR 7/4).
[En]io0PMICCK[OY]
eiaf: N6ota tou E2- avaaxap5 1966.
[EnlIOPMI]gCKOY
16. Ap.M.: 4104 (PI. 74:a, x6-rcaViov).
26. Ap.M.: 5732.
'TY. 0,02 V. Mfx. 0,05 V.
HrX6 p6btvo (7.5YR 7/4). 'TY. 0,02 .t. Mfx. 0,08 .t.
HrjX6ounipuOpoqxtcptvcaf6q (5YR 6/8).
Geia: E2 avaaxacq 1966.
eias: E2- avaaxayp 1969.
[EnlI]OPMIC[CKOY]
[EnliOPMI]C[C]KOY
17. Ap.M.: 5828.
27. 'T . 0,02 .t. Mfx. 0,06 V.
'T+. 0,02 V. M~x. 0,02 V.
Hrj6 p6Vobo (7.5YR 7/4).
HrX6 p6&voq (7.5YR 7/4). eiaf: E2- avaaxacp 1967.
Geac: E2 avaaxayp 1966. [Enl4OPMIC]CKOY
[EnlI]OP[MICCKOY]
28. Ap.M.: 5737 (P1.76:a, entV& Bet&).
18. Ap.M.: 5731 (P1.75:b, entV& apLatep&). 'TY. 0,02 V. Mfx. 0,05 V.
'T+. 0,02 V. Mfx. 0,09 14. Hrj6 unipuOpoq(2.5YR 6/8).
HrjX6p6&voq (7.5YR 7/4). eias: E2- avaaxacp 1968.
E@af:E2 avaaxacn 1969.
[EliOPMmICC]iKOY
[Enilvoipmic[ciK[oyi
29. 'Y+. 0,02 ti. M~x. 0,03 V.
19. 'T+. 0,02 V. Mfx. 0,12 V. HrjX6 unipuOpoqxt-cptwvca6q(5YR 7/8).
HrjX6 unipuOpoqxxtptvcan6q (7.5YR 7/6). eias: E2- avaaxayq 1967.
eiaf: E2c avaaxayq 1967. [Enl4OPMICCK]OY
[EnliO]PMICCK[OY]
30. 'TY. 0,02 .t. M~x. 0,08 ji.
20. 'Y+. 0,02 ti. M~x. 0,08 ti. HrX6 unipuOpoqxvtptvcan6q(5YR 6/8).
HIXMcunipuOpoqxvtptvcan6q (7.5YR 7/6). Avaaxacn 1967.
Geia: 01 avaaxayq 1981. [EnliOPMICCK]OY
[Efl O0P]MIC[C]KOY
31. Ap.M.: 5747 (P1.76:a,x6t& 8et&).
21. 'TY. 0,02 V. Mfx. 0,09 V. 'TY. 0,02 V. Mfx. 0,04 V.
HrjX6 p6Vobo (7.5YR 7/4). HrX6 p6tvoq (7.5YR 7/4).
E@i":E2- avaaxaqp5 1967. EM"o:E2- avaaxacpq1969.
[EnlI0OP]MICCKOY [EnlI0OPMICC]iKOY
212 AMALIAVLACHOPOULOU-OIKONOMOU
241
Hp3p. 'Ent KePX(cvo?caco6 tnv Kaaac'anI. Axprjc 1980y, acX. 31, 32, 2dv. 39:p, y. rta xnv
p6OcaIJ 2tE+ to 6vo.a nctcvu~tou 6xovXta pX. Op9X~voq1955, aeX. 115, aIV. 7 Franke 1961, aeX. 314
Xat ac &6pOova ayppaY~aytiataac6 tv AXp3av(a: Andrea 1976, aeX. 350, 2dv.XVIII:1 Tartari 1977-1978,
aeX. 217, ndv. I:p1-3 Ceka 1982, aeX. 110, ap. 38, 2dv. III:39 Hidri 1986, acX. 110, ndv. XIV:2 xa,
LBME-tcpaTartari 1988, aeX. 96-98, ndv. IV.
242'OEaX cuo to Happvxv spoXpExt to Hpieviaxoc. Fraser xat Ronne 1957, aeX. 167. rta -o
bop~.?(xvf. Fraserxat Ronne 1957, acX.142 (3)- Pape xat Benseler 1911, a-o X Vp. Fraserxat Matthews
1987, a-o X4qva. rta trjv 'HizLpO f3X.Hammond 1967, Index I, aeX. 816.
243 EuayyeXEB'j 1914, acX. 239. IIpp. Cabanes 1976, acX. 552, ap. 29 xat Hammond 1967, aeX. 651,
oait. 2 xat Index I, aeX.816.
244 SGDI, ap. 1359. IIpP. Carapanos 1878, aeX. 57, ndv. 30:1. rta to 8c6tcepo faata tc x ztaXoyo
OL0~t0OLEEXaV2tpoapELaYXPIfaOctX o& yYta nv otxoavomoUaV j tou IEPo6 vcC& trv av(aLXtx5
7tOXOtCX)V
xataaxpoqp5(219) PX-.napatv& aeX. 204, aIV. 174. Dto apa auc6 to 6vo.a 4 op.toaxoq avaypecat
X
86o (popi Xat ntOav6v tp6xEctat Yta tantO6 Xat yyov6.
245 Franke 1961, aeX. 156, 174, ndv.24:V 85
(o00&aII, aELp&20).
246 Axxaprlq1966, acX. 82, ndv.82:p.
247 .tov 4o at. auvavtc6ZcLantopabcLX. laffenbach 1957, acX. 43.
248 rta oc. 206,
-cI jLopyp5 tou nt, tou ?Ltxpov Xat Tou lit xa-cx -cvv EXX1vtvLtXy nepto8o PX. napantv&
aIV. 201, 202, acX. 209, aIV. 229.
249 rPato
vjVOELcuqaoyva px. napantv& aex. 199, aIV. 128. Hpp. Franke 1961, ae). 142.
250 EuayyeXE~lq 1914, aeX.234, 62ou xat aXEtX& napaBcEypa.ta.
Ilpp. Hammond 1967, aeX.735.
251 xXaprj 1966, acX. 82.
252
rta tjv avotxomo86tial ou Iepo6tect& trV atzcaX)`5 xa'caapoqP5 f3. napantv& aeX. 188, aIV. 46.
TA E4PArIEMATA KEPAMIASINAHO TO IEPO THE AnAnNHE 213
CL)V
IOXpa4&wv alla v cst6 u Acavrv. As OsVleGOeEt 6cL 2tp6xeoLyL a r liovaBLx&
GpcZ ppyaylOIarC)ov uto6 y Acadv,
xaOrxyopTa r onota le PpmpcO-ca npOipXe-aL o6-rv
o-riy& rou BouXeu-nptou (E2).25
BIBAIOrPANA
AgoraV = H. S. Robinson, Potteyof theRomanPeriod:Chronology (Atienian
AgoraV), Princeton 1959
AgoraXIV = H. A. Thompson xat R. E. Wycherley,TheAgoraoftAdtns(AtenianAgoraXV), Princeton 1972
Albanien= Albanien.SchatzeausdemLandderSkipetaren, Mainz 1988
Andrea, Z. 1976. "Germimetarkeologjikete viteve 1974-75," Ilina 6, aeX. 331-366
Avtcvlou, A. 1991. A&8&vr. Eu4poX5 HneLP&yrdVaxrv avotxoU6Vaj x-rt~aV6rcav tou Iepo6 trj
Acabdvrj (Ve.t6 to 219 t.X.), AO5vat
Appavut6touXoq,A. 1928. At ypantaE atx5Xat A~Vtp6p8oq HIayasadv,AO5vat
Bereti,V 1977-1978. "Germimene Triport,"Ilina 7-8, aeX. 285-288
Boisacq, E. 1950. Dictionnaire6ymologique dela languegrecque,
Heidelberg
Brommer,R 1988. "RedendeZeichen,"AA [JdI 103], aeX. 69-70
Buschor,E. 1933. Die TondclherderAkropolis, H, Stirnziegel,
Berlin/Leipzig
Cabanes, P. 1976. L' tpire dela mortdePyrrhos &la conquete
romaine (272-1 67 av.J.C.), Paris
1988. "Lesconcours des NAIA de Dodone," Nikephoros 1, aeX. 49-84
Carapanos,C. 1878. Dodoneetsesruines,Paris
Ceka, N. 1982. "Vulaantikembi tjegullane treven ndermjetAosit dhe Genusit," fliria12:1, aEX. 103-130
Xpuaoat6.tou, II. 1980. vIB' Eqpope(a HpotatopLx(V XaL KXaGLX6v ApXaLotrt((wv>>, ALEXt 35, 1980, B'1
[1988], aeX. 307-311
. 1982. <<IB'EcpopetaIIpotatopLxGXa v KXaGLX6v ApXaLto4i(.w>>, ALkX 37, 1982, B'2 [1989],
aeX. 260-263
Cook, A. 1965. Zeus.A StudyinAncientReligionII, New York
A~xapgq, E. 1952. <<Avaaxaccpj E; KaatpEtaaIv avvv>>v, IIpaxtLx& 1952 [1955], aeX. 362-386
. 1960. <To Ikp6v -qq Aca8davv, ALkX 16, 1960 [1962], aeX. 4-40
1963a. <<ApXaL6tTr-eqXaL
xativvtia Hnerpou>>,AeXt 18, 1963, B'2 [1965], aex. 149-157
.1963P. VHHEIPOE. Nexuotavreltov 'Eppaqo>, 'Epyov 1963 [1964], aeX. 57-64
1964a. OLyEVEaXoyLxoE V00oL TCOV MoXoaaGv, ev AOjVaLq
.1964P. vHllEIPOE. Nexuo.tavrelov 'Eqpipaq>>, 'Epyov 1964 [1965], aeX. 51-64
1964y. <ApXaL6tTteq XaL Vvvtida Hne{pou>>, ALkX 19, 1964, B'3 [1967], aEX.305-314
. 1965. <<Avaaxacpij tou Ikpo6tq5 Aca8dnvs, IIpaxtLx&1965 [1967], aeX. 53-65
. 1966. <<Avaaxacpqj -ou Ikpou -c6 Ac6vvq, IIpaxtLx6 1966 [1968], aeX. 71-84
. 1967. <<Avaaxacpqj CouIkpou AtOWvvq>>,
qq IIpaXtLX&1967 [1969], aeX. 33-54
. 1967-1968. vE7vupLt44Loq aG-jnX(j nEpt 8puo0Xatp Eoa)0, XapLGtajpLOV Et Av. K. OpXkvbov,
t6pi. A', AOijVaL, aeX. 386-405
. 1968. <<Avaaxacpqj tou Ikpou tq Aa8dvvo, IlpaxcxLx& 1968 [1970], aeX. 42-59
. 1969a. OAO6vv>>, 'Epyov 1969 [1970], aeX. 19-28
. 1969P. <<Avaaxacpqj tou Ikpo6 tvq Aa8dvvqo, Hpaxczxi 1969 [1971], aeX. 26-35
. 1971. <<Avaaxacpqj tou Ikpo6 tvq A&06vvq, HpaxcxLx&1971 [1973], aeX. 124-129
. 1972a. <<Avaaxaccpj tou Ikpo6 tvq A&06vvq, IIpaxti~x&1972 [1974], aeX. 94-98
. 1972P. <<E3eanpcra>>, ApXa(eq eXX-vLxiq76XELq15, AOijVaL
1973. <<Avaaxacpqj tou Ikpo6 tvq AO 'vvo, IlpaxcxLx& 1973 [1975], aeX. 87-98
. 1974. <<Avaaxaccpj tou Ikpo6 tvq Aa8dvvo, IIpaxtLx& 1974 [1976], aeX. 73-78
. 1976. <<AXipcv-Eqptpa>>, 'Epyov 1976 [1977], aeX. 80-88
. 1980a. VOLapXi voqVLGatoxoatUXTEatv V apXata 'HtELpo>>,HtELpCLtX&XpovLx& 22,
aeX. 21-26
253 rLa tUL
9paGLq tou
OLXOBOVILXiq E2 XaLt aLqsVtUtoLX~ atUyEq Tou PX.napansvc asX. 189, ani. 49.
214 AMALIAVLACHOPOULOU-OIKONOMOU
AMALIAVLACHOPOULOU-OIKONOMOU
OFCLAssIcALARCHAEOLOGY
DEPARTmENT
Universityof Ioannina
Ioannina, Greece
PLATE 60
All"~~~~
cci6 trnvK~atrrpza~Icxvvsvcs
a. Ecpp&TLO14a
A ~~~~~~~,1~
r ~~~~~~~~~~e
b. A'I, 1
A. VLACHOPOULOU-OIKONOMOU:
Tcx Ecppaytwaxa~ Kcpcxallt8v ciio 'o Icp6 'rvg Ac846vvs
PLATE61
1 4' / ~tg
a. AI 2
. .LI z .
~A lZ.. ..a.
.__
ns-Hiv^vt * a. ts'
AA'I,~.2
W.
a.~~~~~~~~~~~~~~~~~~b
4
b. A'IT
PLATE62
a.A'I, 3
PA. p !. r
b. B'I, 1
2
a. BWI, (cit&Vc XapLatcp&) BI, 4 (cu&Vca 3 (x6xm apLap&i) BWI,11
B~CL&)WBI,
(X&tCa BCViL)
loot
Ai- ~ ~ ~ ~ ~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ i
4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'0
._~~~~~~~~~~~~~~~~~~~~1
kk
I L.
b. B'll, 2
7---=
?gad
U Y% >?, 9 -.< ., ,,_ *:.__
"4* '~-Q''-
-A~~~~~~~~~~~~~~~~~~~~~~~~~i:
* 4~~N
ar
%> r jri-.i
4"7 -
..j. .
a.11BlIII0
A. VLACHOPOULOU
OIKONOMOU: ro Iep6 'rracAxcBv)cr
Tcz ISpx'aptwr~a Kepczptt8owclvat6
PLATE67
a. B'M, 13
4X;44
Ao&$irjc
A. VrAcHoPou1LuOuoNCoNMou:Ta Scppaytoapsva KcpaiA3ov cc6 so Icp6 N'Nc
PLATE68
.~~~~~~~~~~~~~~~~~~~~~d l{),rt
.,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~M - -,
9,~ ~ ~~~F t
4.1
A
.~v~t
>4r ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~'
k>k
_ - -+,
b. B'Il, 20 (cxitvc~apLatcp&) B'Il, 30 (cxitvc Bc{L&) B'll, 37 (xixmapatc?p&) B'Il, 36 (xitca 5c{L&)
i A
m~~~~~~~~~~~i
35 (aplCYxCp&)B'III, 25 (SsiL&)
b. BWIII,
.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~F
. ~~~~~~~~~~~~~~~~~~~~~o
' 1~~~~~~~~~~~~~~~~~~~~~~~~~~~1
b. 1711,
3
4r' 'I
; * - . . .~N.
4t.
b.r'II,5
Tct EcppcyayitesctKipctli(&xvcuit To ksp6 Ur A~d6vrv
A. VLACHOPOULOU-OIKONOMOU:
PLATE 72
/r/t.; <~~~~~~~~'794
?S*-4r'>
''S~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4
I' IA~~~~~~~~~~~~~~~~~~1 ~
Al
~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~A
a. P'IV, 1
~~~~~~~ 4~~~~~~~~~~~A
PLATE 74
_0s~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
a. rev,12 (enaVw) rFv, 13 (XaCnd apaGCrEp) rFv,16 (x-cw viaov) r'V, 22 (xdm
Ar.
r'v 8 (t~
b.w r'v1 x;
Tx EcppaytoanaxKrpaotco?~v
A. VLACHOPOULOU-OIKONOMOU: ct6 no krp6 arjs Acob&ovrjc
PLATE 75
'I
. _ r~~~~~~~~~~~~~~~~~~W
a. rFv, 11 (etVcJ apLa-tep&) rFV,28 (en6VJ betL6)- rFV,24 (x6-m aptatep&) rFV,31 (x6xm
.~ iM ~ ~ -L A
7 w _~ - o k ~~._
Jx~~~~~~~~~vj
A. VLACHOPOULOU-OIKONOMOU:
Tz SEppczy~oitrcz
Kcpczpt(&v acut 'ro Icp6 rvdAbv
ARCHITECTURALTERRACOTTASFROM THE
TOWNS OF KAONIA:ANTIGONEAAND BUTHROT
(PLrAEs77, 78)
ACCORDING TO literary sources, Kaonia was situated along the coast in southern
1~ Albania, from the Akrokeraunmountains in the north to the Thyamis river in the
south. Its interior stretchedthrough the valley of the Drinos river to the Aoos river,which
separates Illyria from Epeiros.' Historical and archaeologicalsources reveal an intensely
populated area during antiquitywith great economic and cultural prosperityreflected in
its early and intensiveurbanization.2
Buthrot, Antigonea, and Foenike rank among the most important urban centers.
Systematic, long-term excavations in these towns have shed light on different aspects of
the cities' rich cultural life. Archaeologistshave uncovered monumental architecture as
well as local and imported works of art that reflect high cultural, economic, and artistic
activity.3
Among the decorativefindssome architecturalterracottas,treatedfor the firsttime in this
paper,attractattention. Twenty fragmentsof antefixesare examined, giving (1)classification
by characteristicfeatures,(2) identificationof techniques and prototypes, (3) descriptionby
center of origin and motifs, and (4) approximatedating, based on the type of decoration
and comparisonto relatedfinds.
In Antigonea, although tiles were excavated in great numbers, antefixes are very rare;
all in all only six pieces have been found, of which only two are fairly complete and well
preserved,the othersbeing fragmentary.All are made of red clay mixed with lime granules.
Their compositionand color are the same as those tilesfound in otherbuildings,an indication
not only of the existence of local workshopsfor producing tiles but also of local production
of antefixes.
According to decorative motifs, the antefixes of Antigonea fall into two groups. The
firstvariantof Group 1 includesantefixeswith a palmettein reliefon the front (P1.77:a). The
palmettehas as its heart a bud fromwhich ten petalsradiate,five on each side. The petalsare
arched at their tips and end in a spiral. Below the palmette are two symmetricalspirals,
almost horizontal,accompaniedby other leaves.
The second variant,which is ratherfragmentary,has the same motif: the palmette,with
its heart in the form of a rhombus from which petals branch out on both sides (PI. 77:b).
There are also lateral floral motifs in the form of a spiral. Both of these architectural
terracottasof Group 1 were found on the main street of the agora of the town, which is
well dated to the beginning of the 3rd centuryB.C.4
1 Budina 1969.
2 Prendi and Budina 1970, pp. 67-87.
3 Ugolini 1932; Ugolini 1942; Budina 1972.
4 Budina and Stamati 1989, pp. 135-140.
218 DHIMOSTENBUDINA
The second group of antefixes is composed of four fragments, which are decorated
with floral motifs combined with spirals(P1.77:c-f). These antefixesof Group 2 bear some
resemblance to those found in Buthrot. They have been found in strata belonging to the
3rd centuryB.C.
Dating ofthe antefixesfromAntigoneadoes not constitutea problem,for they come from
a well-dated stratumbelonging to the 3rd to 2nd centuryB.C. Even the name of Antigonea
itself is helpful here, for it suggestsa foundationby Pyrrhusbetween the years 297 and 280
B.C.5 Because of the town'sviolent end near the middle of the 2nd centuryB.C., as a resultof
the war with Rome, the archaeologicaldata do not go beyond this time. The town was
destroyedand there was no later occupation. The antefixesthus belong to the period before
the destruction of the city. Their style and decoration can be compared to the antefixes
found in Foenike,Buthrot,Apollonia, and other Illyriantowns of the Hellenisticperiod.
The architecturalterracottasof Buthrot, in particularthose discoveredin recent years
in differentparts of the town, include fifteen fragmentsof antefixes that are importantfor
their decoration. These objectswere found in mixed strata, and archaeologicaldata is not
availablefor their dating. All antefixesare made of pure clay, yellow-ocher or red or pink
in color. There are several variants, based on decoration. Their common feature is the
type of decoration, a palmette and spirals, similar to the antefixes of Antigonea. By the
stylistictreatmentof these decorativeelements,we can identifyother variantswith human or
animal heads combined with spiralsor floralmotifs.
The first variant has as its decoration a palmette in relief opening in both directions,
radiatingfrom a heart in the form of a rhombus or circle (PI. 78:d). Below the palmette
there are two or more antitheticalspirals. In contrast to the most common style, one of
the antefixes has a palmette in deeper relief (P1.78:). On the front surface is a slip of red
color, characteristicof the Hellenisticperiod.
In the second variant the antefixeshave as decoration a rosette combined with spirals.
The better-preservedfragment(PI.78:a)presentsa rosettewith lines followingthe contours
of the petals. In the center is a circularbud from which nine petals radiate. Below is a
pair of volutes ending in spirals. There are traces of red on the front surface, and the clay
is pure ocher. Another antefix of this variant (PI. 78:e) has the same decoration, but the
color of its clay is different,being darkgray,probablybecause it was overfiredor, more likely,
was a remainderin the local workshop.
The third variant (PI. 78:b, c) of the Buthrot group is composed of antefixes that have
as the main decoration a palmette emerging from a bud. At the center of the palmette
on one of these examples is the head of a bearded man (P1.78:c), probably Zeus. Both
of these examples are made of pure clay that is red to light pink in color. Judging from
the compositionand color of the clay,the antefixesappearto be the same as other tiles found
in abundance in the excavations. In this group we must include five other fragments of
antefixes (PI. 78:g-i), all made of pure clay, yellow-ocher in color. The main design is the
spiralcombinedwith floralmotifs. On two other fragmentswe can see, besides the S-shaped
spiral, a woman's head (PI. 78:i), very damaged but similar in style to an example found
I
Budina 1984, p. 159.
TERRACOTTASFROMTHE TOWNSOF KAONIA
ARCHITECTURAL 219
BIBLIOGRAPHY
Budina, D. 1969. "Disa rezultatete giurmimevene vendbanimete lashta ilire te Kaonise," in Konferenca e dytfe
studimeve II, Tirana
Albanologiike
* 1972. "Antigonee,"Iliria2, pp. 269-348
* 1984. "Le lieu et le r6le d'Antigonea dans la vallke du Drino," in I'Ilyie mrbidionak et I'tpiredans
P. Cabanes, ed., Clermont/Ferrand,pp. 159-166
I'antiquitl,
Butint, Buthrot/Tirana
. 1989. Vorrdmisches
Budina, D., and F. Stamati. 1989. "Disa objektebronxi nga Antigonea,"Monumentet 2, pp. 135-140
Kaltsas,N. E. 1988. rH5XLVE 8laxX durV~eq XEpat6OVls c6 =6 MCy MaxeM6VML, Athens
Prendi, E, and D. Budina. 1970. "La civilisationillyriennede la vall~e du Drino," StudioAlbanica2, pp. 67-87
Ugolini, L. M. 1932. "L'Acropolidi Fenice,"Albanicaantica2
. 1942. "L'Acropolidi Butrino,"Albanicaantica3
Zeqo, M. 1986. "D'ancienstemoignagesde l'art ADurres, Iria 1986:1, p. 185
DHIMOSTENBUDINA
QENDRA E KERKIMOVE ARKEOLOGJIKE
Tirana
Albania
6 Ugolini 1942, p. 192.
7 Virgil 1.55; Aristotle,Hist. Anim.8; Athenaios 9.18, etc.
8 Ugolini 1932.
9 Muzeum Nacional inv. no. 76.
10 ArcheologicalMuseum inv. no. 573.
11 Kaltsas 1988, p. 79.
12 Kaltsas 1988, p. 79; Zeqo 1986:1, p. 185.
13 Budina 1989, p. 21, with earlierbibliography.
PLATE 77
--dO ~ ~ ~ ~ -
a l
'Ad
~~~~~~~~~~~-
.c
DHIMOSTE
BuDINA:
ARCHITETRLC TmuAcorAs FROM nmTow~s OF KAONI
PLATE 78
AA1.
A' .
Ao?
_
LA~~~~~~~~~~~~~I
~. , r; -
Pc
(?LA~ Of e
a~~~~~Atfxsfo uho
~~~~~~~~~~~-oo
NI
DHLMSTENBL1DNA:ARH1;TRLTRAOASFOTETWNOFK
d~~~~~~~~~~d
TRAUFZIEGELMIT RELIEFMAANDER
AUS DEM SCHWARZMEERGEBIET
(LATE 79)
I
Der Beitrag steilt die tberarbeitete Fassung eines Kapitels meiner ungedruckten Habilitationsschrift
dar: Zimmermann 1983, Kapitel VI/5. Teilergebnissewaren bereits im Oktober 1978 auf dem Eirene-
Kongressin Nesebar,Bulgarienvorgetragenworden: "HellenistischeArchitekturterrakotten von der westlichen
SchwarzmeerkUste."Alle Fotosund Vorlagenfur die Zeichnungenstammenvom Autor;die Reinzeichnungen
fertigtenH.-J. Roloffund A. Rehbein, beide Rostock.
2 Vgl. Akerstrom1966; Zeest 1966; Ognenova-Marinova1980.
3 Zu Histriavgl. zuletzt zusammenfassend:Alexandrescuund Schuller 1990 (mit Gesamtbibliographie).
4 Je 1 Stack abgebildetbei Parvan 1916, S. 703-704, Nr. 61:b, mit Abb.; Canarache 1957, S. 171, Nr. 457,
mit Abb.; 5 weitere Stacke in HistriaV, Taf. 11:88-92; 2 weitere bei Zimmermann in Alexandrescu und
Schuller 1990, S. 169, Abb. 8, S. 177, Abb. 23.
222 KONRADZIMMERMANN
und VerzierungcharakterisiertenDachterrakottentypverbinden. Dem steht allerdingser-
schwerend gegentiber,daB nicht einer dieser Hegemones komplett erhalten ist.
Alle aufgeffihrtenFragmente(1-3 1)vertretendurchdie gleiche allgemeineZurichtung-
es handelt sich um Flachziegel mit herabhaingenderFront und Stegen uber den Seiten, die
ein wenig auf die Front umbiegen-und den von eingestreutenFeldern aufgelockertenRe-
liefmaanderdenselben Typ. Trotzdem scheitert die wuinschenswerteweitere typologische
GruppierungdiesesMaterialsdaran,daBsich die Stuckein vielen Punktenunterscheiden,sei
es in Ton, Uberzug und Farbgebung,sei es in der unterschiedlichenFronthoheund anderen
Abmessungen,sei es in Ornamentdetails.Gehoren sie auch alle zu demselbenTyp, a1Bt sich
dennoch bei keinem einzigen Fragmenteine derartigeUbereinstimmungmit einem anderen
feststellen,daBeine Gruppenbildungvoll gerechtfertigtware.5 Deshalb folgt die Anordnung
der Fragmenteim Katalog ihrem Funddatumbzw. anderweitigenBekanntwerden.
Der in der Regel matt ziegelrote, bisweilen nach Gelb (7, 17, 27), Ocker (15, 16)
und/oder Braun (2, 5, 14, 17, 21) neigende, meist verschmutzteTon ist in Abhangigkeit
von der Menge, GroBe und Verteilung der Zuschlage unterschiedlichdicht bzw. poros.
Neben Glimmerkommenweif3liche,schwarzlicheund rotlicheMagerungsteilchen-letztere
auch in groberer Form und vereinzelt als Ziegelsplitteran der Oberflache bloBliegend
(10, 14)-in unterschiedlichenMischungsverhaltnissenvor. Auch finden sich gelegentlich
im Ton Schnecken oder Muscheln (4[?]) bzw. Abdrucke von letzteren (17). Ist das
Ausgangsmaterialsandig,so wirktnoch das fertigeProduktweich und leidet unter erhohtem
Abrieb, andere Stucke sind dagegen sehr hart gebrannt (5, 27). Zwei Fragmente (6, 18)
fallen durch ihre Tonbeschaffenheitaus dem beschriebenen Rahmen: Sie haben zwar
alle die ablichen Zuschlage, dabei aber eine rotlich-violetteGrundfarbemit vielen, fast
zitronengelben Einsprenkelungen-darin ahnelt der Ton demjenigen der Amphoren aus
Sinope;6 es wird sich also bei ihnen um Import von dort handeln. Sie verfugen auch
fubereinen ganz charakteristischen,urspriunglich kraftiggelben, ins Griunlichetendierenden
Uberzug, der alle OberflachenerfaBt,aberunterschiedlicherhaltenblieb.7 Auch an anderen
Stacken finden sich einheitliche Uberzuige,teils in der Art einer gelblichen Engobe (4[?],
7, 14, 16, 23[?]), teils als braun- bis violettroterFarbauftrag(8, 15[?], 19, 27). Gerade
von letzterem sind nur geringe Spuren erhalten (10, 12, 21, 31), so daB sich schwer
entscheiden laft, ob einst immer das ganze Stuck damit uberzogen war. Es kommt
hinzu, daB eine Reihe von Stucken stark versintert ist (3, 5[?], 20, 24, 26), wodurch
ein Uberzug moglicherweiseverdecktwird. Trotzdem mochte man annehmen, daB dieser
Dachterrakottentypursprunglichgenerell farbigeUberzuigebesessen hat.
Auch bei der insgesamteinheitlichenZurichtung,die sich aus zwei groBerenFragmenten
(15, 25) noch am besten erschlieBenlift, begegnet man laufend kleinerenAbweichungen.
5 So sprichtz.B. die unterschiedlicheFronthohedagegen, alle Stacke mit Buchstabenzusammenzufassen,
zumal sich diese auch in ihrerAnbringungunterscheiden;moglicherweisebilden 1 und 19 eine Gruppe, da die
Hohe von 6,0 cm und der Sitz des Buchstabensauf einer leicht angehobenen Platte tibereinstimmen-doch
laBt sich eine solche Vermutung nicht anderweitigerhgrten, solange das von Parvan veroffentlichte Stuck
verschollenbleibt.
6 Vgl. Coja in Histria V, S. 53 zu Nr. 91/2.
7 Dieser lJberzugwird auch bei 14 verwendet,doch unterscheidetsich der Ton von den Sttickenvermuteter
Provenienzaus Sinope.
TRAUFZIEGEL MIT RELIEFMAANDERAUS DEM SCHWARZMEERGEBIET 223
Zu dieser gehort offenbar auch, daB die gelegentlich (6, 26) rechtwinkligzur Front
abbiegenden Seitenflachensonst nach hinten leicht zuruckgenommen,d.h. einwartsgestellt
werden (3, 7, 15, 20), um so ein besseres Anpassen der Traufziegel untereinander zu
erreichen. In einem bestimmten Abstand zur Front wird bei einem Fragment (7) die
Seitenflache sogar noch zusatzlich zuruckgesetzt. Ein anderes Beispiel (15) zeigt noch
zwei weitere Vorrichtungen,die offenbar der Gefahr des Verrutschensvorbeugen sollen:
Einmal springt in Hohe des unteren Maandersteges ein zapfenartiger Wulst uber die
linke Seitenflachehinaus und muifte theoretischam Nachbarstuckauf eine entsprechende
EinkerbunggestoBen sein; zum anderen tritt-in ziemlicher Entfernungvon der Front-
vor die schrag nach hinten zuruckweichendeSeitenflacheein zwar schmaler,hier aber die
gesamte Ziegelhohe einnehmener Steg hervor, dessen Stirn genau im rechten Winkel zur
Frontliegt und folglichdas Pendantam ebenso konstruiertenNachbarstiuckflachigberuhren
soll. Ein ahnliches, mit AnathyrosevergleichbarespunktuellesAneinanderstoBenist auch
andernortsbei Dachziegeln beobachtetworden.12
Nattirlichwerden auch diese am Dachrand liegenden Traufziegelangenagelt gewesen
sein, doch reicht die maximal erhalteneTiefe (25 = 15,9 cm) offenbarbei keinem Fragment
bis in diesen Bereich. Nach Aussage des breitestenFragments(15 = 32,2 cm) mit nur einer
erhaltenenEcke muBdie FrontbreitediesesTraufziegelsmindestens40 cm gemessen haben.
In Olbia, wo auch keine komplettenTraufziegeldieses Typs aufgetauchtsind, hat man eine
Breitevon ca. 55 cm erschlossen.13
Die Fronten aller Fragmenteziert als einheitlicherSchmuck ein Reliefmaander,der-
wie schon beschrieben-aus zwei einzelnen, sich kreuzendenMaanderbandernmit dadurch
regelmaBigwiederkehrendemSwastika-Ornamentund zwischendiesen eingestreutenrecht-
eckigen Feldernbesteht. Die Reliefstegesind durchschnittlich0,5-0,7 cm breit und im all-
gemeinen 0,3-0,4 cm hoch, zeigten ursprtinglicheinen rechteckigenQuerschnitt-einmal
kommt Trapezform vor (25)-mit scharfen Kanten und sind heute meist abgeschliffen
gerundet. Gewohnlich liegt die Grundflacheder gesamten Front auf einem einheitlichen
Niveau. Nur an einem Eckstuck(20) ist die Vorderseitedes oben auf die Front umbiegen-
den Steges auf die Hohe der Maanderstege gehoben, so daB das Ornament an dieser
Stelle weniger aufgesetztals ausgeschnittenwirkt. Die Ansichtsseitenjener die Frontecken
uberragendenStege sind entwedermit einem Eierstabin schwachemRelief dekoriert(6, 7,
15, 26), der seinerseitsnach oben sowie an der zur Mitte hin gewandten Seite gerahmtist,14
oder mit einem spiralartigaufgerolltenPflanzenoranmentgeschmuckt(3 1), wie es auch an
Beispielen aus Olbia vorkommt. Alle diese Reliefornamentemussen aus der Form gewon-
nen sein, was auch daraus ersichtlichwird, daB bisweilen Maanderstege und Oberkante
der Front nicht parallel liegen (20, 21) oder Stege beim Abnehmen der Front verrutscht
(z.B. 15) bzw. anderweitigdeformiertsind (z.B. 8). Zwischen den Maanderstegenkann-
offenbarauf Grund schlechterAusformung-sogar Ton stehenbleiben(2, 3).15 Wegen des
12
Vgl. Buschor 1929, S. 30; Hubner 1975, S. 124 zu Taf. 66:5, 6.
13 Vgl. Bra~inskij,in Zeest 1966, S. 40 mit Anm. 27.
14 Am besten an 7 zu erkennen;bei 6 bzw. 15 scheint wenigstensnoch der Ansatz der seitlichenRahmung
erhaltenzu sein, wahrend 26 in allen Randbereichenzu starkzerstortist.
15 Da es sich bei beiden Ecksttckenum ein kleinesFeld an gleicherStelle handelt, konnte man versuchtsein,
an eine Herstellung aus derselben Form zu denken; dem steht gegenuber, daB dieser Bereich bei 2 bis auf
die Hohe der Maanderstegegeflilltist, bei 3 jedoch nicht.
AUS DEM SCHWARZMEERGEBIET
MIT RELIEFMAANDER
TRAUFZIEGEL 225
Platte.21Die auf die Front umbiegendenStege daruibersind entwederglatt oder tragen das
auch in Histria vorhandene pflanzliche Ornament. Die olbianischen Stucke zeigen, daB
die Reliefbuchstabenimmer nur mit ihresgleichen, niemals mit anderen Fillungen kom-
biniertwerden und daBdie aufPlatten gesetztenLetternbisweilenspiegelbildlicherscheinen
konnen. In Histria scheint dies bisher nicht der Fall zu sein: Denn die Lettern unsym-
metrischerForm wie E (4, 25) und K (28) zeigen die uiblicherechtslaufigeLeserichtung;
und stehen die anderen vorkommenden,ganz symmetrischgebildeten Buchstaben A, A,
H, 0, T allein, ist dies nicht mit Sicherheitzu entscheiden.
Das in Mesambria (Nesebar)gefundene diesbeziiglicheMaterial22ist erheblich reicher
und umfaBtsowohl Traufziegel(unterdenen wie in Olbia auch wieder solche mit ionischem
Kymation und Astragal an der Frontflachevorkommen23)als auch Simen, die beide den
Reliefmaandermit eingestreutenFeldernals Schmuckbenutzen. In den Rechteckentauchen
entwedervierteiligeRosetten (Traufziegel,Simen),24ein einzelnerknopfartigerPunktin der
Mitte des Feldes (Simen)25oder aber Buchstaben(Traufziegel,Simen)26auf. Diese sitzen
hier zwar niemals auf angehobenen Platten, werden dafuiraber-wie in Histria-sowohl
einzeln (Simen nur so) als auch in zwei Letternnebeneinanderangeordnet. Hinsichtlichder
Kombinationsfahigkeitder einzelnen Fullungen bieten die Stucke aus Mesambria ebenso
wenig Neues wie ein in der Nahe von Burgas am See Mandren gefundenes EinzelstUck,27
das auch einen Buchstabenaufweisensoil.
Wahrend der Sinn dieser sicher nicht zufalligen Buchstaben bzw. Buchstabenfol-
gen bis vor kurzem noch unklar war, hochstens apotropaischerCharakter zugestanden
wurde,28hat L. Ognenova-Marinovaunlangsteine Deutung dieser ratselhaftenInschriften
vorgeschlagen.29 Danach konnen durch Positionsbestimmungder Fragmente und durch
solche mit mehreren Buchstaben (fur die sie als umfangreichstes Beispiel ein unpub-
liziertes Stuck aus Odessos (Varna)30mit vier Buchstabenfeldern(A]N-E)E-ET-HP-[..)
heranzieht)auch anderweitig, namlich auf Flachziegeln, Amphoren, Munzen und Stein-
inschriftenganz oder teilweisebelegte Personennamenwie Moschos, Euphamidas,Anthe-
sterios und als thrakischerName Rebas/Rebada gewonnen werden. Der Charakteraller
dieser Denkmalergruppenlegt nahe, daB es sich bei den Genannten weniger um Hand-
werksmeisterbzw. Werkstattbesitzer,sondern um stadtischeBeamte (Magistrate,vielleicht
21 Zeest 1966, Taf. 30:11, 31:1, 4; Akerstrom1966, S. 2, Taf. 1:8. Buchstaben: P-H-B-A-A-A, links- und
rechtslaufig.
22 Ognenova 1975, Abb. S. 55, 61, 65, 87. Ognenova-Marinova1980.
23 Ognenova-Marinova 1980, S. 113-115, Nr. 9-12, Abb. 7-10.
24 Ognenova-Marinova 1980, S. 115-116, Nr. 13-17, Abb. 11 = Traufziegel;S. 145, Nr. 74, Abb. 65 =
Sima.
25 Flow 1915, Sp. 235, Abb. 13; Ognenova-Marinova1980, S. 130-138 passimNr. 48, 48:a, 50-52, 58-60,
62-63, Abb. 37, 38, 40-42, 48-50, 52, 53.
26 Ognenova-Marinova 1980, S. 116-120, Nr. 18-34, Abb. 12-24 = Traufziegelmit Einzelbuchstaben:T,
N, 0, Silben: PO, HP (Ligatur),TI, MI, AA, AN, ET, bzw. Silben-Kombination:MI-AA, AN-E-SET,
S. 143-145, Nr. 72, 75, Abb. 62-63, 66 = Simen mit Einzelbuchstaben:B bzw. deren Kombination: P-H-B.
27
Karajotov 1975, S. 26, 27 obere Abb.; angeblichmit BuchstabenA.
28
Vgl. Bra~inskij,in Zeest 1966, S. 40 mit Anm. 28.
29 Ognenova-Marinova 1980, S. 147-155.
30 Ognenova-Marinova 1980, S. 154, Abb. 75.
TRAUFZIEGEL
MITRELIEFMAANDER
AUSDEMSCHWARZMEERGEBIET 227
40 Ognenova-Marinova 1980, S. 152-153; vgl. Ognenova 1960, S. 225, Abb. 3, 229 mit Anm. 2; Filow
1915, Sp. 235. Die in Nesebar gemachten Funde scheinen generell etwas zu hoch angesetz zu sein; das trifft
z.B. fUrdie Mauern zu, die sich (wieAlexandrescu1982 mitteilt)gut mit der hellenistischenMauer von Histria
vergleichenlassen und deshalb herabdatiertwerden mtssen.
41 Sokol'skij1976, S. 46, Abb. 32:2.
42 Nach Auskunftvon M. Coja bedeutet -1.50 m. das unterste, also das frilheste hellenistischeNiveau
(1. Halfte 3. Jh. v.); aus dem gleichen Niveau stammen-wenn auch an anderer Stelle des Sektors Z2
ausgegraben-von ihr an den Anfang des 3. Jh. v. datierte kannelierteund glatte Schwarzfirnis-Kantharoi
wie z.B. HistriaV, S. 45, Nr. 12, 13, S. 37, Abb. 2, Taf. 2:12, 13; die von ihr vorgeschlageneDatierung des
Traufziegelsin das 4.Jh. v. ist also zu berichtigen.
43 Wie die Fundkomplexe Histria 1972 T ([obiecte] 54-57, 59) zusammen mit dem Keramik-Komplex
Histria 1972 T (29) gefunden, der gemischte Keramikmit Schwarzfirnisund Kanneluren sowie West-Slope-
Ware vornehmlich des 3./2.Jh. v., aber auch ein Stuck des ausgehenden4.Jh. v. enthalt.
4 Fundinventarnicht vorhanden; befand sich in einem Kasten mit Funden Histria 1966 T, der Pseudo-
Kos-Amphorenund Keramik des 1.Jh. v. enthalt, wobei nicht sicherzustellenist, ob der Kasteninhalteinen
Fundzusammenhangwiedergibt.
45 Nach Auskunftvon M. Coja bedeutet "Wohnung mit Pflaster"an diesem Fundplatz ein Niveau mit
Bautrimmern, das uber der Schicht des 4.Jh. v. liegt, also hellenistischzu datierenware.
TRAUFZIEGEL
MIT RELIEFMAANDER
AUS DEMSCHWARZMEERGEBIET 229
moglich,46 Dachterrakotten, die den Namen des Moschos (ca. 3. Viertel des 4. Jh. v.)
tragen,47 mit Olbia, diejenigen mit dem des Euphamidas dagegen mit Mesambria (Ne-
sebar)zu verbinden;auch fir die in Odessos (Varna)und Mesambria(Nesebar)gefundenen,
wohl in der gleichen Matrize geformten Stucke mit dem Namen des Anthesterios ergibt
sich auf diese Weise westpontische Provenienz, die man auch fir den in Olbia, Histria
und Mesambria (Nesebar)gleichermaBenvorkommendenthrakischenNamen Rebas an-
nahmen mochte. Neben jeweils lokaler Produktionvon Dachterrakottenin Olbia, Histria
und Mesambria(Nesebar)bestehen also zwischendiesen Stadten auch Exportbeziehungen,
die nattirlichnicht auf diese beiden Orte beschranktbleiben. Denn fir Olbia48 werden
auf der BasiskleinererUnterschiedein der Zurichtung,Ornamentikund Tonzusammenset-
zung nicht nur verschiedene lokale, vielleicht nacheinander arbeitende Werkstitten, son-
dern auch Import aus Sinope und Herakleia wahrscheinlichgemacht. Dafur bieten die
zwei Traufziegelfragmenteaus Histria (6, 18), die sich durch ihre Tonbeschaffenheitals
Importstuickeaus Sinope zu erkennengeben, eine willkommeneBestatigung.Weitere Aus-
sagen zum wichtigen Problem der Austauschbeziehungenzwischen den Griechenstadten
des Schwarzmeergebietesin hellenistischerZeit lassen sich aus dem hier vorgesteiltenhis-
trianischenMaterialvorerstnicht gewinnen, zugleichwird aber deutlich,welchen Beitragzu
dessen Losung Tonanalysenleisten konnen.
Die in Histria bei den Traufziegelnmit Reliefmaanderdurch kleine Abweichungen in
Zurichtungund Ornament angetroffeneVielfalt macht deutlich, daB3es sich bei ihnen um
Uberresteeiner groBerenZahlvon DachrandernhandelnmuB. In keinemFallgibt esjedoch
Nachrichtendaruber,ob in der Nahe ihrerFundorteauch Gebauderesteangetroffenwurden,
deren Aufbau sie dann mit gewisser Sicherheit zugeordnet werden konnten. Vermutlich
haben die Bauwerke,zu denen sie einst gehorten, nicht oder nur im Sockel aus Stein, im
uibrigenaus verganglichemMaterialbestanden. Denn auch uber das sonstigeAussehen der
Dacher ist nichts weiter zu ermitteln, als daB die zur Front umbiegenden Seitenstege eine
bestimmte Form von Traufkalypterenerforderlichmachen. Von solchen an der Stirnseite
geschlossenenund wohl ebenfallsmit Reliefdekorgeschmuckten,vor allem aber im unteren
AuflageteilausgeklinktenTraufkalypterenhellenistischerZeit sind, im Gegensatz zu Olbia,
aus Histriafisher keine Uberrestebekanntgeworden.Hingegen lassen sich die Wurzelndes
zumindest im Schwarzmeergebietdie tonernen Dachrander hellenistischerBauten haufig
zierenden, hier also besondersbeliebten Reliefmaandersmit eingestreutenFeldernbis zum
normalerweise gemalten Hakenmaander spatarchaischerZeit zuruickverfolgen.49In der
spateren Zeit ist ihm eine groBe Verbreitungals Dekorationsgliedmit Aufnahme sogar in
den Monumentalbaubeschieden.
KATALOG
50 Nach Auskunftvon M. Coja zwischen 1949 und 1951 ohne stratigraphischenZusammenhangim weiteren
beschriftet.
Umfeld der klassischenMauer Z2 gefunden und offenbar 1955 von ihr "rUckwirkend"
51 In der romischenNekropole (M = mormint = Grab),also in Sekundarlageangetroffen.
52 Wie oben Anm. 51.
53 Nach Auskunft von M. Coja in oberster hellenistischerSchicht, angeblich aber nicht in ungestorter
Fundlageaufgetaucht.
TRAUFZIEGEL
MIT RELIEFMAANDER
AUS DEMSCHWARZMEERGEBIET 231
gelber, ins Grunliche tendierenderUberzug, am be- Ton matt ziegelrot;vereinzeltmittelgroBedunkel-
sten an Unterseitebewahrt;an Unterkantezusatzliche rote, auch weiBlicheMagerungsteilchen,etwas Glim-
dunkelroteFarbspuren. mer. Spuren von Braunrotan "Wtrfel"-Punktenin-
Linkes Eckstuck, rechts und hinten gebrochen, zwischenverloren.
tberall starkbestoBen;aber Ziegelplatteerhebt sich Mittelstack, hinten gebrochen; Front stark be-
ein ca. 2,5 cm hoher und ca. 2,2 cm breiter Steg, der stoBen und abgerieben. Reliefmaander mit 2 ein-
5,6 cm auf die Front umbiegt; diese ca. 9 cm hoch. gestreuten Feldern: 5 "WtIrfel"-Punkte,Reste einer
Reliefmaander mit eingestreutem, moglicherweise achtteiligenRosette.
durch Seitenbegrenzung angeschnittenem Feld:
Buchstabe T; Frontsteg dartiber mit teilweise ge- 11. Histria,vor 1957 Figs. 1, 10
rahmtem, seitlich geschnittenemEierstab. (nv. V 2445)
Aufschrift,schwarz: H V 2445.
7. Histria, 1954 SektorX Fig. 8 Canarache 1957, S. 171 Nr. 457, mit Abb.
= =
Br 9,0; H 11,4; T 9,2. = Br=6,1;H=4,7;T= 3,2.
Ton fein, sandig, gelblich bis matt rbtdich;sehr Ton weich, sandig, hellrot; viele kleine weif3liche,
kleine weif3liche, rotliche und schwarzliche Mage- gelbliche, rotliche und schwarzlicheMagerungsteil-
rungsteilchen. Hellgelber Uberzug in Spuren an chen.
Front und Oberseite. Allseitig,auch hinten gebrochenesFragment. Re-
Rechtes Eckstfck, links und hinten gebrochen, liefmatandermit eingestreutem Feld: Buchstaben-
starkabgerieben;weitgehend geglattete,nach hinten gruppe OT.
leicht zurilckweichendeSeitenflache nach ca. 5 cm
12. Histria, 1959 SektorZ2 Fig. 11
abgesetzt und etwas tiefer fortgef'uhrt;uiberZiegel-
(Inv.V 20 531)
platte erhebt sich ein ca. 2,4 cm hoher, in der Breite
differierenderSteg, der 5,9 cm und etwas schmaler Aus Schnitt S4.
auf die Front umbiegt; diese 9,0 cm hoch. Re- Br = 9,9; H = 5,7; T = 8,0.
liefmaander;Frontstegdarubermit teilweisegerahm- Ton sandig, dicht, matt ziegelrot; rotliche Mage-
tem Eierstab. rungsteilchen, Glimmer. Weinrote Farbtupfen an
Oberseiteneben Aufw6lbung.
8. Histria, 1954 SektorX P1.79:d Mittelstack,unten und hinten gebrochen;aufglat-
[Verbleibunbekannt] ter Oberseiterechtsnahe der FrontkleineAufwvolbung
(0,2 cm) als Uberresteines aufragendenSteges;Front
Br = 10,0; H = 5,5; T = 7,4.
starkbestoBen. Reliefmaandermit Rest eines einge-
Ton rotlich. COberalldunkelroter tJberzug, an
streutenleeren Feldes.
Frontnur noch in Vertiefungenerhalten.
mit
Mittelstuck,hinten gebrochen. Reliefmnaander 13. Histria, 1959 SektorZ2 Fig. 12
eingestreutemFeld: vierteiligeRosette.
Aus Schnitt S4.
Br = 14,4; H = 7,3; T = 10,2.
9. Histria, 1954 SektorX P1.79:e
Ton sandig, dicht, hell ziegelrot;wenige weiffliche
[Verbleibunbekannt]
Beimengungen,sehr viel Glimmer.
Br = 4,5; H = 3,7; T = 2,9. Linkes Ecksttick mit stark beschadigter Seiten-
Ton rotlich. flache,hintengebrochen;Frontsehrstarkbeeintrach-
Mittelstuck,auBer oben allseitig gebrochen. Re- tigt.. Reliefmnaander mit eingestreutem Feld: Rest
liefmaander. einer achtteiligenRosette.
10. Histria, 1956 T (43)54 Fig. 9 14. Histria, 1960 T (18 [?])55 Fig. 13
Br = 16,1; H = 7,7; T = 10,2. Br = 15,0; H = 7,0; T = 6,2.
54 Fundinventarnicht vorhanden.
55 Fundinventarnicht vorhanden.
232 KONRADZIMMERMANN
Ton matt ziegelrot bis braunlich; unterschiedlich Mittelstuck, hinten gebrochen, uiberallstark be-
groBe ziegelrote, wenige weiBliche Magerungsteil- schadigtbzw.abgerieben;Unterkanteder Frontdurch
chen, Glimmer. Gelbgrauer tlberzug mit Stich Abrieb nicht komplett originateeFronthohe dehalb
ins Grunliche an Oberseite, Front und deren Un- bei 6,4-6,5 cm). Reliefmaiandermit eingestreutem
terkante;schwarze Flecke vor allem an Ruckseiten- leeren Feld.
bruch modern.
Mittelstuck, hinten gebrochen; Front sehr stark 17. Histria, 1963 (?) SektorZ2 Fig. 16
bestoBen. Reliefmaander mit eingestreutem Feld: Passim.
5 "Wurfel"-Punkte. Coja, in HistriaV, S. 53 Nr. 90, Taf. 11:90 (um 90?
zu drehen).
15. Histria, 1963 SektorZ2 Fig. 14; P1.79:a-c
Br = 7,0; H = 7,8; T = 7,2.
(nv. V 25 536)
Ton fein, gelblich-braun;Glimmer;ein deutlicher
Aus Schnitt S6a, Stid-Profil,-1,50 m.56 Muschelabdruck.
Coja, in HistriaV, S. 53 Nr. 88, Taf. 11:88. Mittelstick, hinten gebrochen, uberallstarkabge-
Br = 32,2; H = 10,0; T = 16,0. rieben, besondersan Unterkante(Fronthohedeshalb
Ton fein, sandig, ockerfarbenbis rotlich; an Fla- vielleichtetwas hoher). Reliefmaander.
chen aufgerauht,sonstporosdurchbrockeligerotliche
Magerungsteilchen,Glimmer. Moglicherweiseton- 18. Histria, 1963 (?) Sektor Z2 Fig. 17
farbenerUberzug (Coja). Passim.
Aus 2 FragmentenzusammengesetzteslinkesEck- Coja, in HistriaV, S. 53 Nr. 91, Taf. 11:91 (steht
stUck,rechtsund hinten gebrochen,uberallbestoBen; auf dem Kopf); Zimmermann,in Alexandrescuund
geglattete linke Seitenflache nach hinten leicht zu- Schuller 1990, S. 177 Abb. 23.
rtickgenommen;in 11,6 cm Tiefe springt ein senk- Ton und Uberzug einschlieBlich Verteilung 6
rechter, 3,4 cm breiter Steg vor, dessen Stirn in nachst verwandt.
der Ebene der linken Frontbegrenzungliegt (rechter Mittelstuck mit glatter Oberseite, hinten gebro-
Winkel);hierim unterenFrontbereichzapfenformiger chen; Front bestoBen. Reliefmaandermit Rest vom
Vorsprungfiberdie Seitenflachehinaus;uber Ziegel- Rahmen eines eingestreutenFeldes.
platte erhebt sich ein ca. 3 cm breiter,heute an seiner
19. Histria, 1966 T (65)57 Figs. 1, 18
Oberflachevollig abgerundeter,noch ca. 2,4 cm ho-
her Steg, der weniger breit und mindestens 6,8 cm Br = 8,7; H = 6,0; T = 5,0.
auf die Frontumbiegt. Reliefmaandermit mehreren Ton matt ziegelrot;unterschiedlichgrol3erodiche
eingestreuten Feldern (von rechts): wegen Bruch Magerungsteilchen, Glimmer (?). Rodicher bis
Fullung unsicher, 5 "Wurfel"-Punkte, funfteilige violetter tJberzug an Oberseite, Front und deren
Rosette, leeres (?) Feld an linker Seite geschnitten; ruckwartigerWolbung; an Front groBtenteilsabge-
Frontstegdarubermit Eierstab. platzt.
Mittelstuck, hinten gebrochen; Front stark be-
16. Histria, 1963 (?)SektorZ2 Fig. 15
schadigt. Reliefmaandermit eingestreuter,leicht an-
Passim. gehobenerquadratischerPlatte(2,5 x 2,5 cm): Buch-
Coja, in HistriaV, S. 53 Nr. 89, Taf. 11:89. stabe H.
Br = 11,0; H = 6,3; T = 6,0.
Ton ockerfarbenbis hell ziegelrot;weif3liche,ver- 20. Histria, 1966 T58 Fig. 19
einzelt rotlicheund schwarzlicheMagerungsteilchen, Br= 13,5;H= 10,4;T=7,2.
Glimmer. Gelblicher tCberzugan Unterkante und Ton dicht, matt ziegelrot; wenige weif3licheBei-
Wolbung kraftiger,an Front nur in den tieferliegen- mengungen, kaum Glimmer;vollkommenweif3grau
den Partienbewahrt. versintert.
59 Zum Teil auBerordentlichschwach, scheinen sich flinf Rosettenblattersicher und mcglicherweise zwei
weitere abzuzeichnen.
60 Wie oben Anm. 45.
61 Vgl. oben Anm. 43.
62 Wie oben Arn. 16.
234 KONRAD ZIMMERMANN
Mittelstuck,auBer oben allseitig gebrochen. Re- abgerieben; Oberflache der Ziegelplatte beidseitig
liefinalander. geringfidgigansteigend. Reliefmaander mit 2 ein-
28. Histria gestreutenFeldern: achtteilige Rosette, 5 "Wurfel"-
Figs. 1, 27
[1983 im Depot aufgetaucht] Punkte.
22~~~~~~~7
A _ _ _~~~~~~* . _
31 11
-- m..- 0 5
woicm
4-~~~~~~~~~~~~~~~~~~~~~f
28 -
~~S1
J-g3 T 1 -_
19 4
25
FIG. 1. Zusammenstellung von Traufziegelnmit Reliefmaanderund eingestreutenBuchstaben: Fronten von
31, 11, 28, 1, 6, 19, 4, 25
236 KONRAD ZIMMERMANN
0 ~~~5
- I____________________cm
I 5
1mw --Walcm
41
A .
cm F 4 3 S h
rechts
FIG.4. 3: Front,Seitenansicht
TRAUFZIEGEL MIT RELIEFMAANDERAUS DEM SCHWARZMEERGEBIET 237
o 5
cm
I _~~~~
FiG. 5. 4: Front, Schnitt
0 5
c mcm
0 5
cm
0 5
cm
F~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~F
9. 10:
FrontLc
FIG. L
kmt~ji cm
0 5
I 'cm
~~~~~~0 3
t
-t ~~~~~~0~ 5
~~___________ cm
0 5
wo mom cm
Schnitt
FIG.12.13: Front,
tullei1XSsS'.7".///-
0 5
cm
I~~~~~~~~~~~~~~~~~~~
ci
1l < Xa
t S :Sm~~~~~~~~~~~~~
vt 1
<^ ~~~~~~~~~
TRAUFZIEGEL MIT RELIEFMAANDERAUS DEM SCHWARZMEERGEBIET 243
I/
cm
cm
t
0 5
-- - cm
1t
FIG.12:rt'icm
0 5 0 5
I cm _ _cm
FIG. 20. 21: Front, Schnitt FIG. 21. 22: Front, Schnitt
0 5
cm
0 5
cm
4.1
1z
0 5
cm
LF
cn
V)4
*
248 KONRAD ZIMMERMANN
I~~~~~~~~~~~~~~~~
An-~~~~~~FG2
T ~~~~~~~~~~WN-mlcm
2 rn Shit
IOI 01 5
Acm
I~~~~~ ci~~~~~ -
250 KONRADZLMMERMANN
BIBLIOGRAPHIE
Akerstrom,A. 1966. Die architektischen TerrakottenAlnasiens(ActaInstituiAdenienses RegniSueciaeXI, Lund
Alexandrescu,P. 1982. Rezension zu Nessebre II in Pontca15, S. 47-55
Alexandrescu,P., und W. Schuller. 1990. Histria.EineGriechnstadt anderrnandischen Schuameerkiiste (Xenia.Kon-
stanzerAlthistorische
Vortrlge
undForschkwgen 25) [darinS. 155-157: K. Zimmermann,"Zu den Dachterrakotten
griechischerZeit aus Histria'1, P. Alexandrescuund W. Schuller,Hrsg., Konstanz
AvP VIII, ii = E. Fabriciusund C. Schuchhardt,Die Insch/rifnwn Pergamon (Altertimer vonPergamon VIII, ii),
M. Frankel,Hrsg., Berlin 1895
AvP IX = E. Boehringer und F. Krauss, Dos TemenosfrdenHerrscherkult, "Prinzessinnen Palais"(Alertimervon
Pergamon IX), Berlin 1937
AvP X = A. von Szalay und E. Boehringer,Die helenistischen Arsenale(Altemer vonPergamon X), Berlin 1937
Billot,M.-F. 1976. "Terrescuitesarchitecturalesdu Musee Epigraphique,"AeX- 31, 1976, A' [1980], S. 87-135
Buschor,E. 1929. Die Tond&herder Akropols,I, Simen,Berlin/Leipzig
Canarache,V. 1957. Importul amforelor
stampilate la Histia, Bucuresti
Coja, M. 1962a. "ActivitateamestesugfreascAla Histria in sec. VI-I f.e.n.," Studiisi cercetdri de istorieveche[si
arheologie]13, S. 19-46
. 1962b. "L'artisanata Histria du VI' au I` si~cleavant notre 6re,"Dacian.s. 6, S. 115-138
DelosV = F. Courby,Portique d'Antigone
oudunord-est (DelosV), Paris 1912
Flow, B. 1915. "Bulgarien,"AA [JdI30], Sp. 218-236
Gneisz, D. 1994. "Die Dachterrakottenvon Aigeira,"in Hesperia Supplement27, pp. 125-134
HesperiaSupplement27 = Proceedings of theInternational Conference
onGreek Architectural
Terracottasof theClassicaland
HelenisticPeriods,Athens,December12-15, 1991 (Hesperia Supplement27), N. A. Winter,Hrsg., Princeton 1994
Histria V = M. Coja und P.Dupont, Ateliers cdramiques (HistriaV), Bucuresti/Paris 1979
Hornbostel, W. 1980. MusewnflirKunstundGewerbe Hamburg: Aus GribernundHeiIgtiimern. Die Antkensammlung
W Kropatscheck, Mainz
Habner, G. 1975. "Dachterrakottenim Magazin des Museums von Nauplia," in 7-ryns VIII, Mainz,
S. 117-136
Kastner,V 1994. "Kleinasienund Griechenland: DachterrakottennacharchaischerZeit aus Pergamon,"in
HesperiaSupplement27, pp. 253-268
Karajotov,I. 1975. "Trakijskakrepost na severnijabrjag na Mandrenskotoezero," Izhutro (Sofija)25:3-4,
S. 25-29
Koch, H. 1915. "Studienzu den campanischenDachterrakotten,"RM 30, S. 1-115
Metzger,I. 1971. "Piraeus-Zisterne,"AeX- 26, 1971, A' [1973], S. 41-94
Miller,Stella G. 1994. "ArchitecturalTerracottasfrom Ilion," in Hesperia Supplement27, pp. 269-273
Ognenova, L. 1960. "Lesfouillesde Mesambria,"BCH 84, S. 221-232
1975. "Podvodnaarcheologijav Neseb~r," Vekove 1975:3, S. 43-48
Ognenova-Marinova,L. 1980. "Tuileset terres cuites architecturales,"in Ness~bre II, V. Velkov,Hrsg., Sofia,
S. 110-155
Phrvan,V. 1916. HishiaIV (Analele AcademitRomdn, Mewe Seqini
Istoece, ser. 2, 38), Bucuresti
Pippidi,D. M. 1962. "Gli scavi nella zona sacra di Histria,"Dacian.s. 6, S. 139-156
Sokol'skij,N. I. 1976. Tamanskitolosi reziden~aChkisaiska, Moskva
Wiegand, Th., und H. Schrader. 1904. Pine. Ergebnisse derAusgrabungen undUntersuchungen in denJahren1895-
1898, Berlin
Zeest, I. B. 1966. KeramiZeskoc proizvodstvo i antinyc keramieskiestroitel'nye material = Archeologija SSSR.
Swd archeologjleskix G 1-20 [darin S. 36-45: I. B. Braginskij,"Ol'vija's, I. B. Zeest, Hrsg.,
isto~nikov
Moskva
TRAUFZIEGEL
MIT RELIEFMAANDER
AUSDEM SCHWARZMEERGEBIET 251
KomNADZImRmANN
UNIVERSrIAT ROSTOCK
Institutfar Altertumswissenschaften
FachgebietKlassischeArchaologie
Universitatsplatz
D-18051 Rostock
BundesrepublikDeutschland
PLATE 79
i 4 .*..
a. 15. Seitenansichtlinks
. 5
^ ~~~..
.,,ls~~~~~~~~~i 9Frn
8..Fronti
d.~~b ele~an
_w_~~~~~~~~~~~~~~~~~~Vn
c. 15. Aufsicht
KA Z SW
KLEINASIENUND GRIECHENLAND
DACHTERRAKOTTENNACHARCHAISCHERZEIT
AUS PERGAMON
(PLurEs
80-83)
Dachziegel
Die Ziegelbestehenaus hellrotemTon und sind auf der Oberseitemit einem dunkelroten
Uberzugversehen.Der Flachziegelist 57 cm breitund 65 cm lang. Die Randersind 5,8 cm hoch
aufgebogen,die OberseitezeigtamhinterenEndeeinenschmalenArretierungssteg
rechtwinklig und
an derVorderkanteeineuntenuiberstehende Verdickungvon etwadreieckigemQuerschnitt.Deren
Vorderseite ist bei den Traufziegeln
haufig mit eingedrucktenSchriftzugen(PaoaLXLxvj,
PaOXeLc&V,
Traufiiegelfragment
Antikensammlung,Inv. P 716 (1616x; ZettelinventarNr. 390)
Swastikamaandermit alternierendem Bliitensternquadrat(vom Stern nur geringe Spuren
erhalten).
Hohe: 7,9 cm; Lange (erh.): 15,6 cm; Tiefe (erh.):8,5 cm.
Traufziegelfragment
Antikensammlung,Inv.P 717
Dekorationwie P 716 abermiteinemgut erhaltenemBlitenstern(mitPunkt
RechtesEckstiick;
in der Mitte).
Hohe (erh.):7,7 cm;LInge(erh.):20,6 cm;Tiefe(erh.):8,3 cm.
Traufziegelfragment
Antikensammlung, Inv.P 718 (155x; Zettelinventar
Nr.390)(P1.80:c)
Dekorationwie P 716.
Hohe (erh.):7,7 cm;Lainge(erh.):14,2cm;Tiefe(erh.):4,5 cm.
Traufziegelfragment
Antikensammlung, Inv.P 719
Dekorationwie P 717.
Hohe (erh.):6,7 cm;Linge (erh.):14,5cm;Tiefe(erh.):9,1 cm.
Die Ziegeldicke betragt allgemein 2,4-2,7 cm -und die Stirnseite war wohl bei allen
aufgefuhrtenBruchstuckenursprunglich7,9 cm hoch. Somit gehorten diese Exemplare
zu einem einzigen Ziegeltypus und wurden vielleicht auch zusammen auf dem Burg-
berg gefunden. Daneben konnten andere Fragmente desselben Typs im Herrscherkult-
Temenos7und im Olivenhainbeim Asklepieiongeborgenwerden.8 Weiterepergamenische
Fundstuckevariieren das Ornament und haben ein kleineres Format-so ein Ziegel aus
dem Herscherkultbezirkmit Kreuzstern-Quadratund ein Maander ohne Blutensternaus
dem Asklepieion.9 Derartige Traufziegel mit reliefiertemMaandermusterwaren beson-
ders im kleinasiatischenAmbiente (Ilion, Assos, Erythrai, Priene, Chios, Samos) und im
Schwarzmeergebietweit verbreitet.10 Das Ornament erschiendabei in verschiedenenVari-
anten (Maanderrapportemit Quadratfeldern,Schachbrettmustern,Buchstabenfeldernund
verschiedene Sternformen)oder wurde um weitere Zusatze bereichert(ionischeKymatien,
Anthemien)." Datiert werden diese Ziegel allgemein in hellenistische Zeit. Sollten die
Maandertraufziegelaus Priene wirklich mit der ersten Phase des dortigen >Ekklesiaste-
rions< zu verbinden sein, dann ware wenigstens an diesem Ort eine engere zeitliche Fix-
ierung eines Maanderziegeldachesin die Zeit vom Ende des 3. bis zum Anfang des 2.Jh. v.
Chr.moglich.12 Das Maanderornamentmit Blutensternenwar in gemalterFormaufZiegeln
aber bereits seit archaischerZeit (Histria)bekannt und einfache reliefierteMaandermuster
tratenin Verbindungmit Simen und Verkleidungsplattenebenfallsschon im 6.Jh. v. Chr. in
Kleinasien auf (Milet, Mylasa, Sardes).13Dies und der Mangel an gesichertenFundzusam-
menhangen erschwerenes bisher, die Anfange der besonderen hellenistischenAuspragung
des Maanderziegelszeitlich genauer einzugrenzen.
Zu den einfachsten reliefierten Stirnziegeln im hellenistischen Kleinasien und im
Schwarzmeergebietzahlen die funfeckigenVerschluBplattender korinthischenKalypteremit
aufgepragtenPalmettenmustern.In Pergamonwurdeein AntefixdiesesTyps im Asklepieion
gefunden:
AvP IX, S. 133 Nr. 1-5, Taf. 42:b; H6he (erh.):7,6-6,7 cm.
8 AvP XI, ii, S. 119, 125, Nr. 638 (Inv.OH 67 T 408/175); H6he (erh.): 7,5 cm; Breite (erh.): 16 cm. Die
beiden zusammengehorigenZiegelfragmentewurdenaus den Verschuttungsmassender langen hellenistischen
Halle im Olivenhain geborgen, die im 1. Drittel des 2. Jh. v. Chr. errichtetewurde und bis in das 4. Jh. in
Benutzungblieb (ebd. S. 32-44, 84-85).
9 AvP IX, S. 133, Nr. 7:b und AvP XI, ii, S. 125, Nr. 639.
10 Zu Kleinasien vgl. Akerstrom 1966, S. 2, 13-15, 35, 100, Taf. 1:7-8, 7:1 und 3, 17:4-5; zu den
Maanderziegelnim Schwarzmeergebietsiehe den Beitragvon K. Zimmermann(1994). Weitere Hinweise auf
Reliefziegel, z.B. auch aus Abdera, Athen und Delos, nennt Billot 1976, S. 105-106 mit Taf. 26:b (Athen).
Reliefmaanderlassen sich auBerdemauch in Suditalien nachweisen (vgl. z.B. eine Sima aus Sybaris: Sibari
IV, S. 355, Nr. 153, Abb. 335).
I Z.B.: SamosXIV, S. 42-44 (A 540), Abb. 80.
12
Akerstrdm1966, S. 100.
13
Akerstrom1966, S. 104, Abb. 33, Taf. 43:1, 44:1, 59:1.
256 VOLKERKASTNER
Pentagonalantefix
Bergama,Asklepieion-Depot(ohne Nr.) (P1.81:b)
Kalypterfragment rotenTon. Der Deckziegelist innenhalbrund
mit Antefixausversintertem
Antefixplatte
undschlieBtmit einerflunfeckigen Flammen-
ab. DasAntefixzeigteineflachreliefierte
palmettefiberliegendenS-Voluten.
H6he: 9 cm;Breite:13,5cm;Tiefe(erh.):9,5 cm.
Palmettenantefix
Antikensammlung,Inv. P 712 (Burgberg,Fundortunbekannt)16(PI.81:a)
SpitzbogigeAntefixplattemit reliefierterneunblattrigerFlammenpalmette.In der linkenunteren
Ecke eine kleine Volute, deren Ranke sich von unten anlaufend im Uhrzeigersinn einrollt. Oben
ein abzweigendesBlatt (?). Unten nach rechtsansteigendeBruchkante,Ausbrucheim Palmettenfeld
und ruckseitigerAnsatz des in einem MittelgratauslaufendendachfbrmigenKalypters. Ton grau
verfarbtmit gleichfarbigemUberzug (sekundarerBrandschaden?).
Hohe (erh.):14,7cm;Breite(erh.):13,5cm;Tiefe(erh.):4,5 cm.
14 Akerstrom1966, S. Taf. 1:3, 36:4, 52:4, 59:3 und 5; SamosXIV, S. 43-44, Abb. 81; Zimmermann 1983,
S. 288, Typ A IV/ 1-2, 290-292. (Datierung2.-3. Jh. v. Chr.),Taf. 24.
15 Fruhe Beispiele aus dem samischen Heraion (Buschor 1933b, S. 22-46); vgl. auch Harl 1971, S. 63-71
(Voluten-Typus).
16 Archiv der AntikensammlungInv. 37 (SculpturBarackeII), Nr. 117.
'7 Akerstrom1966, Taf. 8:4 und 6.
8 DAI Athen, Foto Tro. 168.
'9 AvP II, S. 10 (mit Zeichnung).
DACHTERRAKOTTEN NACHARCHAISCHER ZEIT AUS PERGAMON 257
Auch die jungeren Marmorantefixe des von Attalos II (159-138 v. Chr.) gestifteten
Heratempels besitzen unter der frei gearbeiteten siebenblattrigen Flammenpalmette noch
zwei sich nach auBen einrollende Voluten (Fig. 1:e).22 Bemerkenswert und mit Stirnziegeln
aus Assos vergleichbar ist das kelchfodrmigeHerzstuck dieses Antefixes.23 Weitere allerd-
ings im Palmettenmotiv abweichende Marmorantefixe befinden sich im pergamenischen
Grabungsdepot (P1.81 :d). Der Blattficher zeigtje vier in der Mitte gekerbte Blatter, die bei-
derseits eines ebenso gekerbten senkrecht aufsteigenden Mittelstengels nach auBen schwin-
gen, wobei die Blattenden jetzt abgebrochen sind. Darunter biegen vom Herzstuck ausge-
hend zwei dunne Volutenranken nach auBen und rollen sich nach oben ein. Die gerade
abgeschnittene Unterseite war wohl zum Aufsetzen auf eine Sima gedacht. Sehr ahnliche
Palmettenmotive zeigen auch hellenistische Tonantefixe aus Didyma, Histria und Korinth,24
ohne daB aus dem Erhaltenen eine genauere Vorstellung vom oberen AbschluB des Motivs
gewonnen werden konnte.
Neben der Palmette uibereinem Volutenpaar kennen wir aus Kleinasien als Dachschmuck
die Palmette uber stehenden S-Voluten (Volutenleier).25 Das Motiv begegnet uns schon in
000 go4'.
(r
a 40cOm b
all~~~
'.4..
..~~~~~~~~~~~~~~~~~o 0
/ ~~~~~~~~~~~~~d
He '- N
e f
FIG. 1.a. AndesitantefixInv. 138/1. b. Firstpalmetteaus Marmor. c. AndesitantefixInv. 138/2. d. Antefix
aus >Trachyt<. e. Marmorantefixdes Heratempels. f. Marmorantefixdes ionischen Tempels beim
Gymnasion.
DACHTERRAKOTTEN ZEITAUSPERGAMON
NACHARCHAISCHER 259
archaischerZeit als Akroterin Assos26oder als Reliefmusterauf kleinasiatischenSimaplat-
ten27 und daneben besonders in Kleinasien und Attika als Grabstelenbekronung.28Auf
Antefixen erscheint dieses Ornament im mutterlandischenAmbiente seit etwa 500 v. Chr.
meistmit nahzu waagerechtliegendenS-Voluten(JungererAphaiatempelAigina, Stiermon-
ument auf Delos, Olympia etc.).29Die ostlichenAntefixen zeigen dagegen eine analog den
Stelenbekronungenund Firstpalmetten30gestalteteVariantedes Motivs mit aufgerichteten
S-Voluten, wie sie zum Beispiel in gemalter Form sicher auch die Marmorziegel des Nike-
tempels auf der Athener Akropolisschmtickte.31Ein Antefixfragmentaus dem athenischen
Kerameikos,das nach G. Hubner den Eindruckerweckt,als ware es ,,geradezuvon einem
lonier gemacht",lieBesich hier anschlieBen.32WeitereStirnziegelmit dieserOrnamentvari-
ante wurden in Olynth33und Abdera34gefunden. Dabei kann gerade bei den abderitischen
Dachterrakottenallgemein schon in archaischerZeit-eine starke Abhangigkeitvon klein-
asiatischenund nesiotischenWerkstattenbeobachtetwerden,35was letztlichdie Vermutung
bekraftigt,daB der Ursprungdes >Lyratyps< in ebendiesemAmbiente zu lokalisierenist36
Seiner auBerenForm nach laBtsich diesen Palmettenantefixendes >Lyratyps<vermut-
lich ein weiteresAndesitfragmentaus PergamonanschlieBen:
Taf. 222:g.
35 Vgl. dazu auch Koukouli-Chrysanthanki1970, S. 358.
36 Veilleicht kann man Palmettenantefixemit S-fbrmignach aussen schwingendenVolutenenden (Stdbau
im samischeHeraion: Akerstrom1966, S. 99, Abb. 30, 31; Ohnesorg 1990, S. 189, Taf. 21:a;Athenatempelin
Assos: Akerstrom 1966, S. 13-15, Taf. 5:1) als unmittelbareVorlauferdes >Lyratyps< ansehen, der dann
in den letztenJahrzehntendes 6.Jh. v. Chr. in Kleinasienentstandensein durfte.
260 VOLKERKASTNER
Die Palmette des recht gut gearbeiteten Marmorantefixes lIBt sich mit einem alteren
Stirnziegel aus der Unterstadt von Assos41 vergleichen und steht offenbar dem Anthemion-
ornament eines Antenkapitells vom didymrnischen Naiskos stilistisch sehr nahe. Ahnlich
geformt sind hier wie dort die auBen an die Palmettenvoluten angesetzten Blattgebilde
mit kleinen Knospen.42 Danach ware eine Datierung noch in das dritteJahrhundert und
vielleicht die Verbindung mit einem Tempelbau im Asklepieion denkbar.
Auch die ubrigen oben aufgeftihrten pergamenischen Ziegel entsprechen weitgehend
den hellenistischen Funden aus der Unterstadt von Assos. So ahnelt das Antefix in
Berlin (nv. P 710) in der blutenkelchartigen Gestaltung des Palmetten-Herzstuckes den
Fragmenten Inv. P 4132-P 4135 in Boston.43 Fur die Exemplare aus dem Asklepieion-
Depot konnen wir neben den Ziegeln aus Assos auch noch Stirnziegel aus Larisa am
Hermos und Erythrai zum Vergleich heranziehen. Offenbar war dieser Typus in nur gering
voneinander abweichenden Varianten in den Ktistenstadten der Troas und der Aeolis sehr
verbreitet. Nach der Anordnung der kleinen Palmette zwischen den Volutenleiern kann
40 Frau Gioia de Luca verdankeich die nachfolgendennaheren Angaben zur Fundsituation:Inv. T 59/82
stammtaus dem Gebiet der alterenhellenistischenOsthallen-vermutlich aus einer Schicht, die vor der Anlage
der jungeren Osthofe entstand (aus demselben Gebiet, allerdingsaus gestorter Schicht, ein Fragment eines
entsprechendenStirnziegels);T 61/55 vom Vorgelande des Asklepiostempelsauf der Felsbarre(Gebiet des
sogenanntenAltarhauses)und Inv.T 61/131 wurde als Streufundinnerhalbdes Geviertesder kaiserzeitlichen
Hallen des Festplatzesgefunden.
41 Akerstrom1966, S. 15, 20, Taf. 6.
42 Pfrommer 1989, S. 433-434, Taf. 42:3; vgl. weiterhin Antefixe aus Erythrai(EgythraiII, S. 122, Nr. 89,
Taf. 48:2;Akerstrom1966, S. 36, Nr. 2, Abb. 10:1)und einen Stirnziegelaus Atalanti(ausdem Gebiet Kaphassi
mit Bautendes 2.-i .Jh. v. Chr.),wo die Volutenaberbereitsaus einem Akanthuskelchhervorwachsen(Pariente
1990, S. 770, Abb. 90.)
43 Akerstrnm 1966, S. 15-16, Taf. 7:4, 5; die blutenffirmigePalmetten-Herzstuckesind bei den First-
man wenigstens zwei Hauptformen unterscheiden. Die vielleicht altere Variante besitzt
eine hangende Innenpalmettewie das Asklepieion-Antefixund kommt auch in Assos und
Larisa vor.44
Die zweite moglicherweise etwas juingere aber ebenfalls hellenistische Spielart zeigt
dagegen eine aufrechtstehendePalmetteaufder Anteflxbasiszwischenden unterenVoluten.
Dabei kann diese Innenpalmette(wiebei einem Antefix aus Erythrai)45auch auf nach innen
abzweigendenkleinenVolutensitzen. Als RudimentlaBtsich diesesVolutenmotivauBerdem
auch an dem Asklepieion-FragmentT 59/82 beobachten, obwohl hier die Innenpalmette
haingendangeordnetist.
Die Antefixe mit stehenderLeierpalmettewurden in Erythrai,Larisaam Hermos sowie
zusammen mit reliefiertenMaandertraufziegelnund Firstpalmettenbei der Stidhalle, dem
>Bazaar<, der Agora von Assos gefunden. Dieses zweischiffigeund mehrstockigeBauwerk
steht offenbar in der Tradition der spezifisch-pergamenischenHallenbauten46und durfte
wie diese wahrend der ersten Halfte des 2. Jh. v. Chr. entstandensein. Mit dem Dach der
Sudhalle von Assos ergabe sich auch ein Anhaltspunktfur die Datierung der Fragmente
aus dem pergamenischenAsklepieion. Letztere gehoren sicher zu einem Antefixtyp, der
aufgrund seiner konserativerwirkenden Strukturalter ist als der kleinteiligergegeliederte
Stidhallenstirnziegelaus Assos, andererseitsaber nach den Palmettenformenkaum vor dem
Ende des 3.Jh. v. Chr. entstandensein wird.
Auch das vermutlich zu einer Firstpalmettegehorende Bruchsttick,das 1908 bei der
Pergamongrabunggefunden wurde, laBt sich gut mit dem Firstziegelnder Stidhalle von
Assos vergleichen. Bemerkenswertist hier jedoch der Akanthuskelchunter der Palmette,
der dieses Fragmentmit den nun folgendenpergamenischenFunden verbindet.
Wahrend die bisher behandelten Ziegeltypen sich gut in das ailgemeine conservative
Bild der kleinasiatisch-ionischenBaukeramikeinfugen, verkorpert der abschlieBend zu
besprechendeTypus einen Sonderfall,fur den keine Vorstufenin dieser Region existieren.
Mehrere Fragmente dieses Ziegeltyps traten als Grababdeckungim Arsenalbezirkund in
den Schuttschichtender Trajaneumsubstruktionen auf:
Antefixfragment
FotoinventarPergamon1927 Nr. 108: von einem byzantinischenGrab uiberArsenalIV47
(PI.83:a)
UnterteileinesPalmettenantefixes undVolutenranken.
mitAkanthuskelch
44 Assos: Akerstrom1966, S. 15, 20, Taf. 6; Larisaam Hermos: Akerstrom1966, S. 50, Taf. 17:1.
45 Akerstrbm1966, S. 36, Nr. 1, Abb. 10:2.
46 Coulton 1976, S. 70-71. Die hier vorgeschlageneDatierung um die Mitte des 2.Jh. v. Chr. gibt nur den
Antefixfragmente
AntikensammIung,Inv. P 711 und P 720, P 715; >zwischen den beiden Caesarentempeln<48
(P1.83:b)
Zwei zusammengehorigeFragmentevom Oberteil eines Antefixes mit gesprengterdoppelter
Flammenpalmette(>Rosettentypus<)und Bruchstuckeiner Antefixbasiswie oben.
Antefixfragmente
Bergama, Trajaneumgrabung 1983 (aus umgelagertem hellenistischen Bauschutt in den
romischenSubstrukdionen etwaffinfFragmentevon Palmettenantefixendes gleichenTypswie oben)49
Die aufgeffihrtenFragmentebestehen aus hellrotemTon und besitzen eine graugelbe Engobe.
Die erschlieBbarenMaBe eines vollstaindigenAntefixes betragen: Hohe 24 cm und Breite 18 cm.
Der dachformigeDeckziegel ist 12 cm hoch und 18 cm breit;der Ziegel endet mit einem Mittelgrat
auf der Ruickseiteder Antefixplatte.
Die Gestalt des Stirnziegels lalBtsich aus den Bruchstucken volistandig wiedergewinnen.
Das Relief zeigt einen dreiblattrigen Akanthuskelch, aus dem ein herzformiges Volutenpaar
emporwachst, das seitlich von weiteren sich gabelnden und zu Voluten einrollenden Ranken
gestuitzt wird. Aus den Zwickeln zwischen Haupt- und Seitenranken spriel3en schrag nach
oben gerichtete kleine siebenblattrige Palmetten. Zwischen den oberen Voluten haingt eine
kleine Blute. Uber den Voluten erhebt sich die gesprengte doppelte Flammenpalmette.
Sie besteht aus zweimal sechs grol3en aul3eren und zweimal funf kleineren inneren spitz
zulaufenden gratigen Blattern. In der Palmettenspitze sitzt eine kleine achtblittrige Rosette.
Die Beschaffenheit des Tones, die Abmessungen und die ornamental Struktur dieses
Antefixes entsprechen offenbar den attischen Stirnziegeln des >Rosettentypus<, wie sie
beispielsweise im Kerameikos bei der Pompeiongrabung gefunden wurden. G. Hubner hat
diesen Typus in der Pompeionpublikation vorgestellt und bereits auf seine Abhangigkeit
von den Antefixen des Erechtheions-die aber leider nur durch ihre inferioren romischen
Ersatzstucke uiberliefert sind-hingewiesen.50 Es handelt sich damit also um die hellenis-
tische Fassung eines Antefixmotivs, das im griechischen Mutterland bis nach Makedonien
und daruber hinaus eine auBerordentliche Verbreitung gefunden hat.51 Im kleinasiatisch-
pergamenischen Ambiente ist mit dem Import dieses Typus wahrscheinlich auch die
Ubernahme des Akanthuskelches in der Stirnziegelornamentik verbunden.
48 Archiv der AntikensammlungInv. 37, Nr. 116 (P 715) und Nr. 118 (P 720); auBerdem als weitere
hier zugehorigeFragmente,deren Aufewahrungsortnicht mehr zu ermittelnist: Nr. 119 (Fragmentder linken
oberenVolutemit Palmettenresten,7 cm breit)mit derselbenHerkunftsangabe,die sich auf den Hang zwischen
dem Trajaneumund dem Tempel auf der Theaterterrassebezieht. Bei dem letztgenanntenTempel wurden
zwei weitere FragmentediesesAntefixtypusgefunden: Nr. 121 (linkeSeitenpalmette,7,5 cm breit)und Nr. 149
(Mittelstuckmit linkeroberer Volute, 19 cm hoch).
49 Informationenund Fotosvon diesen noch unpubliziertenNeufunde verdankeich W Raeck.
50 Hubner 1973, S. 71, 121, 127, 143, Taf. 70:6 und in Kerameikos X, S. 233, 236, Abb. 260:a (Z 42); zu
den Erechtheionantefixenvgl. Zuchner 1936, Sp. 323-327, Abb. 14-17.
5' Billot 1976, S. 116-117, Nr. 52 und 52, Taf. 28:e, f; Kaltsas 1988, S. 35, Nr. 68 (ausPella),Taf. 21:b, S. 84
(Typ K V).
264 VOLKER
KASTNER
Das bis zu den ParthenonakroterenzuruckverfolgbareSchema der >gesprengten Pal-
mette< laBtsich im 4.Jh. und 3.Jh. v. Chr. mehrfachnicht nur in der steinernenBauorna-
mentiksondernauch aufStirnziegelnbeobachten(z.B.in Delphi, Olympiaund Kalaureia).52
So wird man den >Rosettentypus< nach seiner Palmettengestaltungdieser Zeit-wie dies
auch schon vorgeschlagen wurde-zuweisen und ihn in fruhhellenistischeZeit datieren
konnen.53
Die in Pergamon gefundenen Antefixe konnen damit und nach den gesicherten
Fundplatzen nur mit einem Dach verbunden werden, das zu einem wichtigeren alteren
Gebaude im oberen Akropolisterrraingehorte. In Frage kame dann wahrscheinlich der
PalastkomplexI in seinem Zustandvor dem Umbau im 2.Jh. v. Chr.
An die Fragmentedes >Rosettentypus<lassensich noch zwei weitere Ziegelbruchstucke
anschlieBen,die das Motiv variierenund deren Fundortnicht mehr zu ermitteln ist. Aber
ausgehend vom Tonmaterial und im Vergleich mit den uibrigenArchitekturterrakotten
unserer Sammlung mochte ich doch annehmen, daB es sich auch um pergamenische
Fundstuckehandelt:
Antefixfragmente
Zwei Stuickeohne Inv. Nr. (Pergamon?) (P1.83:c)
Unterteile eines Palmettenantefixesmit gegenstandigenS-formigenVolutenrankenuibereinem
dreiblattrigenstarkverschliffenenAkanthuskelch.Die oberen Voluten besitzen ein konvexes Auge,
wahrend die unteren spiraligaufgerolltsind und in auBereBlitengebilde auslaufen. Zwischen den
oberenVolutenbefindetsich ein Blutenkelchmit einer kleinenherabhangendenPalmette. Uber dem
Blutenkelchist der Ansatz einer Innenpalmetteerkennbarund rechts daneben strebt das unterste
Blatt des jetzt nur noch im Ansatz erkennbarenauBerenPalmettenflachers nach oben und biegt in
Hohe der Bruchkantenach auBen. Im Zwickelzwischendem Palmettenblattund der Volutenranke
sitzt eine neunblattrigeSeitenpalmette.
Die Fragmentebestehen aus hellrotemTon und sind mit einer dunkelrotenEngobe tiberzogen,
wie sie allgemeinffirdie hellenistischenZiegel in Pergamoncharakteristischist. Der innen gerundete,
auBen dachformigeKalypter(8,5 cm hoch und 16 cm breit)kIuftmit einem kraftigenGrat auf der
Ruickseiteder Antefixplatteaus.
A. Hohe (erh.): 14,6 cm; Breite: 16,2 cm; Tiefe (erh.): 13 cm.
B. Hohe (erh.): 12,2 cm; Breite (erh.): 11,3 cm; Tiefe (erh.):7 cm.
Das Relief dieser Stirnziegel laBt sich offenbar aus der Verkntipfung traditionell klein-
asiatischer Gestaltungsweisen mit dem aus Attika ubernommenen neuen Antefixtyp erklaren.
Die teigig und schwerfaillig modellierten Ornamente verleihen dem Ziegel allerdings einen
altertumlichen Charakter. Nach der groberen Ausfuihrung zu urteilen wird man wohl die
beiden Fragmente als Ersatzstucke fuirbeschadigte Antefixe des >Rosettentypus< eingesetzt
haben.
52 Delphi, marmorneTraufsimades ApollontempelVI (366-320 v. Chr.):Samothrace I, ii, S. 88-89, Abb. 83;
Olympia, nordostlichdes Zeustempelsgefundene Antefixe (>Diadochenzeit<): OlympiaII, S. 143, Taf. 91:6;
Kalaureia,Bau G (1.Halfte 3.Jh. v. Chr.):Wide und Kjellberg 1895, S. 284-285, Abb. 15.
53 Ein bisher unpubliziertesPalmettenfragment,das wahrscheinlichzu einem Ziegel des >Rosettentypus<
geh6ren durfte, wurde im Bau Z 3 (Knigge 1988, S. 93-94) des Kerameikosgefunden. Danach ware dieser
Typ in Athen bereits in der zweiten Halfte des 4. Jh. v. Chr. nachweisbar.
DACHTERRAKOTTEN ZEITAUS PERGAMON
NACHARCHAISCHER 265
' I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
* ~~~~~~~~~~
wo~~~
ZEITAUS PERGAMON
NACHARCHAISCHER
DACHTERRAKOTTEN 267
BIBLIOGRAPHIE
Akerstrom,A. 1966. Die architektonischen Terrakotten Aleinasiens(ActaInstitutAthenienses RegniSueciae XI), Lund
AvP I, ii = A. Conze, StadtundLandschaft (Altertumer vonPergamon I, ii), Berlin 1912
AvP II = R. Bohn, Das Heiligtum derAthenaPolias fikephoros(Altertimer vonPergamon II), Berlin 1885
AvP III, i =J. Schrammen,DergroszeAltarderObere Markt(Altertimer vonPergamon III, i), Berlin 1906
AvP VI = P. Schazmann,Das Gymnasion derTempelbezirk derHeraBasileia(Altiermer vonPergamon VI), Berlin 1923
AvPVII, ii = F. Winter,Die Skuipturen mitAusnahme derAltarreliefs(AlterWimer vonPergamon VII, ii), Berlin 1908
AvP VIII, ii = E. Fabriciusund C. Schuchhardt,Die Inschnften vonPergamon (Altertimer vonPergamon VIII, ii),
M. Frankel,ed., Berlin 1895
AvP IX = E. Boehringer und F. Krauss, Das Temenosfitr denHerrscherkult, "Prinzessinnen Palais"(Altertimer von
Pergamon IX), Berlin 1937
AvP X = A. von Szalay und E. Boehringer,Die hellenistischen Arsenale (AltertimervonPergamon X), Berlin 1937
AvPXI, ii = 0. Ziegenausund G. De Luca, DasAsklepieion, II, Dernordliche Temenosbezirk undangrenzendeAnlagen in
hellenistischer
undfiii~hrmischer
Zeit(AlttimnervonPergamon XI, ii), Berlin 1975
Billot,M.-F. 1976. "Terrescuitesarchitecturalesdu Musee Epigraphique,"Ae)X 31, 1976, A' [1980], S. 87-135
Buschor,E. 1933a. Die Tonddher derAkropolis, II, Stirnziegel,
Berlin/Leipzig
. 1933b. "AltsamischeGrabstelen,"AM 58, S. 22-46
ClarkJ. Th. 1898. ReportontheInvestigation atAssos1882, 1883, New York
Corinth IV, i = I. Thallon-Hill und L. S. King, Decorated Architectural Terracottas(Corinth IV, i), Cambridge,Mass.
1929
Coulton,J.J. 1976. TheArchitectural Development of theGreekStoa,Oxford
Daux, G. 1965. "Chroniquedes fouilles 1964,"BCH 89, S. 683-1007
DidymaI = H. Knackfuss,Die Baubeschreibung (DidymaI), Berlin 1941
ErythraiII = C. Bayburtluoglu,Terracottas in Erythrai(Erythrai II), Ankara 1977
Furtwangler,A. 1906. Aegina,dasHeiligtum derAphaia, Munich
Gardner,E. A., W Loring, G. C. Richards,und W J. Woodhouse. 1892. Excavations at Megalopolis (JHS Sup-
plement 1), London
Harl, 0. 1971. "Studienzu den Stirnziegelnder Peloponnes"(Diss. Graz-Universitat1971)
Hiller, F. 1967. "ZurBerlinerRitzstele,"MarbWPr[1968], S. 18-26
Hubner, G. 1973. "Dachterrakottenaus dem Kerameikosvon Athen,"AM 88, S. 67-143
VOLKER KASTNER
STAATLIcE MusEENzuBERLIN
Antikensamrnlung
Bodestr. 1-3
D-1O178 Berlin
BundesrepublikDeutschland
PLATE 80
a. ZiegelfragmentInv. P 752
4 1
.4~~~~~~~~~~~~~~.
b. TraufziegelInv. P 717
c. P
Traufziegel Inv.
718
KASTNER:DACHTRAKOTTEN
VOLKCER ZEIr AUSPERGAMON
NACHARCHAISCHER
PLATE 81
_4
a. PalmettenantefixInv. P 712
b. AntefixfragmentInv. P 710
W;1r~~~~~~~~~~0
e. FragmentevonBlutenstanden r
aus dem Fundmaterialvon der
Altarterrasse
various palmette antefixes from the Lower City at Assos, but these, like ours, remain
unassigned to specific buildings and consequently resist precise dating.'3 Interestingly
enough, although several of them are typologicallyrelated to those of Ilion, none appears
to have the strutarrangementon the rear.
The most unusual antefix type from Ilion is quite differentfrom those just noted. It
features a bust of Athena in relief against a rimmed, arched background (PI. 84:f). The
goddess wears a crested helmet with horns projectingfrom either side, while a spearhead
appears at her right.14 In the field to either side of her face are three rounded protrusions.
The back preservesthe stump of the usual supportingstrut. Two similar but much more
fragmentarypieces of the type have been found.15 All three come from trenches spread
acrossa distanceof more than a hundredmetersin the Lower City. And all three were found
in mixed fills of Hellenisticto Late Roman date.
Another such antefix is recorded as belonging to the collection of FrankCalvert. Hans
Thiersch in his unpublished, illustratedmanuscriptof this collection documents that the
antefix in question resemblesours but adds an owl to the spearhead at the left of Athena's
head.16 We have no idea where exactly it was found other than generally "at Ilion". But
if nothing else, this variant indicates that more than one mold was involved in production
of the antefixtype. What kind of buildingor buildingsthese antefixesmay have adorned and
what the dating might be remain to be determined.
The tradition of human-headed architecturalterracottasis firmly rooted in Italy, and
indeed Nancy Winter has argued persuasivelythat it emanated from there.17 She has also
pointed out that whereas the notion was transplantedto Northwestern Greece in Archaic
times, it was favored only relatively briefly in that area. In their Italian homeland, by
contrast, human-headed architecturalterracottassurvived,at least in some areas, through
the Hellenistic period.'8 The subject has never been systematicallystudied in the rest of
the Greek world, although sporadic examples have been noted from time to time.'9 A
general impressionbased on cursory observationof unpublishedmaterial suggests that by
Late Hellenistic times antefixeswith human heads, usually small and embedded in vegetal
surroundings,were fairly widespread. But it can also be observed on the basis of isolated
published examples, particularlyin the Black Sea area,20that there is an earlier tradition
of antefixes in certain regions of the eastern Greek world that has heads as the dominant
21
Von Graeve (1970, p. 30, pls. 4:3 and 5:1)sees the antefixesof the AlexanderSarcophagusas unrelatedto
those of the Black Sea region.
22
Contrast the Athena-headed antefixes, where the goddess is more traditionallyhelmeted, collected by
Hubner 1976, pl. 63:1-3 and one from Mesembria(Bouzek 1990, p. 141, fig. 40:1).
23 For a 4th-centuryB.C. horned helmet from Brjastovecin BulgarianThrace, see Trac,no. 114. For Celtic
horned helmets, see Dintsis 1986, p. 156, and for Bactrianrulerswearing horned helmets on coins, Dintsis
1986, p. 79 with note 12.
24
Convenientlyillustratedin Yalourisetal. 1980, no. 150.
25
See PellaI, pp. 39, 51-54. Cows are also particularlyconnected with Athena in Boiotia (see Schachter
1981, pp. 130-134).
26 For discussion,see von Fritze 1902, pp.514-516 and Bellinger 1961, pp. 31 and 54. The ritualsacrificeof
the cow to Athena first appears on coins dated by Bellinger 1961, pl. 4, nos. T80, T81 to the 2nd century
B.C. The fate of the cow is more graphicallyrenderedon coins of Faustinathe Younger (Bellinger 1961, pl. 8,
nos. T166, T167).
27 On the sacrificialcow at the Panathenaicfestivalof Ilion, see testimonia collected by Frisch 1975, e.g.,
nos. 5, 6, and 42.
28
For a discussionof basic Athena types at Ilion, see Miller 1991, pp. 44-45 with notes 18-24.
29
See comments in Pergamon III, p. 172.
TERRACOTTASFROMILION
ARCHITECTURAL 273
BIBLIOGRAPHY
Akerstrom,A.1966. Die architektonischen TerrakottenIeinasiens(ActaInsitid AdeiensesRegniSuedaeXI), Lund
Akurgal,E., and L. Budde. 1956. Voriujiger BerichtuberdieAusgrabwugen in Sinope,Ankara
Bellinger,A. R. 1961. Troy.TheCoins(SupplementaryMonograph 2), Princeton
Blegen, C., C. G. Boulter,J. L. Caskey,and M. Rawson. 1958. TroyIV SettlmentsVIIa,VIb and VIII,Princeton
Bouzek,J. 1990. Studiesof Greek Pottieyin theBlackSeaArea,Prague
Dintsis, P. 1986. Hellnistische
Helne,Rome
Dorpfeld, W 1902. TrqjaundIlion.Ergebnisse derAusgrabungen in denvorhistonschen
undhistonschen SchichtenvonIlion
1870-1894, Athens
Frisch,P. 1975. Die InschriflenvonIion, Bonn
Fritze,H. von. 1902. "Die Manzen von Ilion," in Dorpfeld 1902
Goethert, F. W., and H. Schleif. 1962. DerAhnaempelvonIlon, Berlin
Graeve, V. von. 1970. DerAlexandersarkophag undseineWerkstatt (IstanbukrForschlungen
28), Berlin
Higgins, R. A. 1954. Catalogue of theTerracottas
in theDepartment of Greek
andRomanAntiquities BritishMuseum,repr.
Oxford 1969
Hoepfner,W. 1969. "Zum Entwurfdes Athena-Tempelsin Ilion,"AM 84, pp. 165-181
Hubner, G. 1976. "AntefixaDeorum Athenarum,"AM 91, pp. 175-183
Koch, H. 1912. Dachterrakotten aus Campanien, Berlin
Korfmann,M. 1991. "Troia:Reinigungs-und Dokumentationsarbeiten1987, Ausgrabungen1988 und 1989,"
StudiaTroica1, pp. 1-34
. 1992. "Troia-Ausgrabungen1990 und 1991," StudiaTroica2, pp. 1-41
Laviosa,C. 1954. "Le antefissefittilidi Taranto,"ArchCl6, pp. 217-250
Leaf, W. 1923. StraboontheTroad,Cambridge
Miller,Stella G. 1991. "TerracottaFigurines:New Finds at Ilion, 1988-1989," StudiaTroica1, pp. 39-68
PellaI = D. Papakonstantinou-Diamantourou, IatopmxX entaxon:aet xat Vapcuplat (HIXXca I), Athens 1971
Pergamon III = E. T6pperwein, Terrakotten vonPergamon (Pergamenische Forschungen
III), Berlin 1976
Rose, C. B. 1992. "The 1991 Post-BronzeAge Excavationsat Troia," StudiaTroica2, pp. 43-60
Schachter,A. 1981. Cultsof Boiotia,I, Acheloos to Hera(Buletinof theInstituteof ClassicalStudiesof theUniversfvof
LondonSupplement38.1), London
Schmidt-Dounas,B. 1991. "ZurDatierung der Metopen des Athena-Tempelsvon Ilion," IstMitt41, pp. 363-
415
Thiersch, H. 1902. KatalogderSammiung Calvertin denDardanellnundin Thymbra (Handwrittencatalogue)
Thompson, D. B. 1963. Troy: The Terracotta Figns ofthe HellnisticPeriod(SupplementaryMonograph 3),
Princeton
Traci= Artee culturanele terrediBulgariadale orini alla tardaromanit& (Exhibitioncatalogue,Venice 1989), Milan
Winter,N. A. 1978. "ArchaicArchitecturalTerracottasDecorated with Human Heads," RM 86, pp. 27-58
Yalouris,N., M. Andronikos,K. Rhomiopoulou,A. Herrmann, and C. Vermeule. 1980. TheSearchforAlexander
(Exhibitioncatalogue,Washington/Chicago/Boston/San Francisco 1980-1982), Boston
Zimmermann, K. 1994. "Traufziegelmit Reliefmaanderaus dem Schwarzmeergebiet,"in Proceedings of the
International onGreekArchitectural
Conference of theClassicalandHelenisticPeriods,
Terracottas Athens,December 12-15,
1991 (Hesperia Supplement27), N. A. Winter,ed., Princeton,pp. 221-251
STELLA
G. MILLER
BRYNMAwRCOLLEGE
Department of Classicaland Near EasternArchaeology
Bryn Mawr, PA 19010
PLATE 84
i
r' A-
a.AT 19 b.AT30
c. AT 17 d. AT IO(frontview)
4I,
e. AT 10 (backview) f. AT 21
THE END of Morgantina'sArchaic phase was abrupt.' Most areas of the city, both
intramuraland extramural, show signs of violent destructionby fire. The ceramic
evidence associatedwith these destructionstratadates this event to late in the second quarter
ofthe 5th centuryB.C., and sinceDiodoros(I 1.78.1)tellsus thatthe SikelnationalistDouketios
took Morgantinain 459, it seems reasonableto connect the materialand literaryevidence.2
When Morgantina reappearsphysically,it has been refounded on the Serra Orlando
ridge a short distance to the west of the Cittadella,the site of the intramuralArchaic city.3
The positivefeaturesthat appearat the SerraOrlandosite (thecivic and religiousmonuments
in and around the agora, the domesticstructuresand sanctuariesto the east, west, and north
of the agora),all alignedaccordingto an orthogonalplan, and even the city wallsare products
of an intensiverebuildingprogramperhapsbegun duringthe Timoleontic revivalof Hellenic
Sicily in the second half of the 4th century. This program reached its climax in the second
quarter of the 3rd, when Morgantinawas a large and prosperousoutpost of the kingdom
of Hieron II of Syracuse.4
Between the Archaic and Hellenistic phases, then, the material record at Morgantina
appears to have left two major and potentiallyrelated lacunae. The firstis the issue of the
unknown fate of the population of Archaic Morgantina, and the second is the apparent
absence of monumentalarchitecturedatingto the Classicalphases at either the Cittadellaor
the Serra Orlando sites.
At the Cittadellathere is no evidence for reoccupationuntil the second half of the 4th
century.5The extramuralnecropoleis,however,continued in use into the second half of the
5th century,though at a diminishedlevel in frequencyof use and in the richnessof the grave
offerings.6 In addition, Archaic Morgantina'smost prominentlypositioned and splendidly
1 Thanks are due to the followingindividualsand institutions:Shari Taylor Kenfield, Curatorof Research
Photographs in the Department of Art and Archaeology at Princeton University, for invaluable research
assistance;ProfessorMalcolm Bell Imof the Universityof Virginiaand Directorof the MorgantinaExcavations,
for helpfuldiscussionsas always;Dr. Nancy A. Winterof the AmericanSchool of ClassicalStudiesat Athens, for
arrangingand hosting this conference;the RutgersUniversityCouncil on Research and SponsoredPrograms
and the Departmentof Art Historyat RutgersUniversity,for makingpossiblemy attendanceat this conference;
the American School of ClassicalStudies at Athens, for its generoushospitality.
2 Bell 1988, pp. 313-316; Sjoqvist 1973, pp. 51-52.
3 Perhaps the best complete plan of both the Cittadellaand the Serra Orlando sites can be found in Bell
1981, pl. 1.
4 Bell 1988, pp. 316 and 338.
5 Sjlqvist 1973, pp. 51-52; Sjoqvist 1958, pp. 155-158; Stillwell 1959, pp. 171-173; Allen 1970, p. 378.
6 Lyons, forthcoming, Morgantina StudiesV. See now Lyons, in press, Bolleinodi Archeologia.Thanks to
Dr. Lyons for manuscriptversions of both these forthcomingpublicationsand for many helpful conversations
over the years.
276 JOHN F.KENFIELD
7 Allen 1970, p. 378. For the most in-depth discussionof this building,see Kenfield 1990, pp. 265-274.
8 Sj~qvist 1960b, pp. 292-300; Bell 1981, p.5, note 12; Bell 1988, pp. 314-316. Recent excavationson the
east side of the Lower Agora have also revealed a potter's kiln dating to the second half of the 5th century;
see Bell 1988, pp. 319-321.
9 See, for example, comments by Finley 1968, p. 64.
10 Allen 1970, p. 378, notes 34 and 35.
l Thucydides 4.65 for the Congress of Gela and Diodoros 11.78.1 for the descriptionof the Morgantina
taken by Douketios as apolis axiologos.
12 Diodoros 14.78.7; Sjoqvist 1973, pp. 56-59.
13 Bell 1981, p. 5, note 15; Buttreyetal. 1989, pp. 4-8, pl. 1, figs. 7-10.
14 Bell 1988, pp. 333-334, note 66, fig. 28.
15 Discussionswith completebibliographiesto datesofpublicationcan be found in Ridgway 1970, pp. 12-28;
Holloway 1975, pp. 19-26, figs. 125-15 1.
16 In the lively discussionfollowingthe presentationof this paper,opinionsconcerningthe date of this antefix
as manifestedby its style varied widely,from the 470's to the firstquarterof the 4th century.
THE ARCHITECTURAL
TERRACOTTASOF MORGANTINA 277
fragments the classicism is readily apparent, but the dearth of monumental architecture
dating to the Early and High Classicalperiods led to the assumptionthat these fragments
belonged to a set of classicizingHellenisticfemale-headantefixes.
The discoveryof a nearlycomplete antefixrevealedthat stylisticallythe entire set should
probably be dated soon after Douketios' taking of Archaic Morgantina. This conclusion
seems to entail a number of other assumptions.If, for example, this earlierdating is correct,
then Morgantinamust have been refoundedon the SerraOrlandoridge almostimmediately
after the destruction of the Archaic city, and, since Douketios is known to have founded
and refoundedother cities, it is temptingto add Morgantinato his list.17
With whom, then, did Douketios populate this refounded city? The findspot of this
relativelycomplete antefixsuggestsan answer.Its proximityto the North Stoa I, apparently
the only publicbuildingof Morgantina'sHigh Classicalphase of which there are any remains,
indicatesthat it might have belonged to the roof of that building.18
The scrappyremains of the North Stoa I were discoveredin the excavationsof the east
end of the overlyingNorth Stoa II. 19The strataassociatedwith the North Stoa I contained
coins of Syracuse and Kamarina as well as Attic black-glazedsherds dating to the second
half of the 5th century. The coins and sherds, of course, supply a date for the use of the
building, and the sherdsfurtherindicate that providinga formal settingfor drinkingparties
was one of the principalfunctionsof North Stoa I. The remains of the walls of the building
show that they were about 0.75 m. thick and constructedof stone, mostly rubble, but with
the occasional use of ashlar blocks, especially at the corners. The rooms of the building
seem to have been nearly square in plan and ranged side by side along the northeastern
side of the agora. These rooms do not seem to have communicatedwith each other, but
each apparentlygave onto the open space to the south and perhapsonto a streetto the north.
Each of these featuresof the North Stoa I correspondsto the earlierremainsof a building
on the southeasternside of the Archaic agora, a buildingwhich I have addressedon several
occasionsin other contexts.20The eaves of thisArchaicbuildingwere decoratedsequentially
with two sets of female-head antefixes. The early examples date to the early years of the
third quarter of the 6th century, when a sizable group of Greeks arrived and settled at
the previouslyindigenous site.21 Later,probably in the 470's, the roof was damaged and
17 Diodoros 11.88.6 for Menai and Palike;Diodoros 12.8.2 for Kale Akte. See Bell 1988, pp.319-321; Finley
1968, pp. 63-64. Indeed, the House of the Silver Hoard, an architecturallyundistinguisheddomicile in spite
of the glamor of the eponymousfind of Roman silvercoins depositedin its cistern, is located immediatelyto the
east of the Upper Agora and may date as earlyas the second half of the 5th century. If so, its alignmentsuggests
that the orthogonal plan that determined the layout of the Hellenistic city may have been a feature of the
Douketianrefounding.Bell 1988, pp. 320 and 338, fig. 1;Tsakirgis1983, pp. 42-46, 424. This dissertationhas
been updatedand rewrittenand is to appearas Morgantina StudiesVII, forthcoming,PrincetonUniversityPress.
18 Bell 1988, p. 334, note 66.
19 Sjoqvist 1962, pp. 136-137; Sjoqvist 1964, pp. 138-140; Allen 1970, p. 364.
20 Initialreportson this buildingappearin Sjoqvist 1958, pp. 155-158; Stillwell 1959, pp. 171-173; Sj~qvist
1960a, pp. 133-135; Kenfield 1993b.
21 Kenfield 1993a; Mertens-Horn 1991, pp. 11-12, fig. 3.
278 JOHNF.KENFIELD
repairedwith more up-to-date female-head antefixes.22Clearly there was a desire to keep
the decorationof the eaves of this Archaic civic building as au courant as possible. It seems
reasonableto assumethat this desirewas also appliedto its apparentEarlyand High Classical
successor,the so-calledNorth Stoa 1.23
That a building appears in the early 450's at ClassicalMorgantina, which is more or
less identical in placement, plan, constructiontechnique, decoration, and function to what
was arguablyArchaic Morgantina'smost importantcivic building, suggeststhat Douketios
transferredthe survivingpopulation from the Archaic site to the Classical site.24 Perhaps
it is not accidentalthat the positionof this new building,the North Stoa I, was the best on the
agora for affordinga view of the new civic center immediatelyto the south and a view of
the old city in the distanceto the east.
In the 5th century,with the adventof classicismand its correlativehumanism,a conflation
of forms occurredin the two types of antefixesmost prevalentin the WesternGreekworld.25
The canonical Archaic gorgoneion with its broad, flat face (betrayingits origins in the
graphic arts)and its grimacingmouth, fangs, and protrudingtongue is abandoned for what
is essentiallya female head in much higher relief and more closely related to monumental
freestandingsculpture. Indeed, the only indicationthat this new kind of gorgoneion antefix
reallyrepresentsMedusa is the continued appearanceof flankingsnakes,that feature of the
Archaic gorgoneion which is the last to disappear.The necessityfor the Medusa to inspire
fearin the viewer,thusperformingher apotropaicfunction,is now accomplishednot through
her monstrousappearanceas in the past but throughher wildly demented demeanor.
At first the timeless sobriety of High Classicism is maintained and the dementia is
communicatedthrough the disheveledhair into which the snakylocks of earlier gorgoneia
have metamorphosed. Always in step with development in the more monumental plastic
arts, the face graduallybecomes more mobile, more fleeting in its expression. The lips part,
the eyes are set more deeply into theirorbitalcavities,and the foreheaddevelopsa horizontal
creasethat causesit to projectforwardon two planes. The use of these techniquesenablesthe
sculptorto communicate dementia and thus inspire fear through apparent emotion rather
than through symbol, though the hair remainsinvariablywild and unkempt.
It is interesting that none of these High and Late Classical Medusa head antefixes
appear at Morgantina. In fact, the female-head antefixes under discussion are the only
22 Kenfield 1993b, p. 26; Stillwell 1961, pp. 280-281, fig. 12, pl. 94; Winter 1978, pp. 36-38, pl. 14, figs. 3
and 4; AlbaneseProcelli 1990, pp. 7-16; Mertens-Horn 1991, p. 12. My proposeddate for the antefixesin this
series is considerablylater than the dates proposedby Winter,Albanese Procelli,or Mertens-Horn.
23 It is interestingto note that this civic building on the southeast side of the Archaic agora was, in spite
of its destructionin 459, never furtherdismantledor built over when the area was reoccupiedin the 4th century,
perhaps suggestinga continued reverence for the ruins of the building down into at least the Late Classical
period. See Stillwell 1959, p. 173.
24 There has been much speculationon the specificsof Douketios'program. See, most recently,Rizzo 1970,
pp. 158-169; Adamesteanu 1962. A potter's kiln firing Sikulo-Geometricpottery but associated with Attic
black-glazedsherdsdating to the second half of the 5th centuryhas been discoveredon the southeastside of the
Classical/Hellenisticagora at the Serra Orlando site, thus apparentlydemonstratingthe continued ethnic mix
of the community at the Serra Orlando site. See Bell 1988, pp. 319-321.
25 The best general examinationsof this developmentremain Laviosa 1954, pp. 228, 235-243, pls. LXXII,
LXXIII and Floren 1977, pp. 116-217.
THE ARCHITECTURAL
TERRACOYLAS
OF MORGANTINA 279
These were found dumped into the cistern of the House of the Antefixes, a shabby structureto the roof of
which these beautiful revetments could hardly have belonged. The assumption, then, has been that these
antefixesoriginallydecorated the roof of a nearby and much more splendidlyappointed domicile such as the
House of the Official. See Tsakirgis1983, pp. 152-155; Allen 1970, pp. 361-363.
29 Besides the examples of this antefix discovered in the American excavations at Morgantina, a number
were recoveredby Paolo Orsi and are housed uncataloguedin the Museo ArcheologicoRegionale in Syracuse.
280 JOHN F.KENFIELD
BIBLIOGRAPHY
30 See Tsakirgis1983.
31 Ridgway 1990, pp. 150-154, 180-185 esp. 181, and 315-316, with complete bibliographyto date of
publication.
32 See, for example, Wescoat 1989, p. 110, nos. 34 and 35, the only objects (coins)in the catalogue dated
to the reign of Hieron II; Bell 1981, pp. 43-44. Langlotz 1965, p. 301, no. 155 presents a limestone head
in a baroque style from Megara Hyblaia which he dates to the 2nd century. In light of these Medusa-head
antefixes from Morgantina, it would perhaps be preferableto see this head as an example of the sculpture
of Syracusein the 3rd century.
THE ARCHITECTURAL
TERRACOTTASOF MORGANTINA 281
. 1993b. "AModelled TerracottaFrieze from Archaic Morgantina:Its East Greek and Central Italian
Affinities,"in DelidiaeFictiles:Proceedingsof theFirstInternational on Central
Conference Italc Architectural
Terracottas,
Rome,10-12 December, 1990, E. Rystedt,C. Wikander,0. Wikander,eds., Stockholm,pp. 21-28
Langlotz, E. 1965. AncientGreek Sculpture
of SouthItalyandSicily,New York
Laviosa, C. 1954. "Le antefissefittilidi Taranto,"ArchCl6, pp. 217-250
Lyons, C. L. In press. "Modalitadi acculturazionea Morgantina,"Bolletino diArcheologia
. Forthcoming.TheArchaic Necropolis(Morgantina StudiesV), Princeton
Malkin, I. 1981. "Religionand the Foundersof Greek Colonies"(Diss. Universityof Pennsylvania1981)
Mertens-Horn,M. 1991. "Una nuova antefissaa testafemminileda Akraied alcune considerazionisulle Ninfe
di Sicilia,"BoletinodiArcheologia 66, pp. 9-28
Ridgway,B. S. 1970. TheSevere Syle in Greek Sculpture,Princeton
* 1990. HellenistcSculpture,I, Te Stylesofca.331-200 B.C., Madison
Rizzo, E. P. 1970. La repubbika di Siracusanelmomento diDucezio,Palermo
Sjoqvist,E. 1958. "Excavationsat Serra Orlando (Morgantina):PreliminaryReport II,"AJA62, pp. 155-164
. 1960a. "Excavations at Morgantina (Serra Orlando) 1959: Preliminary Report IV," AJA 64,
pp. 125-135
* 1960b. "Perch6Morgantina?"RedLinc 15, pp. 291-300
* 1962. "Excavations at Morgantina (Serra Orlando) 1961: Preliminary Report VI," AJA 66,
pp. 135-143
. 1964. "Excavations at Morgantina (Serra Orlando) 1963: Preliminary Report VIII," AJA 68,
pp. 137-147
.1973. Syip andtheGreeks, Ann Arbor
Stillwell,R. 1959. "Excavationsat Serra Orlando 1958: PreliminaryReport III,"AJA63, pp. 167-173
1961. "Excavationsat Morgantina(SerraOrlando) 1960: PreliminaryReportV,"AJA65, pp. 277-281
Tsakirgis,B. 1983. "The Domestic Architectureof Morgantinain the Hellenistic and Roman Periods"(Diss.
PrincetonUniversity 1983)
Wescoat, B. D., ed. 1989. Syracuse: TheFairestGreekCit, Rome
Winter,N. A. 1978. "ArchitecturalTerracottasDecorated with Human Heads," RM 85, pp. 27-58
JoHN F. KENFiELD
DEPARTMENTOF ART HISTORY
RutgersUniversity
Voorhees Hall
Hamilton Street
New Brunswick,NJ 08903
PLATE85
4~~~~~~~~~~~1
Cd~~~~~~~~~~~~~~~~~~~~~~~~~~
I 6
Cdt
- dnC 3
Cd
$It~ ~~~~~I a -w CU
CO~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.
14~~~~~~~~~~~~~~~~~~~~"
CU
U~~~~~~~~~~~~~p
LE TERRECOTTEARCHITETTONICHE
DI ELEA-VELIADALL'ETAARCAICA
ALL'ETAELLENISTICA
(PLATEs86-88)
-s"- -----.-
C.)
4.)
"0
4.)
'4
I!
LE TERRECOTTE ARCHITETTONICHE DI ELEA-VELIA 285
trovato poi in M. Napoli, nel corso degli anni '60, un entusiastae attivo ricercatore,teso
ad esplorareil tessutourbano della cittA,la sua estensioneed i suoi rapporticon 1'interno.10
Degli scavi realizzatiin quegli anni rimangonole analisitopograficheed urbanistichedi
M. Napoli"1 (Fig. 1) e, piui recentemente,un quadro di sintesidi W Johannowsky,12mentre
del tutto inediti sono a tutt'oggii materialie i loro contesti di provenienza.
Accanto dunque ad un dibattito storico sempre piui approfondito e ad un quadro
urbanistico via via piiu chiaro, grazie anche alle sistematiche indagini che l'Istituto di
Archeologia dell'Universitadi Vienna va conducendo,13 va annotata questa carenza nello
studio e nella sistemazionedei materialivelini.
Da ci6 e scaturitapertanto la necessity di un ampio programme di studio avviato in
questi anni dalla SoprintendenzaArcheologica di Salerno-Avellino-Benevento,contempo-
raneamentealla ripresasistematicadell'esplorazionedella citta antica.14
Uno dei primi lavori in via di realizzazione interessa le terrecotte architettoniche
rinvenute nel corso degli scavi tra gli anni '30 e gli anni '70, sollecitato proprio dalla
necessity di fornire materialiediti al dibattitorelativo alla comparsadi particolaritipologie
della decorazione architettonicafittile nell'area del basso e medio Tirreno nella seconda
meta del VI sec. a.C. 5
II material e stato organizzato per classi tipologiche dal momento che purtroppo,
soprattuttoper i pezzi provenientidagli scavi piNu vecchi, mancano quasi del tutto i dati di
provenienza. Anche nei casi in cui e stato possibile individuarei contesti di provenienza
o addirittura le quote di profondita del ritrovamento, i dati non sono mai sufficienti
per attribuirele terrecotte architettonicherecuperate ad un determinate edificio ne per
ricostruiredei "sistemi"decorativecompleti. Si pu6 solo ipotizzarela presenza di una serie
di tetti con coperturadi tegole e coppi, decorati, ai lati terminali,da elementi di protezione
fittile, di cui rimangonofrustulipiNu o meno ricomponibili.
ETA ARCAICA
10
Napoli 1966, pp. 191-226; Napoli 1970, pp. 226-235.
l Napoli 1969, pp. 151-163.
12 Johannowsky 1982, pp. 225-246.
13 Krinzinger 1987, p. 19.
14 Tocco 1991.
15 Winter 1978, pp. 27-57; Kastner 1982a; Knoop 1987; BonghiJovino 1989, pp. 666-682; Greco 1990a,
pp. 59-63.
286 GIOVANNA GRECO E MARIA JOSE STRAZZULLA
secondo una tradizione orientale, peraltro confermata dal fatto che la documentazione
archeologicanon attestauna particolareabbondanzadi tegole e coppi di etAarcaica.16
Accanto a questi edifici ve ne erano tuttavia altri che, pur conservando lo zoccolo in
muraturapoligonale e l'elevato in mattoni crudi, avevano un tetto pesante con tegole piane
a bordo rialzato, coppi decorati nella parte terminateda antefissesemicircolarie tegole di
sponda con la faccia iposcopica dipinta. Questi tetti riflettono una tradizione decorative
peculiar dell'areatirrenicala quale si innesta su una tecnica struttivapuramenteionica.17
Le antefissearcaiche sono a lastrasemicircolarecon nimbo baccellato e presentano tre
diversi motivi decorative:palmetta dritta, palmetta pendula, testina femminile. Esse sono
pertinentiad almeno quattro tetti differenti,mentre le variantiindividuateper l'antefissaa
palmettapendulapossono apparteneresia a riparazionidello stessotetto, sia a picitettidiversi.
Le antefissesono tutte a stampo, con ritocchia stecca e con una ricca policromia. Anche
ad un semplice esame autoptico sono distinguibiligli esemplari realizzati in argilla locale
da quelli invece ricavati da un'argilladel tutto differente,di cui'soltanto le analisi in corso
potranno precisarel'area di provenienza.
Anteftsaa palmetterittaentronimbobaccellato(P1.86:a). La lastra,semicircolare,e decorata
con una palmetta a nove lobi convessi, rigidi e allungati, fuoriuscenti da un elemento
semicircolaredi base; un nastro piatto segna la cornice liscia, terminante con un bordo
dentellato,sulla quale era dipinta la decorazionebaccellatain colore bruno e crema.'8
L'antefissaripete esattamente un tipo attestato a Cuma, Ischia e Capua e si inserisce
perfettamentenella seriazione di recente operata da Knoop, il quale individua in Ischia e
Cuma i centri propositiviper questo tipo, collocando cronologicamentela palmetta dritta
senza base (tipo 4) al secondo quarto del VI sec. a.C. e considerandolal'antecedente del
motivo a palmettapendula.19
La discussion sulla priority o meno del tipo di antefissacon palmetta dritta, ripresa
recentemente anche dalla BonghiJovino,20 poggia purtroppo su seriazioni teoriche o su
analisistilistiche,mentre mancano dati archeologicistratigraficisicuramenteaffidabili.
II problema e strettamenteconnesso a quello della nascita del nimbo, considerate dalla
maggioranzadegli studiosiun apportodel mondo greco-orientaleper il confront con alcuni
esemplarinoti da tempo, provenienti da Neandria, Taso e Gordium, che presentano una
cornice dentellata.21
Va tuttavia osservato che antefissesemicircolarisono ben attestate anche nella Grecia
continentale: l'esempio piu"noto e calzante e quello di Sparta, dipinto con un motivo
elicoidale circondato da una cornice liscia dipinta a denti di lupo.22 Antefisse circolari
senza nimbo sono presenti in Sicilia23e in Etruria24gia nel primo quarto del VI sec. a.C.
16 Martin 1970, pp. 93-107; Morel 1970, pp. 131-145; Neutsch 1970, pp. 146-152.
17 Knoop 1987, pp. 186-187.
18 N. inv. 43531. Alt. 13,5; largh. 21,2; argillanon locale.
19 Knoop 1987, p. 139, con bibliografiaprecedent.
20
BonghiJovino 1989, p. 672; Kastner 1982b, pp. 90-100.
21
Akerstrom1966; Greco 1990b.
22 Winter 1990, pp. 13-32, pl. 1:c.
23 Cfr. infra,nota 26.
24 Wikander 1981, pp. 124-125.
LE TERRECOTTE ARCHITETTONICHE DI ELEA-VELIA 287
e durano per un lungo periodo, sino a sviluppareuna grande variety di tipologie in eta
ellenistica.25Consideraredunque la semplice antefissasemicircolarecon palmetta il primo
anello all'originedel tipo puo essere fuorviante,cosi come non sempre e chiaro il rapporto
tra motivo "a rilievo"e motivo "dipinto".
E opinione diffusache il "dipinto"preceda il "plastico"e che pertanto anche la cornice
semplicemente dipinta sia l'antecedente di quella con foglie plastiche. In realty, anche
solo sfogliandoi classicirepertoridi terrecottearchitettonichea nostra disposizione,risulta
evidente come, soprattuttonelle fasi iniziali della sperimentazionee della produzione di un
tipo, il rapportotra "rilievo"e "pittura"sia piuttostoarticolatoe mai costrettoin una rigida
succession di "prima"e "dopo". Basti pensare, per limitarsiall'areagreco-occidentale,ai
materialicumani dove si alternanonello stessolasso cronologico elementi a rilievo o dipinti;
e ad Agrigento e Gela, dove antefissedipinte con palmette coesistono con palmette a rilievo
nel corso della metAdel VI sec. a.C.26
IItentativedunque di porre in uno svilupposerialel'antefissasemicircolarecon palmetta
drittadipinta,poi plastica,a cui si aggiungela cornice primadipintae poi plasticae da cui ha
infine origine il motivo della palmetta pendula con cornice plastica non sembra, allo stato
attuale delle ricerche, suffagato da elementi probanti e puo considerarsisolo un'ipotesi di
lavoro.27
Ad Elea i due tipi coesistono e si collocano entrambi nella seconda metA del VI sec.
a.C., cosi come anche a Pompei, nella decorazione del tempio di Apollo, l'antefissa a
palmetta dritta, in ben quattro varianti e repliche, con nimbo, e datata alla fine del
VI sec. ed e considerate contemporanea al tipo con palmetta pendula.28 Le antefisse
pompeiane sembrano apparteneretutte, per la presenza del listello di base e del nimbo
plastico aggettante, ad una variante leggermente piiurecente rispetto all'esemplareeleate,
che viceversae privo di base, con cornice liscia a curvatureappena accennata.
L'antefissadi Elea, realizzatacon ogni probabilitycon argilladi area vesuviana, si data
ai primi decenni di vita della cittA(540-520) e trovaun confront particolarmentestretto,sia
per modulo che per elementi decorative,con un esemplareproveniente da Capua, datato
dalla Bonghi al secondo quarto del VI sec. a.C.29 e daJohannowskynel terzo quarto dello
stesso secolo.30 In mancanza di elementi oggettivi,non e possibiledecidere se la cronologia
del tipo debba essere appiattita sul dato eleate, oppure se esso sia da considerate il caso
attardatodi modello giAformato nei decenni centralidel VI sec. a.C.
In questa sede interessasottolinearedue riflessioniscaturitedall'analisidell'esemplare
eleate: il tipo a palmetta dritta coesiste e si alterna con quello a palmetta pendula ancora
nei decenni finali del VI secolo, mantenendo il "dipinto"quando il "plastico"e pienamente
affermato;ad Elea l'esemplaree giunto insieme alla prima generazione di coloni ed e un
prodotto di officina campana al quale vanno associate le piuAantiche tegole di sponda,
anch'essedipinte, con ogni probabilityprovenientidalla stessaofficina. Non e da escludere
25 Laviosa 1954, pp. 214-250.
26 De Miro 1965, p. 74, fig. 3; Fiorentini 1977, p. 109 tav.XXV:5; Andren 1940,
pp. CLXII-CLXIII.
27 Kastner 1982a, pp. 23-30.
28 De Caro 1986, pp. 37-38.
29 BonghiJovino 1989, p. 672.
30 Johannowsky 1983, p. 74.
288 GIOVANNA GRECO E MARIA JOSE STRAZZULLA
che qui esso abbia costituito un prototipo da cui sara stata forse ricavata una serie locale,
purtropponon attestatadai frammentia nostradisposizione.
Antefssaa palmettependulaentronimbobaccellato (P1.86:b). La lastra semicircolarepoggia
su una fascia piana dipinta; il motivo centrale e costituito da una palmetta capovolta a 7
lobi fuoriuscenteda un nucleo centrale a bottone plastico; essa e retta da una coppia di
volute dispostein senso orizzontalee legate tra loro da un elemento lanceolato che funge da
riempitivo nel punto di congiunzione. II nimbo e costituito da 17 lobi appena incurvati,
terminanti a giorno, contornati da un sottile bordo a rilevo e poggianti su un cordolo
plastico.31Sono attestatedue variantidistinteper modulo e resa plasticae pittorica,mentre
una variantepiuA tarda(A II c) e caratterizzatadallapalmettaimpressaal negativo sullalastra.
Il tipo e notissimo, attestato da una ricca serie cumana32e presente, con le medesime
caratteristicheall'internodella Campania, nel Lazio meridionalee in Sicilia.33Purtroppoi
materialicumani, ischitani34e capuani,prividi contestodi provenienza,non possono fornire
elementi precisiper l'inquadramentocronologicodel tipo, mentre puntualizzazionimigliori
vengono dalle recenti analisisulle terrecottearchitettonichedi Satricum,Pompei e Himera.
A Satricumla sostanzialeintegritydel sistemadi coperturaconsentea Knoop di attribuire
l'antefissaa palmettapendula entro nimbo al terzo venticinquenniodel VI sec. a.C., mentre
G. Colonna tende a fissareintorno al 520 la seconda fase decorativedel tempio.35
A Pompei alcuni esemplariprovenientidagli scavi di A. Maiuri nell'area del tempio di
Apollo vennero datati dallo studiosoagli inizi del VI a.C.36Una recente sistemazionedi tutti
i materialidel tempio ha permesso a S. De Caro di datarele terrecottearchitettonichenegli
ultimi decenni del VI sec., attribuendoleal periodo di massimo splendor del santuario,37
in associazionecon ceramiched'importazioneche ne confermanola cronologia.
A Himera esemplaridi antefissea palmettapendula provengono dagli scavi nell'abitato
e nell'areadei templi C e D e vengono datatea partiredal terzo venticinquenniodel VI sec.38
Alla diffusaopinione che segna dunque come limiti cronologici per questo particolare
tipo di antefissagli anni tra il 550 e il 520 a.C. si oppone la BonghiJovino, che lo consider
un'elaborazionepiui'antica che "non scavalca il limite cronologico della metA del VI sec.
a.C."39 L'ipotesi della Bonghi poggia sul presupposto che l'antefissa nimbata con testa
dedalica costituisca il terminusante quemper le sequenze delle antefisse a palmetta; dal
3' Alt. 25,8; largh. 30; alt. base 4,7; largh. base 22,5; alt. lobo 5.
32 Cfr. Scatozza 1971, pp. 45-111.
3 Per la carta di distribuzionedel tipo cfr. Cristofani1987, p. 107; Knoop 1987, p. 96.
34 Da Ischia provengono alcuni esemplaridi antefissaa palmetta drittae a palmetta capovolta entro nimbo
spesse volte citati dagli studiosi, ma purtroppo inediti. Nel piccolo museo di S. Restituta sono esposti due
esemplari,di cui uno a palmetta drittaed uno a palmetta capovolta,provenientidagli scavi di Lacco Ameno,
ma privi del relativocontesto. I due esemplari,che rientranoperfettamentenella tipologia cumano-capuana,
sembrano tuttavialeggermente piu recenti e sono certamente di produzione locale. Su questi materiali 6 in
corso uno studio da parte di L. Cicala che ha discusso una tesi di laurea sull'argomento,presso l'Universita
di Salerno.
35 Colonna 1984, pp. 396-397.
36 Maiuri 1949, pp. 120-121.
37 De Caro 1986, p. 33.
38 Bonacasa 1970, p. 222, fig. 12; Epifanio 1977, p. 164, tav. XLIX:1.
39 BonghiJovino 1989, p. 674.
LETERRECOTTEARCHITEITONICHEDI ELEA-VELIA 289
momento che la prima viene datata prima della fondazione di Elea, queste ultime non
possono scendere al di lAdella metAdel secolo.
Se proviamo pero ad affrancarela tipologia dell'antefissacon nimbo, in questa prima
fase di sperimentazione,dal rigido schema seriale che costringeovviamente a determinate
successivepassaggidi una derivazione,si nota come e solo la comparsadel nimbo l'elemento
piuinuovo e dirompente rispetto a tipologie gia circolanti tanto nella Grecia continentale
quanto in quella orientale e occidentale. Sostanzialmentei motivi decorative(palmetta
dritta,palmettapendula e testafemminile)seguono un percorsoquasi paralleloe in molti casi
coesistononello stessolassocronologico.40Saranecessariodunquespostareil nostrointeresse
sul problema della comparsadel nimbo e sull'areadove tale innovazione e avvenuta.41
Ad Elea l'antefissaa palmettapendulacoesiste,come giAsi e detto, con quella a palmetta
drittae costituisceanch'essaun portatodella officinecumane. E interessantenotare come in
succession ad esemplariimportaticompaiano quelli in argilla locale che documentano la
ripetizionedel modello (varianteC). Una verificacronologicadi notevole interesseproviene
dai recentisondaggistratigraficirealizzatidall'IstitutoArcheologicodi Vienna.42 Nella zona
bassa della citta e stata individuata,negli stratipiuiprofondi a contatto con la sabbia sterile,
una serie di struttureabitative caratterizzatedallo zoccolo in opera poligonale a giunture
curve e dall'elevatoin mattoni crudi, con un tetto pesante rinvenutoin crollo: oltre a tegole
ed a coppi semicircolari,sono stati recuperateesemplariquasi integridi tegole di sponda con
tracce di pitturaed alcuni frammentidi nimbo pertinent al tipo II b dell'antefissaa palmetta
pendula. L'associazionecon frammenticeramicisia d'importazioneche prodottilocalmente
consente una datazione delle struttureagli anni tra il 540 ed il 520 a.C.
Antefissaa testafemminile
entronimbobaccellato(Fig.2). La testa poggia su una fasciapiana di
cui rimane parzialmentel'attacco. II busto femminile, frontale, con capelli a bande sulla
fronte e treccioline sul collo, e incorniciato da un nimbo a 17 lobi leggermente incurvati
e poggianti su un cordolo plastico. Una ricca policromiane completa la decorazione.43
L'esemplare,noto da tempo, aveva circolatonei testi scientificisolo come frammentodel
busto e della testa.44 Nella recente ricognizione nei magazzini di Velia abbiamo avuto la
fortunadi recuperatetre frammentidel nimbo con attacco, che hanno consentitoil restauro
e la ricostruzionecompleta del tipo.45
L'antefissa non pone grossi problemi di inquadramento cronologico e tipologico,
trovando confronti molto stretti con esemplari cumani e capuani,46 mentre si discosta
leggermente da quelli satricani. II Kastner data l'esemplare velino intorno al 530 a.C.,
sulla base di convincenti confronti con l'acconciaturadei capelli delle korai dell'Acropoli,
40 Cristofani1987, p. 106.
41 Greco1990b.
42 Ringrazio il prof. F. Krinzingerper avermi illustratogli ultimi risultati,inediti, delle sue ricerche. Una
nota preliminarein Tocco 1991.
43 N. inv. 43130 + 86397 + 86395. Alt. att. 24,5; largh. ric. 34.
4 Sestieri 1955, n. 2174; D'Agostino 1974, p. 197, tav. 74; Riis 1981, p. 18, tav. 1, 5E (erroneamentedetta
da Capua).
45 I frammentiprovenivanotutti dal riempimentodel grande muro di terrazzamentorealizzatoal momento
della costruzionedel tempio sull'acropoli:cfr. Morel 1970, p. 144.
46 Koch 1912, taw. VIII, IX.
290 GIOVANNA GRECO E MARIA JOSE STRAZZULLA
52 "I Focei".
53 Ridgway 1984, pp. 134-135; Buchner 1985, pp. 82-83.
54 I1 nimbo, trasposizioneplastica delle baccellature dipinte di tradizione orientalizzante, puo essere as-
sunto come modello dei processi "combinatori"tipici delle officine italiote. L'ipotesi che esso costituisse
un'elaborazionedella piuanota palmetta e statagia espostain Greco 1990b.
55 Knoop 1987, p. 186.
56 Colonna 1980-1981, p. 160; Frederiksen1984, pp. 174-179; Johannowsky 1969, p. 280.
57 Lepore 1967, pp. 151-170.
58 Pugliese Carratelli 1966, pp. 155-163.
292 GIOVANNAGRECO E MARIAJOSEtSTRAZZULLA
L'ETA CLASSICA
G. GPEco
ETA ELLENISTICA
ANTEFISSE
64 Sulle attivitaedilizie nella citta nel III e nel II sec. a.C. cfr. Johannowsky 1982, pp. 236-238; sulla vitality
del porto nel II sec. a.C., BencivegnaTrillmich 1984, pp. 20-33; Leiwo 1985, pp. 494-499 circa la presenzadi
negoiatoreseleati a Delo.
65 Schleuning 1889, pp. 184-190; Mingazzini 1954, pp. 21-60; Gallo 1966, pp. 366-377; Johannowsky
1982, pp. 235-237.
66 Johannowsky 1982, pp. 235-236. Tale osservazionesi potrebbefacilmenteestendereanche ad altre sigle,
quali AlL. oppure HPM.
294 GIOVANNA GRECO E MARIA JOSE STRAZZULLA
/ / \~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I -I I
I ~ ~~~
t, / I ~ ~~~~~~~~~~
,~~~~~~~
FIG.
68
3: Antefissaellenlisticaa palmettaritta e nimbo baccellato
67
Se ne conservano7 frammenti,pertinentia due diversevarianti. L'altezzaricostruitadel tipo e di cm. 25.
L'argillae locale.
68 Supra,n. 59-60.
DI ELEA-VELIA
LETERRECOTTEARCHITETTONICHE 295
dell'ambienteitaliotasin dagli inizi del V sec. a.C.69e si fa evidente nel corso del IV sec. a.C
anche in centri campani, come Frattee Poseidonia,70quale esito di un processo di parziale
penetrazionedi elementi lucani.71In questachiave risultadi estremointeresseriscontrarelo
stesso fenomeno in un centro come Elea, che tradizionalmenteviene considerate, insieme
a Neapolis, una sorta di roccaforteculturaledella grecita contro le pressioni dei Lucani, i
quali nel corso degli ultimidecenni del V secolo avevanogia occupato la vicina Poseidonia.72
Un esempio opposto e riscontrabileanche nel caso di un'antefissa apalmetta,conbordoa dec-
orazione (P1.86:d; Fig.4), che viceversasi presentacome una ripresaparticolarmente
lenticolare
elegante degli antichi motivi.
69 Popolianellenici,pp. 76, 129, tavv. 22, 56; Greco 1977, pp. 131-146; Mazzei 1981, pp. 17-39; Greco
1982, pp. 74-75, figg. 18, 19.
70 Greco 1990a, p. 73, figg. 95, 96.
71 Cfr. Zancani Montuoro 1965-1966, pp. 36-37; Lepore 1967, pp. 194-215; Cordano 1971; Pugliese
Carratelli1973, pp. 3-6; Pugliese Carratelli1974, pp. 7-12; PontrandolfoGreco 1979, pp. 36-37.
72 Per una discussion del passo di Strabone(6.1.2)relativoalle lotte contro i Lucani e i Poseidoniati,Lepore
1966, p. 263.
296 GIOVANNA GRECO E MARIA JOSE STRAZZULLA
672; tav. 125, nn. 808-809; tav. 134, nn. 880 (per il motivo con perle e astragali);tav. 84, n. 433; tav. 89,
nn. 488, 495; tav. 102, n. 583; tav. 104, n. 634; tav. 118, nn. 737-739; tav. 119; tav. 122; tav. 123, nn. 792-795;
tav. 139, n. 923; tav. 141, n. 928 (motivodel kyma ionico).
75 Cfr. il medesimo motivo di perle e astragaliinsieme al kyma ionico in area metapontina, sia su alcune
terrecottearchitettonicheche nelle arule: Lo Porto 1966, p. 151, fig. 11 (sima),p. 154, fig. 14 (arula);Mertens
1974, tav. 53 (tetto B, tetto A); Mertens 1983, tav.XXV:c (tempioionico).
76 Sono stati raccolti piu di 20 frammenti,pertinentiad altrettantiesemplarie provenienti sia dall'acropoli
che dalla parte bassa della citta (scavi 1935, 1953, 1973). In tutti gli esemplaril'argillae locale.
DI ELEA-VELIA
LETERRECOTTEARCHITETTONICHE 297
77 Si vedano ad esempio alcune antefisseda Assos: Akerstrom 1966, pp. 13-15, tav. 7:7, 8:1-3, di incerta
datazione.
78 Cfr. Laviosa 1954, pp. 249-250, tav. IXXVIII:2-4; Pensabene e Sanzi Di Mino 1983, p. 136, n. 247,
tav. 65, dove si menzionano anche alcuni esemplariinediti dalle terme di via S. Aloe a Vibo Valentia.
79 9 frammenti, pertinenti ad almeno 7 pezzi diversi. Nella variante A l'altezza e di cm. 15, la larghezza
16. L'argillae locale.
80 Koch 1912, p. 29, fig. 38, tav.V:2, 4; Bedello Tata 1990, p. 104, tav. XV:1, 2.
81 Per la testina cfr. Bedello Tata 1976, p. 68, tav. XVIII:1, tipo A XXIII A 1, riproducente, sia pure in
dimensioni maggiori, un busto con caratteristicheanaloghe. Le figurine di eroti trovano confronti generici
nel material capuano edito. Si veda ad es. MollardBesques 1986, p. 9, tav. 6:d, p. 10, tav. 7:e, p. 11, tav. 8:d.
82
N. inv. 20089; alt. 14; largh. 19; argillalocale.
298 GRECOE MARIAJOSESTRAZZULLA
GIOVANNA
a quattropetali (P1.87:b). Tanto la forma, quanto la decorazionecentraletrovano riscontro,
piu che non in altreterrecottearchitettoniche,nellaproduzionedi aruleche, nel corso del IV
e del III secolo a.C., appare caratteristicadel centro campano.83
SIME
Di particolareinteressetra il material eleate di etd ellenisticae inoltre la presenza di
un cospicuo gruppo di sime fittili.
I frammentisuperstitisono pertinentia non meno di 10 tipi diversi,distintil'uno dall'altro
sia per le dimensioni che per il profilo e gli elementi decorative,che ne rendono probabile
1'attribuzionead un arco cronologico abbastanzaampio e ad una serie di edifici differenti
tra loro. Ciononostante sembra possibile, almeno per la maggior parte di essi, enucleare
alcune caratteristichecomuni che costituisconol'unico elemento valido, nella quasi totale
mancanza di dati di scavo, per un inquadramentodi massimanell'ambitodella produzione
fittiledel centro.
In almeno 7 dei 10tipi che sono staticlassificatil'altezzaoscillatracm. 17 e cm. 20. IIpro-
filo e generalmentebombato e la sima termina superiormentecon una fasciapiana frontale,
risultandopriva della tacca per 1'inserimentodel coronamentosoprastante,che caratterizza
invece pezzi analoghiin ambientelaziale ed etrusco. La decorazionee costituitadi regola da
palmette alternatea fiori di loto, collegati tra loro da motivi a spirale(P1.88:a). In un unico
caso e sicuramentedocumentatala presenzadi un doccione passante,probabilmentea pro-
tome leonina.84Altri tipi presentanoun motivo di palmette alternatetra loro e differentiate
nella disposizionedei lobi (P1.88:b), oppure inframezzatea tralci naturalistici(P1.88:c).
In linea di massimasembrapossibileaffermareche le sime di Elea-Velia,sia per struttura
che per motivi decorative,appaiono collegate ad una tradizionefittilesquisitamentemagno-
greca, che trova i suoi precedenti nella numerosa serie di sime, prevalentementeutilizzate
nella decorazione templare,attestatein siti quali Crotone, Caulonia e Metaponto.85
Pi in particolare potremmo registrar l'influenza della vicina cultura poseidoniate,
nell'adozionedel profilo a kyma recta verticalecon una larga e prominente curva inferiore,
qui attestatosino dallafine del VI secolo a.C. nella sima litica del tempio di Athena86e desti-
nato a mantenersia secoli di distanzanella sima in terracottadel tempio italico del Foro.87
In definitive, allo stadio attuale della nostra conoscenza, le sime di Elea, insieme
con il tipo di antefissa semicircolare a palmetta che abbiamo sopra ricordato (Figs. 5,
6), sembrano rappresentarequanto di piuistrettamentelegato alla cultura magno-greca il
centro abbia prodotto nel campo della decorazionearchitettonicafittile. Esse risultanoassai
diverse dagli esemplaritardo-classicied ellenisticidi area campana di Fratte di Salerno,88
83 Perle aruledel Museo Campano si veda Bedello Tata, Casolo, e Baroni 1990, pp. 58-69, taw. 10-15. PiA
in generateVan Buren 1918, pp. 15-53, in part. p. 40, n. 4, per un tipo simile alla nostra antefissa;Jastrow
1938, pp. 1-28; Jastrow 1946, pp. 67-80; Ricciotti 1979.
84 N. inv. 20095, tipo C II, alt. 18,5.
85 Per il material di Metaponto cfr. Mertens 1974, pp. 216-224, taw. XLIX, LII, LIII; per Crotone,
Mertens 1986, pp. 200-206, 213-221.
86 Cfr. Shoe 1952, pp. 171-173, tav.XXX:3, 4.
87 Shoe 1952, tav. XXX:5: cfr. Krauss 1939, pp. 14, 35, taw. 7, 34:2, 35:1.
88 Greco 1990a, p. 66, figg. 71, 72.
LETERRECOTTE
ARCHITETTONICHE
DI ELEA-VELIA 299
Pompei, Foro triangolare89e di Capua, caratterizzatidalla costante di un inserimento tra
le palmette di testine femminilicon berrettofrigio o arricchitedi volute e viticci.90Le forme
capricciose delle palmette e l'eleganza naturalisticadei petali dei fiori di loto consentono
un inquadramento generico della maggior parte di questi tipi nel corso della media etA
ellenistica(III-II sec. a.C.): maggiormentesignificativein questo caso e il confront con una
sima a piccola gronda leonina, decorata con palmette e fiori di loto, sempre da Fratte,9' la
cui probabilecronologia(IIIsec. a.C.) trova confront nei dati stratigraficiemersi da recenti
scavi a Pompei.92
Un diversoinquadramentocronologico,cosi come un differenteriferimentoambientale
deve essereproposto per l'ultimotipo di sima considerate(P1.88:c).
La sima, a profilorettilineo,e limitatasuperiormenteda un kymaionico abbastanzarego-
lare soprauna filadi dentelliquadrati.IIcampo decorativee occupatoda un motivo di grandi
palmette a 7 petali, quellacentralea petali introflessi,le due lateralia petali alternativamente
introflessied estroflessi.Le palmette fuoriesconoda un calice stilizzatoe piatto e sono alter-
nate ad elegantitralciondulatida cui pendono piccoli fruttidi melograno. IIritrovamentodi
un esemplareintegro permette di escluderela presenza del doccione leonino.93
Le caratteristichegeneralidel tipo (profilorettilineo,bordo dentellato,forma inorganica
della palmetta) riportano ancora una volta all'ambientecampano e trovano confront in
alcuni esemplari di Capua94e, soprattutto,di Pompei, databili nel corso del I sec. a.C. e
della prima eta imperiale:95in particolaresi veda la sima a doccione leonino, decorata con
motivo di palmette e tralcidi acanto che fuoriesconodai lobi, anch'essacon profilorettilineo
e bordo superior a dentelli,96o ancoraquella con doccione a forma di mascherae palette
di forma molto simile a quella del nostro esemplare,dai cui petali fuoriescono sottili tralci
vegetali.97
ACROTERI
89
Le terrecottedel tempio dorico del Forotriangolaredi Pompei sono ancorainedite. Una descrizione,priva
purtroppodi apparato fotograficoin Richardson,Jr. 1974, pp. 281-290, con una datazione eccessivamente
bassa, al II sec. a.C.
90 Bedello Tata 1978, pp. 210-218, taw. IXXXI, IXXXII.
91 Greco 1990a, p. 69, fig. 84, con ampia discussion sull'evoluzionedella sima.
92
Bedello Tata 1984, p. 251, tav. 141:4 (anterioreal II sec. a.C.).
93 Alt. 23,5, largh. 51. Numerosi frammentiprovengono dallo scavo dell'abitatonel quartieremeridionale
della cittA.
94 Koch 1912, tav. XXIII, XXIV; De Franciscis1952, p. 311, fig. 3.
95 In mancanza di un'adeguata edizione del material il punto di riferimentorimane ancora la vecchia
edizione di von Rohden 1880, pp. 6-15, figg. 3, 4, 6, 7, 9, 10; taw. VII-IX.
96 Von Rohden 1880, p. 32, tav.VII:1, dalla casa dei Niobidi (alt. 32).
97 Von Rohden 1880, p. 14, fig. 9 (alt. 22).
300 GIOVANNAGRECOE MARIAJOSE STRAZZULLA
di una scriminaturacentrale. Inferiormenteil pezzo presentauna terminazionearrotondata,
a sezione obliqua, che doveva formarela fasciapiatta di contorno dell'acroterio.98
L'utilizzazionedi acrotericircolaria decorazionedella testatadel columen, ampiamente
documentata in Sicilia e Magna Grecia, cosi come in ambiente campano sin dall'eta
arcaica,99si prolunga in eta classica ed ellenistica: oltre allo splendido pezzo da Fratte di
Salerno con Eracle in lotta col leone nemeo, probabilmenteeseguito da un'officinaitaliota,
che ne costituisce l'esemplare di maggior rilievo,100vanno ricordati alcuni tipi capuani,
con testa femminile coperta da berretto frigio e a maschera, quest'ultimo di esecuzione
estremamentecorsiva.'I'
La cronologia del nostro esemplare, unicamente basata sui pochi element che si
possono trarredal trattamentostilisticodella barba, a boccoli corposi e resi con un rilievo
profondamente chiaroscurato,sembrerebbecircoscrivibiletra la fine del IV ed il II seco-
lo a.C. II diametro ricostruibile(circa cm. 40) ne consente l'attribuzioneipotetica ad un
edificio di medie dimensioni.102 Altrettanteincertezze permangono per quanto riguarda
l'identificazionedel soggetto: a livello di pura ipotesi potremmo pensare ad una ripresa
del tipo dell'Acheloo, ampiamente documentato in area campana in eta arcaica,103nella
versioneumanizzatacon fluentebarbaa boccoli che si affermanell'iconografiadella divinity
sin dalla fine del V secolo.104
Un secondo frammentoe pertinentead un grossomuso di cinghiale,plasmatoa mano, 105
di cui si conserva solo la parte anteriore, con il pelame a folte ciocche ondulate, eseguito
con fitti colpi di stecca, le narici ottenute mediante due impressioniconcave contornate da
due linee semicircolarie l'attacco della dentaturanella bocca spalancata(P1.88:d). Sia per
fattura che per dimensioni il pezzo appare identico all'esemplarepiu completo rinvenuto
a Frattedi Salerno, databilealla fine del IV-inizi III secolo e consideratedi probabileofficina
italiota, che ipoteticamente G. Greco attribuiscea un edificio sacro, forse il medesimo cui
appartenevail celebre acroteriocircolarecon Eracle in lotta con il leone nemeo.106
Un'idea piu precisa della sua originariastrutturae funzione ci pu6 essere fornita da
un doccione angolare integro di Pompei,107presumibilmenterinvenuto nell'area del Foro,
dove avrebbe decorato uno degli edificipubblici di eta sannitica. Per quanto di dimensioni
leggermente minori, 08 esso rivela la medesima cura e la stessa freschezzadi modellato dei
98
N. inv. 7488; alt. att. 14, largh. att. 21; argillalocale.
99 Kastner 1990, pp. 251-264; Greco 1990a, pp. 63-65.
00 Greco 1990a, p. 69, fig. 78.
101 Koch 1912, pp. 76-77, tav.XXII:3, figg. 85, 86.
102 Nettamente inferiore all'esemplarecon Eracle di Fratte (Greco 1990a, p. 68: diametro interno 62/64,
esterno 78/81), si accosta viceversaper dimensionial tipo arcaico con testa di Acheloo, rinvenutonello stesso
sito (Greco 1990a, pp. 63-65: diam. ricostruito44 circa).
103 A Capua: Koch 1912, pp. 74-76, tav.
XXII:2; a Fratte: Greco 1990a, pp. 60-63.
104
Cfr. Isler 1970, pp. 40-43, nn. 10-12; n. 91, tav.XII; n. 92, tav. XIII; n. 272, tav.XXI.
105 N. inv. 9455; alt. att. 9, largh. att. 14,5; argillalocale.
106
Greco 1990a, p. 68, fig 79. La popolaritadel motivo nella decorazionearchitettonicadi Frattee attestato
dallapresenzadi altridue frammentidi modello identico, ma realizzatiin tufo (pp. 81-82, nn. 44-45, figg. 117,
118). Nel pezzo di Frattel'altezza e di cm. 45, la lunghezza totale di cm. 52.
107
Von Rohden 1880, p. 32, tav. IV.
108 Alt. 40; lunghezza testa 35.
LETERRECOITE
ARCHITETTONICHE
DI ELEA-VELIA 301
nostri esemplari. La testa in questo caso emerge da una lastrapiatta quadratadi base che
doveva essere affrancataallo spioventedel tetto, ad una delle estremitAangolari.
In conclusion il quadro che emerge dalla rapida analisi che abbiamo effettuato delle
terrecottearchitettonichedi Elea in eta ellenisticaci consente di enucleare alcune tendenze
assaichiare: una di esse, senza dubbiola piuicolta e raffinata,si rifaa modelli di sicuramatrice
magno-greca, che filtrano elaborazioni ampiamente sperimentatenelle officine dell'Italia
meridionalee che trovanoin questoperiodo in Tarantoil loro maggiorecentro di diffusion.
La seconda tendenza riconfermainvece la gravitazionetirreno-campanadi Elea che aveva
caratterizzatoil momento arcaico della cittA: essa si mantiene assai viva e costante anche
in etAellenistica,come documentanoi numerosiconfrontiche si sono di volta in volta istituiti
con Capua, Pompei e Frattedi Salerno.
Resta da comprendere,come compito principal della ricercaancorain corso, l'effettivo
ruolo sostenuto da Elea tanto nella recezione quanto nella diffusionedi questi modelli: il
volto culturaledella citta in questo periodo e assai composito, con apertureal piu raffinato
linguaggiofigurativedi improntagreca che coesistonocon manifestazioniformalipiu corsive,
caratteristichedei centri campani ormai pienamente sannitici. Cio rende assai suggestive
e probabile l'ipotesi che la produzione delle terrecotte eleati non sia rimasta confinata ad
un livello prevalentementericettivo, ma che, viceversa, essa abbia rivestitoun importante
ruolo di elaborazioneautonoma e di successivediffusion dei nuovi tipi.
M.J. STRAZZULLA
BIBLIOGRAFIA
GIOVANNA
GiEco
II
DEGLISTUDID'NAPoLIFEmRnuco
UNMWERSTA
Dipartimentodi Discipline Storiche
via Porta di Massa, 1
80133 Napoli
Italia
~~~~~4,. 5 0~~~~~~~~~~~~~~~~
U "00~~~~~c
C~~~~~~~~~~~~~~~~~~~~~~~~~~c
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~41 ''
AC
~
009qC ~
PLATE87
I~~~~~I
e talci vegetali
b. Antefissaad echmnocon testmna
U 2i *, A__
LCi
_d'1 So ,0 C
. 5
0-5 ~ ~ ~
?aR_5
8; Liz i~~~~~~U
EIN ARCHITEKTONISCHES TERRAKOTTAFRAGMENT
HELLENISTISCHER ZEIT AUS REGGIO CALABRIA
89, 90)
(PLATES
M MUSEO NAZIONALE von Reggio Calabria ist ein im Stadtgebiet von Reggio
Calabria,das sich ilber den Ruinen der griechischenKolonie Rhegion erstreckt,gefun-
denes Terrakottafragmentaufbewahrt(Pls. 89, 90),1 welches bis jetzt fast keine Beachtung
gefunden hat. Von R. Spadea wurde das Stuck unter den architektonischenTerrakotten
von Reggio Calabria erwtihnt, ohne auf dessen Funktion einzugehen:2 "Esse sono di eta
ellenistica,come un bel frammentoin cotto con parte di una figura,sedutaprobabilmentesu
una roccia."
Das Bruchstuckbestehtaus zwei im Winkelvon etwa 90 zueinanderstehendenTonplat-
ten, deren originaterechte Seitenkanteund hintere Kante erhalten sind, wahrend der linke
Teil und der obere Abschnitt abgebrochensind. Im Gegensatz zum hinteren Rand ist der
rechte Rand des Fragmentsnicht gerade geschnitten,sondern der Ton wurde an der Kante
wenig sorgfaltigmit der Hand verstrichen.In die rechte Seitenkanteist im Bereichder Kante
zwischen den beiden Platten eine waagrechteHohlung eingelassen. Bei einer horizontalen
Lage der unteren Platte betragt die Breite des Fragmentes 22,3 cm, die Hohe 16,7 cm
und die Tiefe 23,8 cm. Die obere Platte ist 2,5 bis 3 cm dick, die untere ungefahr 4 cm.
Die AuBenseitender Platten sind mit Reliefs verziert. Die nicht fur die Ansicht bestimmte
Innenseite ist auf sehr grobe Weise geglattet. Die Oberflache der AuBenseitenwird von
einer Schicht fein geschlammtenTons gebildet. Reste einer Bemalung sind nicht sichtbar.
Die Oberflacheder unteren Platte wird von unregelmaiBiggeformten Buckelngebildet,
welche durch Kehlungen geringerTiefe getrennt sind. Die hochsten Teile der Buckel sind
ungeftihrin einer Ebene angeordnet. Die Anordnungvon Buckelnvon verschiedenerGroBe
und Formentsprichtder Darstellungvon Felsenin der klassischenund hellenistischenKunst.3
Auf der Vorderseite der oberen Platte ist in hohem Relief eine menschliche Figur
wiedergegeben, welche an der Kante ein wenig auf die untere Platte tibergreift. Die
Relieftiefe betragt 8-9 cm. Von der Figur sind der untere Teil des Bauches, der Ansatz
des rechtenBeines, das linkeBein mit Ausnahmedes FuBesund ein Teil des linkenArmes im
Bereichdes Ellbogenserhalten. Das rechteBein war zur rechtenSeite der Figurausgestreckt.
Der linke Oberschenkelist ganz leicht zu seiner Linken gedreht. Der linke Unterschenkel
I
Inv. Nr. 3266 c. Far die Moglichkeit zur Untersuchung dieses Stuckes dankt der Verfasser Liliana
Costamagna, Silvana Cilione und Beatrice Nucera von der SoprintendenzaArcheologica della Calabria,
fur die Genehmigung zur Publikationder SoprintendentinElena Lattanzi.
Der Verfasserdankt ferner allen, welche sich nach dem Vortragin Athen an der Diskussionbeteiligten,
wobei besonders die Beitrage von Rudolf Kanel und Volker Kastner zur architektonischenFunktion des
Fragmentesund von Gerhild Hubner zur Ikonographiehervorzuhebensind.
Die Photographienwurden vom Verfasseraufgenommenund von GerhardFeitzingervergroBert.
2 Spadea 1987, S. 94.
3 Carroll-Spillecke1985, S. 56-63.
306 PETER DANNER
ist angewinkeltund zur rechten Seite der Figur gerichtet. Die Figur ist also in sitzender
Haltung mit zur Seite gestrecktenBeinen und auf den linken Ellbogen gestUtztemK6rper
dargestellt. Der Bauch ist nicht von Gewand bedeckt. Vom Ellbogen verlauft das um
den Arm geschlungene Himation in einem Bogen zum Ansatz der Beine. Ein Zipfel des
Himations haingtvom Ellbogen herab. Die Beine sind zur Ganze vom Gewand verhullt.
Da die Reliefdarstellungdie beiden im Winkel von etwa 90? zueinander stehenden
Platten zur Ganze ausfuillt,waren beide Seiten flurdie Ansicht bestimmt. Aufgrund der
Position der Figur steht fest, daB der obere Abschnitt des Bruchstuckessenkrecht und
der untere waagrecht angeordnet war, sodaB beide Ansichtsseitennur bei einem tieferen
Standpunktdes Betrachterssichtbarwaren. Das weist ebenso wie die Tonqualitatund die
Dicke der Platten auf eine architektonischeFunktionhin. Wahrscheinlichhandelt es sich
um das Fragmentder Verkleidungeines aus Holz bestehendenArchitekturelementes,dessen
Unterseite nicht oder nicht zur Ganze auf einem anderen Baugliedauflag.
Obwohl aufgrund der fragmentarischenErhaltung und des Fehlens von Vergleichs-
beispielen ein sicheresUrteil nicht moglich ist, kann in Erwagunggezogen werden, daB3das
Terrakottabruchstuick aus Reggio Calabriader Verkleidungsplatteeines column oder mutulus
angehort. Diese Form architektonischerTerrakottenist nicht nur in Etrurienund Latium,4
sondern, wie der Sarkophagdes Athleten aus Tarent5 zeigt, auch in der Magna Graecia
nachgewiesen. Der horizontale Teil des Bruchstacksbedeckte also vielleicht die aus der
Tympanonwand ragende Unterseite, der vertikaleAbschnitt hingegen die Stirnseite eines
in Langsrichtungdes Gebaudes angeordnetenDachbalkens.
Wegen des Erhaltungszustandesist die Rekonstruktionund Deutung der dargesteilten
Szene problematisch.Das Motiv der aufeinem FelsensitzendenFigurentsprichtDarstellun-
gen der Aphrodite6sowie anderer Gottheiten und Menschen mannlichen und weiblichen
Geschlechts7in der klassischenund hellenistischenKunst. Ohne Kenntnisdes Gebaudes,zu
dessenAusstattungdas Terrakottafragment gehort,und ohne KenntnisweitererBruchstiucke
ist eine Identifizierungder Figur nicht moglich.
Wenn auch nicht mit Sicherheitfeststeht,welche architektonischeFunktiondas Bruch-
stuckaus Reggio Calabriahat, so ist es mit groBterWahrscheinlichkeitdas Zeugnis fir eine
bisher einzigartigeForm von Architekturdekoration,welche nach der Bildung der Falten,
die mitjener von tarentinischenTerrakottastatuetten vergleichbarist, der fruhhellenistischen
Zeit angehort. Es zeigt ebensowie die DrachenkopfeausReggio Calabria,8welchen ahnliche
Beispielein Sandsteinaus Gela entsprechen,9daB auch in hellenistischerZeit in der Magna
Graecia neue Formen architektonischerTerrakottengeschaffenwurden. Es sind also auch
die architektonischenTerrakottenein Zeugnis dafur,.daB in hellenistischerZeit im Bereich
der westgriechischenArchitekturdie schopferischeKraft nicht erloschen ist, auch wenn
sich in der Tempelarchitekturam Ende des 5. Jhs. v. Chr. ein Nachlassen der Kreativitat
4 Andren 1940, S. LX-LXII.
S Mura Sommella 1987, S. 50-54, 63 Nr. 17.
6 LAMCII, i 1984, S. 92-95, Nr. 857-898, s.v Aphrodite(A. Delivorrias).
7 Carroll-Spillecke1985, S. 56-63.
8 Spadea 1987, S. 94.
9 Gela: Adamesteanuund Orlandini 1956, S. 346, Abb. 7, 8. Morgantina: Allen 1973, S. 78, Nr. 5, 6,
Abb. 17-26.
AUSREGGIOCALABRIA 307
TERRAKOTTAFRAGMENT
EINARCHITEKTONISCHES
feststellen laBt.10 Vielmehr wurden in der hellenistischen Zeit, besonders im 3. Jh. v.
Chr., eigenstandige Architekturformen,z. B. Theaterbauten11und Grabnaiskoi,12sowie
Architekturelemente,wie das sikeliotisch-korinthischeKapitell,'3 und Bauplastik,wie der
alsMittelakroterdienendeAdlerausMegaraHyblaea,14und die Atlantenund Karyatiden,'5
geschaffen,welche zum Teil eine starkeAusstrahlungaufdie mittelrepublikanischeBaukunst
Mittelitaliensausdbten.16
BIBLIOGRAPHIE
Adamesteanu,D., und H. Dilthey. 1992. MacchiadiRossano, Ilsantuario
dellaMelfits.Rapportopreliminare,
Galatina
Adamesteanu,D., und P. Orlandini. 1956. "Gela (Sicilia).Ritrovamentivari," MSc,S. 203-401
Allen, H. L. 1973. "MorgantinaSea Monsters,Lions and 'Baroque'SicilianWorkshops,"OpRom9, S. 73-84
Andren, A. 1940. Architectural Terracottas
from Etrusco-ItakTemples(ActaInstitutiRomaniRegni SueciaeVI),
Lund/Leipzig
Bonacasa,N. 1985. "Le artifigurative.L'ellenismoe la tradizioneellenistica,"in Sikanie,G. PuglieseCarratelli,
Hrsg., Milano, S. 277-347
Carroll-Spillecke,M. 1985. Landscape Depictions in GreekReliefSculpture,Frankfurtam Main
Carter, J. C. 1975. The Sculpture of Taras(Transactions of theAmneican Philological
Association,
n.s. 65, no. 79),
Philadelphia
D'Amicis, A., A. Dell'Aglio, E. Lippolis,und G. A. Maruggi. 1991. Vecchi Scavi.NuoviRestauri,
Taranto
Isler, H. P. 1981. "Contributiper una storia del teatro antico. E teatro greco di Iatas e il teatro di Segesta,"
AumAntCI 10, S. 131-164
. 1985. "Una cariatidedal teatro greco di Solunto,"SicArch18:59, S. 65-70
Klumbach, H. 1937. Tarentiner Grabkunst, Reutlingen
Lattanzi,E. 1987. Il MuseoNazionaledi ReggioCalabria, Roma/Reggio Calabria
Lauter-Buf6,H. 1987. Die Geschichte dessikeliotich-korinthischenKapitells,
Mainz
LMC II, i = Lexiconkconographicum Mythologiae ClassicaeII, i, Zurich/Munchen 1984
Lippolis, E. 1987. "Organizzazionedelle necropoli e strutturasociale nell' Apulia ellenistica. Due esempi:
Taranto e Canosa," in RimischeGriberstrafren, H. von Hesberg und P. Zanker,Hrsg., Munchen, S. 139-154
Martorano,F. 1991. "Catalogodei modelli locresi,"in I ninfeidi LocriEpizefiri.Architettura. Cultierotici.Sacralit4
delleacque,F. Costabile,Hrsg., Catanzaro,S. 63-94
Martorano,F., und F. Costabile. 1991. "I modelli fittilidei ninfei,"in IninfeidiLocriEpizefiri,
E Costabile,Hrsg.,
Catanzaro, S. 45-62
10
Mertens 1984, S. 202-205; Mertens 1990, S. 383.
11 Isler 1981.
12
Carter 1975; Lippolis 1987.
13
Lauter-Buf61987.
14
MigaraHyblaeaIV, S. 26, Taf. 46:1, 3; 64, 65.
15
Schaller 1973, S. 69-72, Nr. 180-183, 185-187, 189; Schmidt-Colinet 1977, S. 227-228, W 28-32;
S. 242-244, M 2, 3, 5, 7-9.
Canosa: Pensabene 1990, S. 292-293, 315-316, Nr. 97-99, Taf. 131:1-5. Tarent: Klumbach 1937, S.
66-67, Nr. 158, Taf. 25. Ferner sind im Museo Archeologico Nazionale von Tarent das Bruchstuckeines
Atlanten (nv. 6162) und einer Karyatide aufbewahrt. Vaste: D'Amicis et al. 1991, S. 149-158. Macchia di
Rossano: Adamesteanu und Dilthey 1992, S. 77, Taf. 53. Montescaglioso: Lattanzi 1987, S. 194. Solunt:
Pace 1922, Sp. 225-226, Abb. 17, 18; Isler 1985, S. 65-70. Monte Iato: Ribi und Isler Kerenyi 1976, S.
13-48, Taf. 1-19. Segesta: Rizzo 1923, S. 103, Abb. 45. Centuripe: Bonacasa 1985, S. 332, Abb. 355, 356,
389-392. Syrakus:Aten. V, 208b (SchiffHieronsII.);Rizzo 1923, S. 97-105, Abb. 42-44, Taf. 6 (Theater).
Karyatiden sind ferner auf Modellen von Nymphaen wiedergegeben: Lokri: Martorano 1991, S. 80,
Nr. C 1.1, Abb. 135-139. Tarent: Martoranound Costabile 1991, S. 51, 58, Abb. 71:a, b.
16
Lauter-Buf61987, S. 90-91.
308 PETERDANNER
PETERDANNER
UNIWERSTXT
SAIZIRURG
Institutfir KlassischeArchaologie
Residenzplatz 1
A-5020 Salzburg
Austria
PLATE 89
now-~~_ ]
d. Ansichtvon unten
.44
a.V
>0 u~~~~~~~~~~~~
C S
AL,~~~~~~
4V
EINIGETARENTINERARCHITEKTURTERRAKOTTEN
(PiAris 91-93)
anderplanenRuckseitederAntefixplatte.
FormdesKalypteransatzes mehrerecmunterhalb
Diesesetzenjeweils
derOberkantedesAntefixesanundwirkenteilweisewieabgeschnitten.
DasKielerAntefixB 349besitztnochein
zuruckspringendes von ca.13 cm Lange.
Kalypterfragment
b Lt. Munseilnotierungen;die Tonfarbwerte
ergebensich aus dem am ObjektdominantenFarbton,der mog-
lichstan Bruchkanten
derTerrakottaerschlossenwerdensoilte,um etwaigeUberlagerungen durchKalk,Farb-
resteoderVersinterungen vgl. auchdie mikro-und makroskopischen
auszuschlieBen; sowiepetrographischen
Untersuchungen bei Knoop 1987,im Tafelteilund in AppendixTeil B, S. 227, sowieahnlicheMaBnahmen
beiWiederkehr 1990,S. 119.
C
ZurDatierungsproblematik
s. weiterunten.
Erscheinung;s. hierzu z.B. Mertens-Horn 1988, S. 115, die von "alternativer"Technologie spricht;vgl. auch
Mertens 1990, S. 374-375, sowie kurzlichTodisco 1990, S. 915.
14 Vgl. MunsellSoil ColorCharts,Baltimore 1975. Zu den Einschrankungenbei der Verwendung dieser
Provenienz.18 Ebenso deutet auch die bis auf eine Ausnahme (fur das Antefix Kiel B 407)
relativeinheitlicheTonfarbwertgebungim Farbbereichvon "verypale brown"(= 1OYR8/4)
auf die Produktiondieser Antefixe in Tarent, da fir eine Vielzahl anderer tarentinischer
Terrakottenahnliche Farbwertemitgeteiltwurden.19 Farbuberzugsrestefinden sich nicht
auf den Kieler Antefixen,jedoch Reste weiBerEngobe20aufVorder- und Ruckseite,so dal3
antikerFarbauftragbestandenhaben wird.21
Die erstmaligeVerwendungvon plastischenReliefkopfenauf Antefixplattensowie der
Verlauf ihrer Weiterverbreitungist bis dato nicht eindeutig gekldrt.22Aus der bekannten
Pliniusstelle23uber den Koroplasten Butades aus Sikyon laBt sich eine Reliefkopfantefix-
produktionnicht eindeutig erschlieflen,zumal fur den erwahnten ProduktionsortKorinth
direktbisher keine archaologischeEvidenz besteht.24 Die ersten Nachweise lassen sich im
7. Jh. v. Chr. vielmehr fur den Nordwesten Griechenlands,in Thermos, Kalydon, Korfu
(Korkyra)sowie im AitolischenTaxiarchosund Sak6nina25erbringen, der jedoch in dieser
Zeit unter korinthischemEinfluBgestanden hat.26 Schon bald darauf-falls nicht sogar
18 Vgl. Laviosa 1954; Herdejurgen 1978, S. 84; Kastner 1989, S. 122.
19 TerrakottaprodukteandererHerstellungszentrenbesitzen i.d.R. sowohl eine andere Tonfarbwertgebung
als auch eine andere Tonstrukturals die tarentinische,s. hierzu Higgins 1954, S. 9. Da zwar mit einem
"Export"von Tonmatrizen gerechnet werden darf,jedoch nicht mit dem Fernhandelvon groBen Mengen
unbearbeitetenTons bzw. verarbeiteterTerrakotta,ist i.d.R. bei Auffindungvon Terrakottaobjektenmit einer
Produktionvor Ort oder in der naheren Umgebung zu rechnen. Knoop 1987, S. 208 mit Anm. 536, glaubt
hingegen an einen Fernhandel selbst von groBerenTonmengen, auch wenn er keine Nachweise erbringen
kann. Der ebda. zitierte Stibbe 1984, S. 138, nennt allerdingskeinen naturwissenschaftlichenBeleg fPr den
angeblichenTransportvon Ton aus Lakoniennach Lipari. So ein Vorgangist hochstensbei kleinerenMengen
sowie eventuell bei einer "Erstausstattung" von auswanderndenTopfern und Koroplastenvorstellbar,die in
tonarme Gegenden migrieren. Ansonsten ist wohl nur der Transportvon Modeln, Hohlformen und leichten
Hilfsmaterialienwie den Pigmenten vorstellbar. Auch Knoop 1987, S. 208, konzediert "matrix-linkswith
numeroussites." FUrdie lokale Produktionim Hinterland,die auch "MetapontinerTypen" des spaten 6. bzw.
5. Jh. v. C. verwendete, s. z.B. Adamesteanu 1990, S. 146. Bei diesen "MetapontinerTypen" durfte es sich
eher um Antefix-TypenTarentinerUrsprungshandeln, s. hierzu z.B. Wuilleumier 1939, S. 425 mit Anm. 4;
Adamesteanu 1974, S. 254, Taf. 31 und KYstner1989, S. 123. Der Matrizenhandel-zumindest im 4.Jh.
v. Chr.-geht prinzipiellvon Tarent in das Umland aus,jedoch nicht zurack, so auch Giuliani 1988, S. 164.
20
Koch 1912, S. 12. Zur Fragedes Engobe-Brandes,s. ebda.;mit GegenmeinungSteingraber1980, S. 232.
21
S. Brandes 1988, S. 46-48; Wuilleumier1939, S. 425-426; Higgins 1954, S. 7: "The absence of colour ...
may be explained by the fugitivenature of the pigmentsemployed."
22 Winter 1974; s. Brandes 1988, S. 21-33.
23 Pliny,NatHist35,151-152; vgl. hierzu auch Mertens-Horn 1978, S. 30-31; Winter 1978, S. 29; Torelli
Abb. 41; Mertens-Horn 1978, S. 55-61, Abb. 22, 23. Taxiarchosund Sak6nina: Van Buren 1926, S. 59-60
(Sak6nina/Konitza: frahes 7.Jh. v. Chr.),S. 62 (Taxiarchos:7.Jh. v. Chr.),S. 138-144, Taf. 34-38; Winter
1978, S. 28, Abb. 1.
26 Zur Frage des korinthischenEinflussesin Nordwestgriechenland,s. Mertens-Horn 1978, S. 31, S. 63, die
den korinthischenEinfluBf flr die Erfindungder Dachterrakottenvon Thermos und Korfuaus historischenund
typologisch-stilistischenGrunden ausschlieBenwill, wahrend D'Andria 1990, S. 285, dieses Gebiet generell
im 7.Jh. v. Chr. als "Corinthiancolonieswith their naturalhinterland"sieht. Die starkdominantekorinthische
Kunstlandschaftdieser Zeit mit pragendem EinfluBauf die nordwestgriechischenReliefkopfantefixebetont
312 BERNDBRANDES-DRUBA
m.E. zu recht Wallenstein 1971, S. 31, S. 37-40; Winter 1990, S. 23 halt zwar den Einsatz von in Korinth
ausgebildetenArbeitern far moglich, das Konzept und die Ausfohrungsartseien jedoch ". . . foreign to ...
the pure Corinthian system produced at Corinth," so daB sie im Atolischen ein "regional roofing system"
vermutet. Diese Differenzierungkann sich aber nur auf die allgemeine Anlage der Dachkonstruktionund
nicht auf die Besonderheitender Verwendungvon bestimmtenAntefixtypenbeziehen.
27
Neils 1976; Winter, 1978, mit Fundkarten;Kastner 1989, S. 115, Aim. 5-6.
28 Van
Buren 1923, S. 80-82, Abb. 66 (6.Jh. v. Chr.),S. 147, Nr. 49; Phillips 1968b; Winter 1978, S. 31,
Taf. 8:1, 2, und 4; 9: 1;Adamesteanu 1982, S. 311, Abb. 10.
29 Winter 1978, Taf. 9:1-3. Ahnliches gilt auch fuirdie Reliefkopfeaus Apollonia, Afrati und Ephesos (?)
bei Mertens-Horn 1978, S. 63, Abb. 19-26. Es ist schwerlich vorstellbar,daB im 7. Jh. v. Chr. schon
eine "dadalischeKoine" fuirReliefkopfantefixevorgelegen haben k~nnte, welche von Kleinasien (Ephesos)
fiber Kreta (Afrati)und Nordwestgriechenland(Thermos, Kalydon und Korkyra)bzw. Albanien (Apollonia)
bis nach Westgriechenland(Etrurienund Unteritalien)reichte. Knoop 1987, S. 204, sieht den Hauptbereich
Etruriens mit Murlo und Acquarossa am Ende des 7. Jh. v. Chr. sogar .. . almost cut off from Greco-
Campanian influences....."
30 Im Laufe der zweiten Halfte des 5. Jh. v. Chr. wird Tarent auch im Bereich der Arula-Produktionin
Unteritalien fuihrend,deren Wirkungnicht nur dort sondern auch bis uiberdie Adria nachgewiesen werden
kann;s. hierzu: van der Mejden 1990, S. 130.
31 Graepler 1984, S. 88. Coarelli 1970 hatte hierzu noch eine andere Auffassung.
Tarent hat sich kulturell wohl noch bis zum Ausgang der Romischen Republilcbehaupten konnen:
Metzler 1985, S. 21, und Graepler 1990, S. 90, sowie Steingraber1990, S. 79. Allgemein zum EinfluBRoms
auftarentinischesKulturlebenbei Pape 1975, S. 8.
32 Jucker 1961, S. 199, sieht den Ursprungdes Motivs der Einzelkbpfeder unteritalischenund etruskischen
Dekorationskunstin orphischen Unsterblichkeitsvorstellungen.Demnach mulBtendie orphischen "Heiligen
Reden" schon ab der Mitte des 7.Jh. v. Chr. im etruskisch-italischenRaum bekanntgewesen sein, wenn man
die frohenAntefixkopfedieserLandschaftenhierdurchstimuliertsahe. Das halte ich far wenig wahrscheinlich.
33 Wallenstein 1971, S. 31.
34 Winter 1978 datiert die Antefixe aus Murlo rechts frilh schon in die Mitte des 7. Jh. v. Chr.; vgl. z.B.
fur etruskischeKopfprotomentypenGiglioli 1935, Taf. 161:1; Riis 1981.
EINIGETARENTINERARCHITEKTURTERRAKOTTEN 313
Morgantina: Darsow 1938; Orlandini und Griffo 1970, Nr. 62; Kastner 1982, S. 175; Kenfield 1990,
berichtetfur Morgantinavon mehreren "sets"von spatarchaischenGorgonenantefixenaus den Ruinen eines
Naiskos, der nach seinen Aussenmassenwohl eher ein groBeresSanktuardarstellteals ein gewohlich unter
diesem Terminus zu verstehendesGrabmalin Form eines kleinen Tempels;ebenso strittigdas bei Bartoccini
1936, S. 200-201, Abb. 105, aufgeftlhrterunde archaischeGorgonenantefixaus dem Ende des 6.Jh. v. Chr.
aus einem angeblichenNaiskoskontext;s. hierzu auch: Herdejurgen1978, S. 85, Anm. 3.
Metapont: Vgl. Mertens 1975; Mertens 1977; Mertens 1979.
43 In spatklassisch-frthhellenistischer
Zeit scheinenbisweilenauch Privatgebaudemit Antefixengeschmnnckt
worden zu sein, s. Brandes 1988, S. 77-79 mit weitererLit. Orsi 1933, S. 73, und Herdejurgen 1978, S. 84,
Anm. 4, vermuteten das schon fur die klassischeZeit; m.E. besteht hier jedoch eine zeitliche Abfolge in der
Erweiterungdes Gebrauchsvon Reliefkopfantefixenin Unteritalienan bestimmten Gebaudetypen,wahrend
dagegen Winter 1978, S. 53, meint, daB "(the)employment of the human face as decoration does not seem
to be connected with any specific type of building." Fur das indigen besiedelte Serra di Vaglio am Basento
im 6. und 5.Jh. v. Chr. s. Orlandini 1971, S. 288, Taf. 46:2, sowie Adamesteanu 1990, S. 146: "The main
axis was flankedby rectangularbuildingsof either religiousor public character."Damit scheinen aber keine
Privatgebaudegemeint zu sein.
44 Zur Grabungsliteratur,s. Herdejurgen 1982, S. 19, Anm. 2, und Graepler 1984, S. 86, Anm. 3.
45 Kastner 1982, S. 111. Diese sind eventuell als "wertlose"Objekte nach den Ausgrabungengar nicht
erst mitgesammeltworden, wahrend die Antefixe als zumindest asthetischansprechendffir Sammlerzwecke
galten. Bartoccini 1936, S. 193-201, Abb. 105, berichtet for eines der frfheren Antefixtypen Tarents von
Funden im Zusammenhangmit Akroterfigurenund anderenArchitekturdetails,die auch auf ein gr6fleresund
reprasentativeresGebaude als einen Grabnaiskoshinweisen k6nnten, da bisher fur das 6./5. Jh. v. Chr. noch
kleineglaubhaftenNachweisefur TarentinerNaiskoierbrachtwerden konnten;s. hierzu auch Lohmann 1979,
S. 7-8, mit Anm. 44-48.
46 Detaillierte Angaben sind derzeit nicht ermittelbar; s. Laviosa 1954, S. 217, Anm. 1. Dittmers-
Herdejurgen 1979, S. 815, Anm. 2, hatte aufgrund der Fundangabe von Bernab6 Brea 1940, S. 476, fir
zahlreicheAntefixe unterschiedlichenTyps und Stils an einem Fundplatzgemeint, "es handelt sich offenbar
um eine Art Depot." Hierbei wird es sich allerdingsum eine Sekundarverwendunggehandelt haben.
47
Polybios8,28.
48 Drago 1956, S. 32. Eine topographischeUbersicht Tarents ersteilte der mit archaologischen Sicher-
ungsaufgabenbetraute Viola 1881, Taf. 2. Neuere Studien zur Topographie Tarents befinden sich u.a. bei
Wuillemier1939, s. Plan mit Fundortvon "Statuetteset antefixes";Neutsch 1956, S. 198-199; EAAVII, 1966,
S. 605, s.v.Taranto (Degrassi);Herrmann 1966, S. 274-292; Lo Porto 1970, S. 366, Taf. 59-64, sowie zuletzt
Graepler 1989, S. 39-46 mit Lit.
EINIGE
TARENTINER
ARCHITEKTURTERRAKOTTEN 315
knapperBerichtvor, der auBerder reinenTatsachevon FundenbestimmterTypen und ihrer
Auffindungim NekropolenbereichTarentskaum weitere brauchbareAngaben erhalt.49
Der einzige bisher zeichnerischerfolgteRekonstruktionsversuch mit den in Tarent ver-
wendeten Antefixtypenzeigt ein Traufsimadachin Metapont.50 Es handelt sich in diesem
Fall um runde Antefixtypen der archaischen Zeit mit Gorgonengesichtern,wie sie etwa
ein Sttick im Museum von Brindisi zeigt.51 Ein von Carter zeichnerisch rekonstruierter
Naiskos spatklassischerZeit nach Funden aus Tarent, Via Umbria, zeigt die Dachkonstruk-
tion nicht im Detail; die hufeisenfdrmigenAntefixplattenam Traufrandsind nur sehrfluichtig
widergegeben,deren Dekor uiberhauptnicht zu erkennen.52Dennoch kann eigentlich auf-
grundvon Anzahl, GroiBeund Fundsituationder Antefixe53kaum Zweifelbestehen, daBdie
zahlreichenTerrakotta-Antefixfunde in Tarent nicht von groBerenReprasentationsbauten
wie Tempeln stammen, sondern von den bedauerlicherweisenicht mehr in situ nach-
weisbaren Naiskoi der TarentinerNekropolen.54 Die von Winter geauBerteVermutung,
daB diese Antefixe als Grabgeschenke,eventuell in Sekundarverwendung,gegolten haben
konnen, laBtsich m.E. nicht aufrechterhalten,auch wenn CarterAntefixein hellenistischen
GrabanlagenTarentserwahnte.55Hierbei wird es sichjedoch wohl um durch den Einsturz
49 Viola 1881, S. 543-544: "Si trovanonel terreno,accanto alle tombe, e in tombe gia frugate,et serviranoa
decorare la parta superior degli ipogei"; schon Wolters 1925, S. 10, hatte diese unzureichendenAngaben
kritisiert. Moglicherweisebestand der sichtbareOberbau dieser "ipogei"zumindest zum Teil Holz, dessen
Verwendungin der Magna Graecia sogar fur Tempelbautengesichertist, vgl. Mertens 1990, S. 378, fur Cir6.
Dieses warde das weitgehende Fehlenvon Architekturteilender Naiskoi erklaren.Ebenso auch Lippolis 1987,
S. 146, Anm. 24.
50 Mertens 1979, Abb. 4, vom dortigenTempel D.
51 Ohne Inv.-Nr.;s. Brandes 1988, Kat. Nr. II 7.
52 Carter 1975, S. 194, Taf. 70. Ein neuerer Rekonstruktionsversuch fur einen tarentinischenNaiskos bei
Lippolis 1987, S. 146, Abb. 32: dieser zeigt an den Uangsseitenjedochkeine Antefixe.
53 Laviosa 1954, S. 220-221, schlieBtaufgrundder geringen Gr6Be der tarentinischenAntefixe aus, daB
diese an groBen Tempelbauten angebrachtwaren (zumal zu bedenken ist, daB der Kalypteransatzauf der
Antefixruckseiteedlichecm unterhalbder OK des Antefixesansetzte,wie bei den Kieler Stuckenzu beobachten,
s.o.); ahnlich argumentierenOrsi 1933, S. 73, ffr die in Crimisa (Cir6)gefundenen "tarentinischenTypen"
sowie Herdejurgen 1978, S. 84. Ebenso sprechen der Fundort sowie die hohe Fundzahl in Tarent gegen
eine solche Verwendung,die in starkemKontrast zu den wenigen Fundrestenan Architekturdekorationvon
Tempelanlagensteht, die auBerdemin Westteilder Stadtim Gebiet der antikenAkropolisdokumentiertsind;s.
auch Wuilleumier 1939, S. 542, der in bezug auf die Graber meint: "La plupart des antefixes doivent avoir
la meme provenance."
54 Klumbach 1937, bes. S. 95-99, mit Architekturdetailsvon Grabbauten, jedoch ohne Antefixfunde.
Vermutlich wurden die Reste der Grabnaiskoials Baumaterialwiederverwendet;s. Graepler 1989, S. 41.
Allgemein scheint die Verwendungvon Antefixen an Grabbautenaber schon seit archaischerZeit verbreitet
gewesen zu sein, so Manfredini 1969, S. 78; speziellfur Tarentwird diesesbehauptetin EAAVII, 1966, S.611,
s.v.Taranto(Degrassi):"Ainaiskoiarcaicipossiamoattribuirele antefissecon Gorgoni e Sileni, altrielementi di
decorazione fittile...."; Wuilleumier 1939, S. 542, berichtet von drei Gorgonantefixmasken,welche noch in
situ am Grabbau gefunden worden sein sollen. Bedauerlicherweisegilt aber fur Tarent in Hinblick auf die
Antefixe das Wort von Coulson 1990, S. 11: ... .the presence of architecturalterracottasoften provides the
only evidence for the existence of buildings."
5 Winter 1978, S. 54, Anm. 95; Carter 1975, S. 103-111; Carter 1973.
316 BERNDBRANDES-DRUBA
56 Graepler 1984; vgl. auch D'Amicis 1984. Lunsingh Scheurleer 1986, Nr. 71, hatte bei diesem m.E.
singularenStack ohne Bildplattean ein Votiv gedacht, doch handelt es sich hierbei um kein Antefix; haufig
dienen Antefixe und andere Ziegelfragmenteauch als VerschluBim Sepulkralbereich,wie etwa Pagenstecher
1912, S. 18, berichtet,wo ein GefB in einem angeblichin das 3. bis 2.Jh. v. Chr. zu datierendeMetapontiner
Grab von einem "tarentinischen"Stirnziegelbedeckt wurde, vgl. auch Koch 1912, S. 7-8, Abb. 13, Taf. 20,
mit einem Firstziegelmit aufgemaltemMedusenhauptaus einem Kindergrabdes 3. Viertel des 6.Jh. v. Chr.
(jetzt in Paris,Louvre D 159). Fur das archaischeSatricums. Knoop 1987, S. 210, mit Anm. 542, mit einem
Fund von zwei Antefixenin einem Votiv unterhalbdes erstenTempels. Es soll hier zwar die Verwendungvon
Antefixenals Grabbeigabeffiur sie spatklassischeund hellenistischeZeit in Tarent nicht generell ausgeschlossen
werden, doch ist aufgrund der bisherigen Befunde weder eine durchgangigenoch weitverbreitetederartige
Praxis nachweisbar. Die gleiche Vorstellung wie Winter 1978, S. 54, Anm. 95, von einer Verwendung
dieser klassischentarentinischenAntefixtypenals Grabbeigabe,eventuell ebenfalls in Sekundarverwendung,
hatte ich anfanglich wegen der teilweise wie abgeschittenwirkenden Kalypteransatzean der Ruckseite der
Kieler Antefixe (s.o.), doch kann es sich hierbei auch um normale Bruchkantenbzw. durch im Kunsthandel
erfolgte "Verschonerungsmaf3nahmen" handeln. AuBerim Grabkontextfanden sich Antefixe-hier wohl als
Fullmaterial-auch in Zisternen: s. Orsi 1933, S. 71 und Foti 1968, S. 154, Taf. 10:1 (aus Reggio Calabria).
57 Carter 1970, Taf. 31, Abb. 11. Die leider recht unscharfeAufnahme zeigt ebenso wie bei Kiel B 349 ein
langeres Kalypterfragment(es handelt sich bei diesem Antefix um den gleichen Typus wie bei Kiel B 347).
Zur Kritik an Carters Fundvergeseilschaftungsbeobachtungen und Analysen, s. Dittmer-Herdejurgen1979,
S. 815, sowie Lippolis 1987, S. 142.
58 Dieses unternahmen u.a. schon fruh Holwerda 1899, bes. S. 131; Watzinger 1899; Pagenstecher 1912;
und Klumbach 1937, IX und S. 95-99, welcher immerhin aufgrund der Funde von architektonischen
Einzelgliedernin Tarent eine Rekonstruktionsolcher kleinen Grabtempelfur moglich hielt, sie jedoch nicht
zeichnerischdurchfuhrteund auch keinerleiAntefixproblemediskutierte,wahrend Smith 1976, S. 215, die
Architekturdetailsunvereinbarmit den Naiskosdarstellungenin der apulischenVasenmalereides 4.Jh. v. Chr.
fand; s. hierzu auch Wuilleumier1939, S. 553, sowie Lohmann 1979, mit weiterfuhrenderLit. Nach wie vor
wird auch in der neueren Lit. der Architekturdarstellung in der unteritalischenKeramikabbildenderCharakter
zuerkannt,wobei uberlieferte tarentinische Architekturfragmente mit diesen Wiedergabenverglichenwerden,
so z.B. bei Engemann 1973, S. 16; Lippolis 1987, S. 139-154, sowie auch Prange 1989, S. 48. In Vorbreitung
ist zu diesen Fragen die Dissertationsarbeit des Verf. Uber Architekturdarstellungen in der unteritalischen
Keramik.
59 Smith 1976, S. 215, hatte-wohl aufgrund solchartigerHypogaen in der apulischen Chora sowie des
Mangels an oberirdischen Grabbauten in Tarent-sicher zu unrecht vermutet, daB "the development of
funeraryluxury both at Tarentum and in Apulia was a subterraneanaffair";s. fur diese Hypogaen z.B. die
Abb. 11 bei Langlotz 1957, S. 411, sowie Bartoccini 1936, S. 187, Abb. 100. De Juliis 1989, S. 30, hatte
bei den monumentalenKammergraberneine Unterbrechungvom Ende des 5. bis zum Ende des 4.Jh. v. Chr.
EINIGETARENTINERARCHITEKTURTERRAKOTTEN 317
solche Grabmaler ersetzten.60 Trotz der groBen Anzahl von Abbildungen der Naiskos-
Grabbautenab dem 2. Viertel bis zum Ende des 4. Jh. v. Chr. auf apulischen Vasen kIBt
sich mit den bisher aufgefunden Teilen von Tarentiner Grabbauten keine detailgenaue
Ubereinstimmungerkennen. Carter sprach sogar von "strikingdifferences"zwischen den
erhaltenen Naiskoselementenund den Darstellungenauf den Vasen,61 doch hat Lippolis
inzwischen fur diverse Details der FunerararchitekturEntsprechungenin der apulischen
Keramiknachgewiesen.62
Dennoch bleiben Darstellungenvon Antefixenauf apulischenVasen auBerstselten und
sind keinesfallsausreichendfir eine Beurteilungder realen Terrakottaobjekte.63
vermutet. Gerade Tarent jedoch scheint einen kontinuierlichenund parallelen Gebrauch verschiedenster
Grabmalstypenin dieser Zeit zu bieten; s. hierzu Steingraber1990.
60 Vgl. Orsi 1917, S. 106-108, Abb. 10, und Lohmann 1979, S. 20, sowie jungst Kaeser 1990, S. 199.
Offensichtlichwar die Verwendungvon aufwendigenGrabbautenauBerhalbder Kulturzentrenwie Tarent
schon aus finanziellen Grunden die Ausnahme, was die groBe Anzahl von Grabmalvasenim Hinterland
erlautern warde. Carter 1990, S. 430 hat far die "ruralnecropolis"des landlichen Pantanello, der bisher
groBten systematischausgegrabenen und dokumentiertenNekropole im Gebiet Metapontos, anhand von
255 GrAbernnachgewiesen,daB dort keine Naiskosbautenverwendetwurden, aber alle sonstigenbekannten
Grabtypen (s. Carter 1990, S. 435, Abb. 13). Ebenso auch Cerchiai 1982, S. 289 for die Nekropole von
Locri Epizefirides 7. bis 4. Jh. v. Chr., sowie Gualtieri 1982, ffir Roccagloriosa im lukanischenHinterland
des 4./3.Jh. v. Chr.
61
Carter 1975, S. 15-16.
62
Lippolis 1979, Abb. 33-37 und Taf. 16-19.
63
Zumeist nur in der Form als "Giebelantefix",z.B. im Tympanon der Naiskosdarstellungauf dem
Volutenkraterin Tarent. Museum, Inv.-Nr. 9280 ( Trendall and Cambitoglou 1982, S. 771, 24/77); vgl.
auch im von der apulischenVasenmalereiabhangigenLukanischenauf dem Volutenkrateraus Montesarchio
in Salerno (= Trendall 1970 4/556e). Diese "Giebelantefixe"(Antepagmenta)sind wohl von den seit dem
frtihen 6. Jh. v. Chr. bekannten scheibenformigenAkroterenmit figUrlichenDarstellungen(vgl. Lauter-Bufe
1974, S. 215, Nr. 11, Abb. 2) herzuleiten, wie es z.B. ein westgriechischenStuck aus Salerno bei Neutsch
1956, Abb. 110, zeigt; s. auch Goldberg 1982, und zuletzt Kastner 1990.
Bei den apulischen Darstellungen von Naiskoi sind mir die VolutenkraterNeapel, NM H 2287,
Inv.-Nr.82358 (Trendalland Cambitoglou 1982, S. 784, 24/249, Taf. 290:2, 3) sowie Pavia, IstitutoArcheo-
logico, ohne Inv.-Nr. (Lohmann 1979, Kat.-Nr. A586, S. 245, Taf. 51:2 statt 51:3) bekannt, auf denen die
Art der AntefixdarstellungUiberwaagerechtem Gebalkstilcknicht nur fur die Naiskosdarstellungin Apulien
auBerstselten ist und eine Vorliebedes Pavia-Malers(Amphoren-Gruppe,ca.330-310 v. Chr.) zu sein scheint.
Die wenigen weiteren apulischenBeispielemit Antefixdarstellungenbetreffenden Tempel- und Theater-
bereich, z.B. der Artemistempelauf dem VolutenkraterNeapel, NM H 3223, Inv.-Nr.82113 (= Trendall and
Cambitoglou 1978, S. 193, 8/3), ca. 370 v. Chr., der KelchkraterTarent, NM Inv.-Nr. 52.665 (= Trendall
and Cambitoglou 1978, S. 39, 2/24, Taf. 12:la, b), beide mit sehr starkstilisiertenund in Tarent bislang nicht
nachgewiesenen Antefixtypen, sowie der apulische VolutenkraterBari, Museum Inv.-Nr. 3648 (= Trendall
and Cambitoglou 1978, S. 210, 8/144), um 350 v. Chr., mit den dreiecksfbrmigenAntefixen auf der linken
Traufseiteeines Tempels (eventuelltibermalt?),und das bekannteTheater-Fragmentin Wurzburg,Martinvon
Wagner-Museum,Inv.-Nr.H 4696 und 4701, ca. Mitte 4.Jh. v. Chr., welches runde archaisch-frmhklassische
oder archaisierende(wie z.B. in Crimisa im 5./4. Jh. v. Chr., s. Orsi 1933, S. 68 mit Taf. 9) Antefixtypen
zeigt; vgl. zu diesem Typus z.B. Herdejurgen 1978, S. 89, Kat. C4 (Anfang5.Jh. v. Chr.),sowie Herdejurgen
1978, S. 90, Kat. C5 (Mitte5.Jh. v. Chr.);Adamesteanu1960, S. 171, Taf. 9:b und K.istner 1982, Abb. 13-17.
Vom ionischen Peripteraltempeldes frahen 5.Jh. v. Chr. NachbarstadtMetapont (s. Mertens 1979) stammen
die die hiermit vergleichbarenAntefixtypen, die sicher auf tarentinischenVorgaben beruhen, auch wenn
Pagenstecher 1912, S. 15, nicht allzu wortlich genommen werden durfte, "alles, was in Metapont zutage
318 BERND BRANDES-DRUBA
kommt, ist tarentinisch."Zu der von Tarent ausgehendenVerbreitung,s. Orsi 1933, S. 71; Wuilleumier1939,
S. 425; Kingsley 1981, S. 52, sowie Kastner 1982, S. 126. Unklar, ob es sich um stark stilisierteAntefixe
handelt, bleibt die Darstellungauf dem apulischenVolutenkraterin der Slg. Marotti, Inv.-Nr. 19; s. DeJuliis
1982, Taf. 48:2 (unten).
64
S. z.B. fir die attische Keramikmit Darstellungenvon Brunnenbautenbei Glaser 1983, S. 181.
65 Zu
diesem Begrifl,vgl. Giuliani 1986, S. 14.
66 Zur Wirkungund Aussagekraftvon Bildinhaltenin der Antike,s. Langlotz 1957, S. 398, 405-42 1;Martini
1986, S. 95, sowie letztens Stahler 1990, S. 206.
67 Herdejurgen 1982, S. 19. Es ist mit ein entscheidendesVerdienst von Carters Untersuchungen der
tarentinischen Plastik, die bis dato gelaufigen Vergleiche mit den ktinstlerischenEntwicklungen anderer
Landschaften, insbesondere Attikas, zu relativieren;s. hierzu Dittmers-Herdejurgen1979, S. 815. Auch
Himmelmann 1986, S. 194, mahnte kurzlicherneut zur Vorsichtbei stilistischenVergleichenzwischen diesen
Landschaften.
68 Vgl. z.B. Brandes 1988, Kat. Nr. II:13, 14, 19, 20.
EINIGE TARENTINER ARCHITEKTURTERRAKOTTEN 319
69 Zur Terminologie und Technik des Reproduktionsprozessesbei der Terrakottaproduktionmit Hilfe des
Matrizeneinsatzes(Hohlformen),s. Brandes 1988, S. 55-62, 91-92; fernervgl.: Nicholls 1952; Nicholls 1984,
S. 24; Olbrich 1979, S. 18. Ein Hilfsmittelin der Beurteilungder relativenAbfolge von Antefixserienaus
Matrizen bietet der Trocken- und Brennschwundbzw. die Abnahme der Plastizitat:s. hierzuJastrow 1941;
Hitzl 1985, S. 13. Wiederkehrin Schwinzer und Steingraber 1990, S. 118. Zu den Einschrankungenbei
diesem Verfahren,s. Giuliani 1988.
70 Zusatzlichsind typologischeBeurteilungender Kieler Antefixe vorgenommen worden, s. Brandes 1988,
S. 80-91. Zur typologischenEntwicklungdieser Gattung zuletzt auch Krauskopf1990, S. 25-26.
71 Laviosa 1954, S. 217, Anm. 1, nennt 120 Typen und zeigt davon eine Auswahlvon 50 Typen.
72 S. z.B. bei Herdejurgen 1978, S. 84, 86. Winter 1978, S. 45, versuchte, die inhaltlicheDeutung der auf
den Antefixen erscheinendenKopfen uber die Zuweisung an bestimmten Gottheiten gewidmete Tempel zu
fuhren, ein Unternehmen, das m.E. bei der haufig unklarenFundsituationund der fehlenden literarischen
ltberlieferungin Unteritalienzu keinen eindeutigenErgebnissenfuhrenkann. Ein ahnlicherAnsatz auch bei
von Vacano 1980, S. 470, sowie bei Mertens-Horn 1978, S. 63. Vgl. dagegen Robinson 1923, S. 3. Eher
als der Bezug auf eine bestimmte Gottheit scheint mir der apotropaischeSchutzcharakterdieser Stucke an
einem Gebaude wichtig. Fur den ZeitraumunsererKieler Antefixeentfielezudem auch diese hermeneutische
Moglichkeit, da diese Stucke hochstwahrscheinlichvon Sepulkralgebaudenund nicht mehr von Tempeln
bestimmbarerGottheiten stammen;vgl. Brandes 1988, S. 106-107.
73 Kastner 1982, S. 115. Zur Entwicklungdes Gorgoneions,s. Floren 1977, mit weiterfuhrenderLit.
74 Typusvariationenliegen vor bei Kiel B 346 bzw. B 351 sowie Kiel B 348 bzw. 408.
75 Herdejurgen 1982, S. 112, undJucker 1961, S. 200 mit Bedenken, ". . . ob sich die bildhaft angedeutete
Idee fur Darstellerund Betrachterjedesmalzu einer benennbarenGottheit verdichtethaben."
76 "Artemis-Bendis":Die Bandbreite bisheriger Benennungen allein fur diesen Typus zeigt die folgende
Auswahl: Viola 1881, S. 544, Nr. 47 ("Omphale");Daimmler 1883, S. 200-201 ("Artemis");Walters
320 BERNDBRANDES-DRUBA
Konvention hinsichtlichder Benennung: so kann Kiel B 347 als "Manaden"-Typus,B 349
als "Pan"-Typus,B 350 als "Gorgo (-Medusa)"-Typusund B 407 als "Hermes"-Typusbe-
zeichnet werden, doch ist keiner dieser Typen ikonographischeindeutig in dieser Erschei-
nungsformbelegbar.77Nur das Antefix des Typus "Gorgo-Medusa"stammt noch aus dem
5.Jh. v. Chr., als die tradiertenikonographischenSchemata erst aufzubrechenbeginnen.78
Die uberliefertenTypen gewinnen im Verlaufdes 4. Jh. v. Chr. in Unteritalien ein "neues
damonisches Gesicht",79dessen InterpretationAhnlicheSchwierigkeitenbereitet wie auch
die der zahlreichauftretendenBusten in der unteritalischenVasenmalereiinsbesonderedes
letzten Drittels des 4. Jh. v. Chr.80 Eventuell bestehen hier Zusammenhange in der Ver-
wendung dieser diversen Typen menschlicherund damonischer Protomen und Gesichter
in den verschiedenen Gattungen des spaitklassisch-fruihhellenistischen
Apuliens,81die sich
nach wie vor einer konkretenakzeptablenDeutung entziehen.82
1903, D 662 ("Herakles");Laviosa 1954, S. 241 ("Medusa");Langlotz 1960, S. 13, Anm. 15 ("Erinys");
Schauenburg 1974, S. 176 (gegen verschiedene Deutungen); Carter 1975, S. 103 ("Perseus").Die Aussage
von Kenfield 1990, S. 272, daB "lion-headedantefixes are rare in the Greek world," gilt hiermit zumindest
ffr Tarent so nicht.
"Io":vgl. Schauenburg 1989, S. 121.
77 Brandes 1988, S. 116-137. Hierzu kurzlichauch Krauskopf1990, S. 26.
78 Winter 1978, S. 57: ".. . antefixes lost their initial meaning"; Kastner 1982, S. 115, nimmt aufgrund
der fortschreitendenHumanisierungdes Damonengesichtes eine Zunahme des dekorativenCharaktersan,
doch laBt sich diese Uberlegung nicht beweisen. Ich sehe eher-wie schon Herdejtirgen 1978, S. 84, es
andeutete-eine enge Verbindungmit desJenseitsglauben,der sich in diesen damonisch-dionysischenKopfen
ausdrackt.
79 Herdejurgen 1983, S. 4. Schon Wuilleumier 1939, S. 427, erwahnt, daB "plusieurstypes feminins
apparaissenta partirde 400 av.C."
80 Z.B. apulische Loutrophore im Privatbesitz, Slg. Arden, Scottsdale, Arizona, des Meo-Evoli-Malers
(= Trendall and Cambitoglou 1982, S. 935, 28/140, Taf. 368:5). Hier erscheintzum einen eine FrauenbUste
im Profilein einem Naiskos und zum anderen eine weibliche Baste mit Flugeln oberhaib dieses Grabbaus,
was bei diesem Maler auch am VolutenkraterTarent, Museum Inv.-Nr.61465, auftritt;apulische Situla in
Ruvo, Slg.Jatta 1371 (DAI Inst.-Neg. Nr. 64.1231), mit dem ebenfallsgeflugeltenjugendlichenKopf, der aus
einer Blute hervorwachst;Frgt. ehem. KunsthandelRom (hier P1.93:c nach DAI Inst.-Neg. Nr. 63.482) mit
einer am Hals abgeschnittenenMadchenbuste. Ein durchausdem Kieler Antefix B 347 vergleichbarerKopf
auf einer apulischen Oinochoe in Bochum, s. Kunisch 1972, S. 138-139, Nr. 119, aus der Mitte des 4. Jh.
v. Chr. (wohl eher spater),der ebda. als "Erinnye"benannt wird.
81 S. z.B. auch die Kopfe auf einer tarentinischenSilberschalebei Wuilleumier 1939, Taf. 19:1. Ahnliche
Phanomene existierengleichzeitigauch in anderen Kunstdandschaften, s. z.B. die 24 griechischenTerrakotta-
Appliken des spaten 4. Jh. v. Chr. "probablythe ornament of a wood sarcophagus, . . . reportedly from
Naukratis/Egypt,"Auktionskat.Sotheby'sNew York, 18.06.1991, Nr. 84. Diese Typen konnten rein formal
durchaus auch tarentinischerProvenienz sein; vgl. etwa die Gorgoneionapplikenin De Julis 1985, S. 395,
Nr. 37, 38.
82 Hierzu zahlen auch die im 4.Jh. v. Chr. auf den apulischenVolutenkraterenin den Rotellen auftretenden
Gorgonen- und sonstigen Masken, s. z.B. Tarent, PrivatbesitzRagusa (DAI Inst.-Neg. Nr. 68.4633), oder
CVAWarschau, Kationalmuseum 4 [Polen 7] 297 IV Dr, Taf. 18:1-5. Hierzu insbesondere Giuliani 1988, mit
Taf. 45-47. Allgemein zu dem Problem der (Ranken-)Kopfe, die laut Trendall 1984, S. 5: "typicalof the
developed Apulian style" sind: Jucker 1961, bes. S. 197, und jungst Stahler 1990, S. 204. Mehrfach dazu
auch: Schauenburg 1957; Schauenburg 1961, S. 97, und Schauenburg1989.
EINIGETARENTINERARCHITEKTURTERRAKOTTEN 321
BIBLIOGRAPHIE
Adamesteanu,D. 1960. "Gela,"NSc 14, S. 67-246
* 1974. "L'attivitAarcheologicain Basilicata,"XIVAttiTaranto, S. 247-259
* 1982. "Si's e Metaponto,"ASAtene 60, n.s. 44, S. 301-313
* 1990. "Greeksand Natives in Basilicata,"in Descoeudres 1990, S. 143-150
Anderson,J. R. 1883. "Antefixesfrom Tarentum,"JHS 4, S. 117-121
Andressai,G., A. Alessio, und A. D'Amicas. 1986. LesorshelUnistique de Tarente(Kat. Tarent),Tarent, S. 6-65
Andr~n, A. 1940. Architectural from Etrusco-ItaliTemples(ActaInstitutiRomaniRegni SueciaeVI),
Terracottas
Lund/Leipzig
AttiTaranto = AttidelConvegno di StudisullaMagnaGraecia, Tarent/Neapel
Bartoccini,R. 1936. "Taranto,"NSc 12, S. 107-232
Belli, C. 1970. II Tesorode Taras,Rom
Bernab6 Brea, L. 1940. "Taranto,"NSc 18, S. 426-505
Brandes,B. 1988. EineGruppe unteritalischerAnteftxein derlitelerAntikensammlung,Kiel
Brauer,G. C. 1986. Taras:ItsHistoryandCoinage, New York
Breitenstein,N. 1941. Catalogue of theTerracottasin theDanishNationalMuseum,Kopenhagen
Carter,J. C. 1970. "ReliefSculpturesfrom the Necropolisof Taranto,"AJA74, S. 125-137
. 1973. "The Figurein the Naiskos,"OpRom9, S. 97-104
*1975. TheSculpture of Taras(Transactions oftheAmerican n.s. 65, no. 79), Philadelphia
PhilologicalAssociation,
1990. "Metapontum-Populationand Wealth,"in Descoeudres 1990, S. 405-441
Cerchiai, L. 1982. "Sesso e classi di eta nelle necropoli greche di Locri Epizefiri,"in La mort,les mortsdans
lessocits anciennes (KongressberichtNeapel), G. Gnoli undJ.-P.Vernant,Hrsg., Neapel 1982, S. 289
Coarelli, F. 1970. "Il dibattito,"XAttiTaranto, S. 201-203
Cook, R. M. 1974. Rezension, CR,n.s. 24, S. 159-161
Corinth IV, i = I. Thallon-Hill und L. S. King, Decorated ArchitecturalTerracottas
(CorinthIV, i), Cambridge,Mass.
1929
Coulson, W D. E. 1990. Welcomingaddress,in Hesperia 59, S. 11-12
Damgaard Andersen, H. 1990. "The Feline Waterspoutsof the Lateral Sima from the Upper Building at
Poggio Civitate, Murlo," OpRom18, S. 61-98
D'Amicis, A. 1984. "Taranto,"Taras4, S. 47-84
D'Andria, F. 1990. "GreekInfluencein the Adriatic,"in Descoeudres 1990, S. 281-289
Darsow,W. 1938. Sizilische Berlin
Dachterrakotten,
Dejuliis, E. M. 1982. "L'attivitaarcheologicain Puglia,"XXIIAttiTaranto, S. 503-531
* 1985. Glioridi Tarantoin etaelenistica(Kat. Tarent u.a. Orte), Mailand
* 1989. "Die tarentinischeGoldschmiedekunst,"in Kat.Tarent, S. 23-27
Dejuliis, E. M., und D. Laiacono. 1985. Taranto. Il MuseoArcheologico, Tarent
Descoeudres,J.-P., Hrsg. 1990. GreekColonists andNativePopulations: Proceedingsof theFirstInternationalCongress
of ClassicalArchaeology,Sydney1985, Sydney
83 Ich danke der American School of Classical Studies at Athens, insbesondere N. A. Winter, for die
Gelegenheitzum VortragdiesesArtikels,der aus meinerunpubliziertenMagisterarbeit1988 an der Universitat
Kiel hervorgegangenist (s.u.). Dieser Aufsatzist Konrad Schauenburgin Dankbarkeitgewidmet.
Abbildungsverzeichnis:Pls. 91-93:b: Kiel, AntikensammIung(PhotosJ. Raeder); P1. 93:c: Frgt. Pri-
vatbesitz (DAI Rom Inst.-Neg. 63.482). Ich danke B. Schmaltz und J. Raeder von der Antikensammlung
der Kunsthallezu Kiel fir die Publikationserlaubnis sowie M. Prange fUrseine Unterstutzung.
322 BERNDBRANDES-DRUBA
Kingsley,B. M. 1976. "TarentineTerracottasin theJ. Paul Getty Museum" (Diss. University of California,
Berkeley 1976)
. 1981. "CoroplasticWorkshopsat Taras,"GetyMusj9, S. 41-52
Klumbach, H. 1937. Tarentiner Grabkunst, Reutlingen
Knoop, R. R. 1987. AntefixaSatricana, Assen
Koch, H. 1912. Dachterrakotten ausCampanien, Berlin
. 1915. "Studienzu den campanischenDachterrakotten,"RM 30, S. 1-115
Krauskopf,I. 1990. "Der Schild der Parthenos und der Typus der Medusa Rondanini," in Schwinzer und
Steingraber1990, S. 22-34
Kunisch, N. 1972. KatalogSammlung Funckein derRuhr-Universitat Bochum,Bochum
Langlotz, E. 1957. "Vom Sinngehalt attischerVasenbilder,"in EineFreundesgabe-Festschrif fiir RobertBohringer,
E. Bohringerund W Hoffmann, Hrsg., Tubingen, S. 397-421
. 1960. Dertriumphierend Perseus,Koin
Lauter-Buf6,H. 1974. "Entstehungund Entwicklungdes lakonischenAkroters,"AM 89, S. 205-230
Laviosa, C. 1954. "Le antefissefittilidi Taranto,"ArchCI6, S. 217-250
Lenormant,F. 1881-1882. "Notes archeologiquessu1rla terre d'Otrante,"GazArch 7, S. 25-127
LindrosWohl, B. 1984. "Three Female Head Antefixesfrom Etruria,"GetpyMusj 12, S. 111-118
Lippolis, E. 1987. "Organizzazionedelle necropoli,"in Riimische Grdberstrassen,
Kolloquium Miunchen 1985, H. v.
Hesberg und P. Zanker,Hrsg., Munchen, S. 139-154
Lohmann, H. 1979. Grabmaler aufunteritalischen Vasen(Archdologische
Forschungen7), Berlin
Lo Porto, F. G. 1970. "Topografiaantica di Taranto,"XAttiTaranto, S. 343-383
LunsinghScheurleer,R. A. 1968. Grnek in hetklein,Amsterdam
Manfredini,A. 1969. "Terrecottearchaichetarentinedel Fondo Giovinazzi,"RivStCl17, S. 75-89
Marconi, P. 1931. Himera,Rom
Martin, R. 1965. Manueld'architecturegrecque I, Paris
Martini, W 1986. "Zwei erotische Vasenbilder,"in FestschrJiftfr KonradSchauenburg, E. Bohr und W Martini,
Hrsg., Mainz, S. 95-100
Mertens, D. 1975. "Metaponto,"BdA60, S. 26-49
. 1977. "Der ionische Tempel in Metapont,"Architectura, S. 152-162
. 1979. "Der ionische Tempel von Metapont,"RM 86, S. 103-140
. 1990. "Some PrincipalFeaturesof West Greek Colonial Architecture,"in Descoeudres 1990, S. 373-
382
Mertens-Horn,M. 1978. "Beobachtungenan dadalischenTondachern,"JdI 93, S. 30-65
. 1988. Die Liiwenkopf-Wasserspeier desgriechischen
Westens im 6. und5. Jh. v. Chr.(Mitteilungen desDeutschen
Archaologtischen RdmischeAbteilung.
Instituts, Ergdnzungsheft28), Mainz am Rhein
Mertens-Horn, M., und L. Viola. 1990. "ArchaischeTondacher westgriechischerTypologie in Delphi und
Olympia,"in Hesperia 59, S. 235-250
Metzler, D. 1985. "Zur Geschichte Apuliens im Altertum," in Apulien.Kulturberiihrungen in griechischer
Zeit
(Kat. Monster),K. Stahler,Hrsg., Munster,S. 14-24
Neils, J. 1976. "The TerracottaGorgoneia of Poggio Civitate (Murlo),"RM83, S. 1-29
Neutsch, B. 1956. 'ArchaologischeGrabungenund Funde in Unteritalien,"AA [JdI 71], S. 195-450
Nicholls, R. V. 1952. "Type, Group and Series,"BSA47, S. 217-226
. 1984. "La fabricationdes terrescuites,"HistoireetArchdologie 81, S. 24-31
NuoviQuaderni 1 = Studiin onoredi FilippoMagi,Istitutodi Archeologicadi Perugia,Hrsg., Perugia 1979
Olbrich, G. 1979. Archaische StatuetteneinesMetapontinerHeligtums (StudiaArchaeologica23), Rom
Orlandini, P. 1971. "Aspettidell'arteindigena in Magna Graecia,"XIAt6iaranto,S. 273-308
Orlandini, P., und P. Griffo. 1970. g*rgento,Florent
Orsi, P. 1917. "LocriEpizefiri,"NSc14, S. 101-167
. 1933. Templwm ApollonisAlaeiad Crinisapromontorium, Rom
324 BERNDBRANDES-DRUBA
BERNDBRANDES-DRUBA
DUEVESBEKER WEG 4
D-24105 Kiel
BundesrepublikDeutschland
PLATE 91
V~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~V
I'~~~~~~~~I
a. B 346
= .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
JWI `A_
b. B 347 c. B 348
A. F-A
I _
QR s
c. B350 d. B351
BiERN BRANDES-DRUBA:
EDNGETARENTER ARCHITEKTURTERRAKOTTEN
PLATE 93
c. Fragment. Privatsammlung 6
1@ d 7 ~~~~~~~~~3
4 5 6 7 9 1f
A LATECLASSICALSIMAFROM HERACLEA
IN LUCANIA
(PATEs 94, 95)
I N 1970 the Soprintendenza Archeologicaof Basilicata excavated the area of the so-called
Archaic Temple in the valley to the south of the hill of Policoro (ancient Siris-Heraclea)
(Fig. 1).1 The fragmentaryconstructionthat appeared(Fig.2), until today known only from
preliminaryreports,was identifiedas a temple,2but its dating remained controversial.The
mixed constructiontechnique of the foundation walls and some architecturalfeatures of
the ground plan (the hypotheticalrow of columns on the wall in front of the pronaos, the
possible buttressesof the naos) appeared to indicate an Archaic construction,but without
any certainty.3 Unfortunately,the reexaminationof data from the excavations has not so
1 This ongoing research is extracted from my thesis work on Greek archaeology at the University in
Florence. I am grateful to the Soprintendente of Basilicata,Dr. Angelo Bottini, and particularlyto Prof. Dinu
Adamesteanu for permission to study this material. A special thanks to Dr. Madeleine Mertens-Horn for
her confidence in me and good advice as always;to Dr. Nancy Winter for her real kindnessin allowing my
attendance at this conference; not least to Prof. Luigi Beschi, my first teacher, for all that he taught and still
gives me and for his patient humanity.Thanks are also due to Nunzia ArcutiArmento for help and advice with
photos, and Carmela Petracconefor inkingthe site and temple plans.
2 Now, only the foundations of the building, probably a peripteral temple set in the so-called area B
(ca. 32.80 x 15.20 m.), remain. For the bibliography,see Adamesteanu 1969, pp. 199-200; Adamesteanu
1970/1971, pp. 136-138; Adamesteanu 1970, pp. 483-484, pl. LXXXII; Adamesteanu 1974, p. 97; Mertens
1976, p. 185; Adamesteanuand Dilthey 1978, p. 518; Mertens 1980a, p. 45; Adamesteanu 1980, pp. 87-88;
Adamesteanu 1982, p. 307; Gullini 1983, pp. 237-238, pl. XI: 1; Adamesteanu 1985a, p. 101; Adamesteanu
1985b, p. 63, fig. 28; Edlund 1987, p. 112;Torelli 1990, p. 11.
3 Gullini (1983, p. 238) talkedabout a Proto-Archaictemple, "forsedatabileanche nel primo quarto del VI
secolo,"with a wooden peristasisand wallsof unfiredbrick. As faras the mixed buildingtechniqueis concerned,
here documented by the presence of some big blocksof sandstoneand conglomerate,approximatelysquared,
and of simple pebbles, its use is commonly acknowledgedin the most ancient buildings (cf. Dinsmoor 1950,
VII, pp. 223-224) and in smallerones generally(cf.Fiorentini
p. 58) as well as in domesticstructures(cf. Olynthus
1985, pp. 104-105, buildingsI, II, V-VIII; see also Mertens 1980b, pp. 341-343). In any event, it cannot be
excluded that in some areas like this, which lack good buildingstone (cf. Mertens 1981, p. 116; Mertens 1990,
p. 374), the use of such a techniquecontinuesin some importantbuildingsas well. The hypotheticalsolutionof
a row of columns in frontof the pronaosseems to recallcloselythe similarsolutionadoptedfor the firstphase of
Temple B (Heraion)in Metapontum, begun in the second quarterof the 6th century B.C. and unfinishedin
its originalplan (forthis temple, see Adamesteanu 1974, pp. 27-28; Mertens 1973, pp. 201-203; Adamesteanu
1976, p. 156;Mertens 1976, pp. 172-174; Mertens 1980b,pp.329-336; Gullini 1983, p. 240). The incomplete
state of the "ArchaicTemple" plan, however, requirescaution. For the supposed buttressesof the naos, see
Mertens 1976, p. 185. The Archaic material recovered from the excavation (mainly some Late Daedalic
terracottastatuettesand a few fragmentsof Ionic bowls) cannot be clearly connected with the architectural
structurein question(see note 4 below). Forthis Archaicmaterial(particularlythe statuettes),see Adamesteanu
1981, p. 87, pl. VII:2; Rolley 1980, p. 186, pis. XXXV:2, XXXVII:3; Adamesteanu 1982, p. 307, figs. 5,
6; Orlandini 1983, p. 335, fig. 310.
328 LUISA VIOLA
I~~~~~~~~~~~I
0 1005D
A-
FIG.1. Policoro, site plan of Heraclea. 1. Museum; 2. City walls; 3. House quarter;4. Potters'quarter;
5. 'ArchaicTemple";6. Temple of Demetra; 7. so-calledBaron'sCastle;8. City area;9. Necropolis
(fromAdamesteanu 1985a, fig. 43)
O O
c
00 > C> C c
00
i _t
bb
LO
co
OC)
wo~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4
C) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C
>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C
cC<n IO
C4
0 ','(I)
-"U H
330 LUISAVIOLA
far yielded a clear dating of the monument.4 A series of trial excavations undertakenin
1980 immediately south of the temple, moreover, revealed the preponderantpresence of
Late Classical and Hellenistic ceramic and coroplastic material,5 as previously found in
1970 throughoutthe area.6 A dating of the structuresin the 4th century B.C. thus becomes
more likely.7
From the 1970 excavations emerged, but without precise details, 21 small fragments
of sima decoratedin relief (Pls.94, 95), today in the Museo della Siritideat Policoro.8 With
all probability,the fragmentsbelong to the same roof.9 Five of them (14-18, Pls. 94:d, e,
95:d, e) are likely part of a lateral sima, whereas a sixth fragment (19, P1.95:e) appears to
belong to a rakingsima.1I
4 For the excavated area, the excavationjournal (13 April-26 June 1970) offers no help, lacking plans,
drawings,and photographicdocumentation.
5 These excavations, carried out to the foundationsby Antonio De Siena and SalvatoreBianco, revealed
no Archaic material but instead, direct evidence of the Bronze Age. I particularlythank Dr. De Siena for
information. For the same excavations,see also, recently,Chiappavento,forthcoming.
6 Among the non-Archaic material of 1970, it is possible here to mention particularly(with only a few
examples cited here for each type): vase fragments in Gnathian style (Inv. 202.004, 202.008, 202.013);
fragmentaryskyphoiand unguentariain blackpaint (Inv.202.007, 47366); votive terracottasas female heads,
with or without polos (Inv. 47427, 47457), or belonging to the so-called Artemis Bendis type (Inv. 47552);
childrenwith himation and phiale (Inv.47432, 47448); a male head of the Dionysos-Hades type (Inv.47467);
pinakeswith presumableheroized dead men (Inv.38765).
7 At this point the resultsof the excavationscarriedout in the same area since 1986 by PerugiaUniversity
will probablyreveal very importantinformation:see Pianu 1991; Pianu, forthcoming.
8 Photos of the materials on Plates 94 and 95:a, b, d, and e were kindly supplied and authorized by
the Archaeological Soprintendenzaof Basilicata (photographerN. Arcuti Armento). The numbers for the
negativesare as follows: Plate 94:a, Neg. no. 94385; b, Neg. no. 94381; c, Neg. no. 94399; d, Neg. no. 94384; e,
Neg. no. 94382; Plate 95:a, Neg. no. 94379; b, Neg. no. 94378; d, Neg. no. 94383; and e, Neg. no. 94380.
The photographson Plates 95:c and f were taken by me.
Besidesthese fragments,the excavationhas produceda greaternumberofArchaic architecturalterracottas
(revetmentplaques, geison revetments,simas, antefixes)and few Late Classicaland Hellenistic antefixes. For
some of these terracottas,see Adamesteanu 1980, pp. 87-88, pls. III-VI, VII:1; Rolley 1980, pp. 181, 183,
pls. XXVI:1, XXXVI:2; Adamesteanu 1982, p. 307, fig. 7; Orlandini 1983, pp. 336, 335, 402, figs. 324,
332, 422; Rolley 1983, p. 218; Orlandini 1985, p. 109, fig. 58, pl. 47; Olbrich 1986, p. 137, fig. 15;Viola 1990.
9 The fragmentswere divided into two differentseries, one (1-13) characterizedby the presence of the
identical palmette and crown of the vertical plaque, the other (14-21) by the presence of the same type of
lotus with hangingbuds (see below,text). But it is evidentthat both seriesbelong to the same roof, both for color
and quality of clay, which is very similar in all fragments,and for the same measurementsregisteredfor the
reconstructeddecorativemotifs(H. palmetteand lotus 0.115 m., measurementstaken on 1 and 19). Moreover,
only some millimetricvariationsdistinguish,in 3 and 14, the height of the lower part of the sima, calculated
from the top of the horizontal S-motif volutes to the lower plane of the vertical plaque (H. lower part in 3,
0.048, in 14, 0.044 m.) At this point, we can add the examination of the same volutes, which below the
palmette are always arranged downward, whereas the volutes below the lotus are always disposed upward
(see 14, 15, 19, 21), thus forming a horizontalS-motif as a clear connecting element between the decorative
patterns. Finally,it is possible to note that the bead and reel of some fragmentsof the first series (1, 2, 4)
are very similarto that of the second series(18, 20); on all the fragmentsin which the profileof the lower part of
sima is still visible (3, 14, 15, and 19), the angle between the vertical plaque and the pan tile corresponds
closely (see catalogue),on the whole referableto an average of 85 degrees.
10 That three fragments in particular(14-16) belong to a lateral sima is possibly proven by the scale of
the acanthus leaf, smaller than in another fragment (19), evidence of a device whose aim is to make room
SIMAFROMHERACLEAIN LUCANIA
A LATECLASSICAL 331
0 0.05m
a I_0=00= b
The reconstructionof the profile (Fig. 3:a) suggests a slab ca. 0.23 m. high, decorated
with an anthemion and bead and reel. The sima is completed at the top by a crowning
molding with ovolo, decoratedwith egg and dart, projectingfillet, small cavetto, and small
taenia, from bottom to top. The reconstructedanthemion (Fig. 3:b) shows a palmette with
eleven leaves, two of which near the center are bent toward the central leaf palmetteeof
"exploded"type), and a lotus flowerwith four petals risingfrom two acanthusleaves, with a
The free mixing of the analyzed elements is directlycomparableto that free structural
composition characteristicof all terracottaroofs of Magna Graecia and particularlycon-
nected with a local productionarea. Confirmationis providedby an unpublishedfragment
of sima (P1.95:f),'8 likewisefrom the Heraclea area, more preciselyfrom the temenos of a
country shrine excavated at the left bank of the Sinni river (Petrullaarea),19and recently
dated between the end of the 5th century and the first half of the 4th century B.C.20 The
fragmentreproducesin identicalform, althoughon a smallerscale, the profileand decoration
of the observed crown.21 It can be said, then, that such a structurefor a sima is peculiar
to the region of Heraclea.
The compositionalscheme of the anthemion, with lotus and side buds, appears com-
pletely new in the repertory of anthemia represented on Western Greek simas. Vegetal
elements of secondary importance (flowers,buds) alongside the lotus are known in some
Attic fictile simas and stelai datable from ca. 430 B.C., although in none of them are flowers
or buds pendant.22 This last feature is alreadyvisible alone on some Attic vases, or vases
of Attic style, from the period between the second and third quarter of the 5th century.23
Thus the anthemion could be a scheme of Attic inspiration since the relations between
Heraclea and Attica were active in the lastpart of the 5th century,afterthe foundationof the
city in 433-432 B.C. by the Panhellenic colony of Thurii, cofounder of Heraclea together
with Tarentum.24The same scheme is, however,renewed by the presence of some Magna
Graecianvegetal motifs such as the palmette, a properlyItaliot creation.25The type, which
appearsin ceramicsbetween the end of the 5th and the early4th centuryB.C.,26is knownon
some Tarentine and Metapontineantefixeswhere the motif appears,nevertheless,as a later
1990, p. 231, pl. 31:d), as is a fragmentof an architecturalterracottafrom Satyrion, considered to be of the
same period (Lo Porto 1964, no. 41, p. 253, fig. 71:5).
18 Policoro, Museo della Siritide. Inv. no. 43396. H. 0.074 m. L. 0.127 m.; H. eggs 0.023 m., projecting
fillet 0.007 m., H. cavetto 0.012-0.018 m., H. taenia 0.006 m. Pinkish orange clay, with composition and
slip very similar to that of 1 (see the catalogue). There are a few traces of black on the frames of the egg
and dart, and red for the darts.
'9 Bini 1989, pp. 15-16, 19-21, with earlierbibliography.
20 Bini 1989,p. 20.
21
It has also the same shape of darts in the egg and dart, a shape that in some fragments (1, 7-10, 12,
and 13) can appear more rounded because of the distributionof color. The sima of the fragment from the
Petrullaarea could probablybelong to the pediment of one of the two small aediculesof the temenos.
22 Compare sima XIX of the Akropolis,datableto about 430 B.C. (Buschor1929, pp. 40-41, fig. 45, pl. 8), or
the later Athenian stele of 394 B.C. (M6bius 1968, pp. 24-25 [9:d]).
23 I suggest only a few of the numerous possible examples: CVA,Firenze3 [Italy 30], 112:1, 3; 113:1, 3
(460-450 B.C. and 465-460 B.C. respectively);CVA,Berln 3 [Germany 22], 111:5-6 (toward450 B.C.); CVA,
Altenburg 2 [Germany 18], 69:5-7 (third quarter of the 5th century). The comparison with the scheme of
the sima in question is, however, only good for the manner in which the stems of the vegetal elements are
arrangedon the cups mentioned.
24 It is possibleto thinkthat the Attic potters'workshopwas active in Heraclea between 430 and 420 B.C. and
was led by the Policoro Painter. For the Policoro Painter and the grave in Heraclea that took its name from
him, see, in particular,Pianu 1989, with earlierbibliography.For the Attic influence on coinage, coroplastics,
and ceramics,see brieflyOrlandini 1985, p. 109. For the historyof Heraclea, see Sartori 1967.
25
So recognized alreadybyJacobsthal 1927, p. 178.
26
Among the earliest representations,see Trendall and Cambitoglou 1978, no. 24, p. 39, pl. 12:1(b),
associatedwith the Painter of the Birth of Dionysos (end of the 5th century B.c.-385 B.C.); Andreassi 1979,
no. 43, pp. 91-94, Lecce Painter(betweenthe firstand the second quarterof the 4th centuryB.C.).
334 LUISA VIOLA
creation than the example on the Policoro sima.27 The same motif, forming a secondary
frieze of palmettes only, occurs, moreover, in a later rendition on a revetment plaque in
the Museum of Bari dated to the 3rd century B.C.28
As far as the lotus floweris concerned,its shape is alreadyenlargedbut stillvigorous.29It
is possibleto recall a fictilesima from Tarentumor Metapontum,now in the Getty Museum
and dated to the last quarterof the 5th century,for the manner in which the four petals are
disposedand enlarged,supportedby two acanthusleaves, smooth here as well.30 The relief,
however, is more compact here and the indentation of the acanthus leaves simpler.3' For
the shape of the buds, very simple indeed, only general comparisons are possible, still in
an Attic sphere of the beginningof the 4th century32but more decidedlyin some productsof
Tarentinetaste datedfrom the firsthalf of the same century.33Finally,the bead and reel with
globular beads at the top of the anthemion can be considered a regular feature from the
Classicalperiod.34
In conclusion, the sima of Policoro seems to fall within the limits of a typically local
production, still marked, if possible, by some Attic influences of the Classical period but
more directlyby Tarentine influences.35The lack of clear stratigraphicaldata preventsthe
formulationof a hypothesisof a well-defineddate. In light of the analysispresentedhere, it is
possible, however, to suggest a date that can span the whole 4th century. Consideringthe
relationshipof the Policoro sima to the sima fragment from the Petrulla area, perhaps as
a hypothesisone can suggest a date within the firsthalf of the same century. At the present
state of research,it is not possibleto attributethe sima to any specificbuilding;its placement
on the 'Archaic Temple"roof remainsto be discussed.36
27 For some differentvariants,cf. Laviosa 1954, no. 52, p. 249, pl. LXXVIII:4 (Hellenisticperiod). Among
the Metapontineantefixes,unpublished(itwas not possibleto supplythe inventorynumbershere), is preserved
a variant, probably earlier than the Tarentine type because of its more balanced proportions,the fuller, less
tired and stiffshape of the leaves, and the same base slab, now quite semi-ellipticalin the model from Taranto.
28
Rossi 1983, p. 113, pl. 71.
29
For the development of the lotus on architecturalterracottasin the Late Classical period, see Le Roy
1967,p. 166.
30
Mertens-Horn 1988, no. 74, pp. 154, 205, pl. 80, Encl. 2:m.
31 On the contrary,it is important to notice on the Policoro sima the sharp projection of the denticles, or
thorns, on the rounded end of the acanthusleaves. For the developmentof the acanthusin the Late Classical
period, see Mobius 1968, p. 90.
32 See the Athenian stele of 394 B.C. previouslymentioned (note 22 above).
33 Compare the vegetal frieze of a fragmentarysilverrhyton from Karagodeua~ch(South Russia),probably
a Tarentine imitation, 370 B.c.-end of the 4th century (Pfrommer 1982, p. 151, fig. 9); or a couple of elix
earringsfrom a Tarentinegrave, ca. 350 B.C. (Schojer1986, no. 103, pp. 178-179); or the more famous golden
diadem from Santa Eufemia, ca. 304-289 B.C. (Lippolis1986, pp. 112-113). Cf. also a sima from Amphipolis,
last quarter of the 4th century (Kaltsas 1985, no. 83, pp. 99, 150, pis. XXVIII, IET); or, still, some vases
of Lucanian production by the Primato Painter,after 350 B.C. (Trendall 1967, nos. 931, 963, pp. 167, 170,
pis. 73:4, 75:5). For the developmentof the bud in the Late Classicalperiod, see Mobius 1968, p. 91.
34 Among the Western Greek terracottasimas, one of the first examples of a bead and reel with globular
beads is on the horizontal gable sima of the Selinuntine Temple C, recently dated by Mertens-Horn to the
last quarterof the 6th century B.C. (Wikander1986, no. 47, pp. 41-42, figs. 3, 12, with earlier bibliography;
Mertens-Horn 1988, pp. 81-82, pl. 18, Encl. 1:b).
35 Concerning the importanceand diffusionof Tarentineartisticinfluencein Magna Graecia and elsewhere
in the 4th and 3rd centuryB.C., see Orlandini 1983, pp. 481-524.
36 On this subject,see note 7 above.
SIMAFROMHERACLEAIN LUCANIA
A LATECLASSICAL 335
CATALOGUE
Descriptions of the individual elements of each fragment (in profile and decoration) have been given from
bottom to top. All dimensionsare in meters.
Preserved: very small portion of left-end side of Preserved: small portion of top part of vertical
verticalplaque, with an undamaged edge. Traces of plaque. Taenia entirely splintered. Colors well pre-
black on thirdleaf from the left. served.
Profile: hardly visible tapering of vertical plaque, Profile: molded crown of vertical plaque, as in 1
clearly observed in 1. A vertical recess, cut as an (H. ovolo 0.028, projectingfillet0.010, cavetto0.032).
obtuse angle on the edge of the left-end side, for Decoration: as 1 (moldedcrown).
fittingin of the flange on a contiguousslab.
10. Sima PI. 95:a
Decoration: half-palmette, here partiallyvisible,
following the left-side edge of plaque; for its type, Inv. no. 35468.
see 1. P.H. 0.104, p.L. 0.127, Th. 0.042-0.057.
Unpublished.
7. Sima PI. 95:a
Fabric:similarto 9.
Inv. no. 38766. Preserved: portion of top part of vertical plaque.
P.H. 0.112, p.L. 0.095, Th. 0.039-0.056. Taenia entirelysplintered. Colors well preservedon
Unpublished. egg and dart.
Fabric:pinkishbeige clay,with many smallinclusions Profile: molded crown of vertical plaque, as in
of fired brick and few of stone. Pale yellow slip on 1, with the same bead and reel (Diam. beads 0.01;
front face. H. ovolo 0.028, projectingfillet0.011, cavetto 0.032).
Preserved: small portion of top part of vertical Decoration: as 1 (moldedcrown).
plaque. Top plane splintered. Colors well preserved.
11. Sima P1.95:c
Profile: molded crown of vertical plaque, as in 1,
with the same bead and reel (Diam. beads 0.014; Inv. no. 47247.
H. ovolo 0.029, projectingfillet 0.009, cavetto 0.032, P.H. 0.082, p.L. 0.058, Th. 0.040-0.057.
taenia 0.008). Unpublished.
Decoration: as 1 (molded crown). The red lance Fabric:similarto 8.
points painted on eggs of the egg and dart are clearly Preserved:very small portion of left-end side and
visible here. All the beads of the bead and reel are top part of verticalplaque with an undamaged edge.
yellow. Cavettoand taenialargelybrokenaway. Colorsbadly
preserved.
8. Sima PI. 95:b
Profile:ovolo (H. 0.028), projectingfillet(H. 0.010),
Inv. no. 38767. and very short part of cavetto; for the type, see 1.
P.H. 0.097, p.L. 0.081, Th. 0.037-0.068. Edge of left-end side completelyflat.
Unpublished. Decoration: as 1 (egg and dart).
Fabric: reddish orange clay, with small inclusionsof
12. Sima PI. 95:b
fired brick and few of stone. Pinkishyellow slip on
front face. Inv. no. 47255.
Preserved: small portion of top part of vertical P.H. 0.066, p.L. 0.103, Th. 0.039-0.052.
plaque; back side splintered. Few traces of black on Unpublished.
framesof egg and dart and red on darts. Fabric:similarto 8.
Profile: molded crown of vertical plaque, as in Preserved: small portion of top part of vertical
1, with the same bead and reel (Diam. beads 0.013; plaque. Cavetto and taenia broken away, back side
H. ovolo 0.028, projectingfillet0.009, cavetto 0.033, entirelysplintered. Colors badly preserved.
taenia 0.008). Profile:ovolo (H. 0.029), projectingfillet(H. 0.011)
Decoration: as 1 (moldedcrown). and very short part of cavetto;for the type, see 1.
Decoration: as 1 (egg and dart).
9. Sima PI. 95:c
Inv. no. 47244. 13. Sima PI. 95:c
P.H. 0.087, p.L. 0.069, Th. top horizontal plane Inv. no. 47256.
0.069. P.H. 0.047, p.L. 0.049, Th. 0.012.
Unpublished. Unpublished.
Fabric: pinkish orange clay, with small inclusionsof Fabric:similarto 9.
fired brick and very few of stone. Light beige slip on Preserved:very smallsplinterof top part ofvertical
front face. plaque. Colors well preserved.
A LATECLASSICAL
SIMAFROMHERACLEAIN LUCANIA 337
Profile: ovolo (H. 0.029) and projecting fillet Decoration: decorative scheme observed in 14
(H. 0.010); for the type, see 1. and 15, identical in elements and scale but here
Decoration: one egg and two contiguous darts of recognizable on the left side. Partiallyvisible (from
that egg and dart observedparticularlyin 1. left): bud, stem, first acanthus leaf, and lotus petal;
on the rightof bud, see the leaf apex of a presumable
14. Lateralsima PI. 95:d palmette,very likelyas in 14.
Inv. no. 35471.
17. Lateralsima PI. 94:e
P.H. 0.104, p.L. 0.132, Th. 0.062-0.041.
Unpublished. Inv. no. 47246.
P.H. 0.07 1, p.L. 0.069, Th. 0.05 1.
Fabric:similarto 1.
Unpublished.
Preserved:centraland lowerpartofverticalplaque
and very small portion of pan tile. Colors badly Fabric:similarto 16.
preserved. Preserved: small portion of central part of verti-
Profile: pan tile and front face of vertical plaque cal plaque. Back side broken away. Colors badly
form an angle of 87 degrees. preserved.
Decoration: anthemion, only partially visible, Profile:not specifiableat present.
formed by (from left): lotus flower (pistil originally Decoration: see 16. Recognizable here: lower
painted red and side petals, two on each side, alter- part of bud and of stem painted black, and a small
nating black and red) rising from two smooth, in- part of horizontalS-motif, also black.
dented acanthus leaves, originallyblack, on a down- 18. Lateralsima PI. 94:e
ward dropping calyx and the upward curving vo-
Inv. no. 47248.
lutes (only one here visible) of two black horizontal
P.H. 0.102, p.L. 0.078, Th. 0.062.
S-motifs;palmette likely of the exploded type as in 1
Unpublished.
(see here from left, also the red thirdand blackfourth
leaf). A pendant bud, on an upright,originallyblack Fabric: clay similar to 1. Pale yellow slip on front
stem arisingfrom the volute of a lotus, dividesthe two face.
describedmotifs. Preserved:portionofcentralpartofverticalplaque.
Back side largely splintered. The hole for the lion's
15. Lateralsima P1.95:e head waterspoutis only partiallypreserved. Traces
Inv. no. 47245. of black on lotus petal and acanthusleaf.
P.H. 0.089, p.L. 0.049, Th. 0.063-0.041. Profile:not specifiableat present.
Decoration: on the left of the hole (restoredDiam.
Unpublished.
ca. 0.082), an acanthus leaf of the type observed on
Fabric:similarto 1. 14-16, but extended to the bead and reel (Diam.
Preserved: small portion of lower part of vertical beads 0.0 13), of the same type on 1. The apex of
plaque. Pan tile largelybrokenaway. Traces of black the second lotus petal is hardlyvisible directlybelow
on the volute and stem of the bud. the bead and reel (cf. lotus on 14).
Profile: pan tile and front face of vertical plaque
form an angle of 84 degrees. 19. Raking sima P1.95:e
Decoration: lotus flower,here barelyvisible,of the Inv. no. 47250.
same type and scale observed in 14. Partiallyrec- P.H. 0.096, p.L. 0.072, Th. verticalplaque 0.046-
ognizable (from left): fourth petal, second acanthus 0.041.
leaf and volute, stem of the right bud. Unpublished.
Fabric:similarto 1.
16. Lateralsima P1.94:d
Preserved: small portion of lower part of vertical
Inv. no. 47254. plaque. Pan tile broken away. Traces of black on
P.H. 0.107, p.L. 0.118, Th. 0.058. horizontalS-motif and acanthusleaf.
Unpublished. Profile: pan tile and front face of vertical plaque
Fabric: clay similarto 1. Pinkishyellow slip on front form an angle of 85 degrees.
face. Decoration: lotus flower,here partiallyvisible, ob-
Preserved:portionofcentralpartofverticalplaque. served on 14, although on a larger scale. Recogniz-
Back side broken away. Few tracesof black on stem. able (from left): second acanthus leaf and upward-
Profile:not specifiableat present. curvingcompletevolute. The absence of the bud has
338 LUISAVIOLA
to be noted here, as well as the S-motif, consequently and third petal, bead and reel (Diam. beads 0.012)
more raisedand less distancedfrom the acanthusleaf of the type observedon 1.
than on 14, 15, and 17.
21. Sima Pl.95:e
Inv. no. 47253.
20. Sima PI. 94:e P.H. 0.048, p.L. 0.044, Th. 0.023.
Inv. no. 47251. Unpublished.
P.H. 0.045, p.L. 0.054, Th. 0.046. Fabric:similarto 1.
Unpublished. Preserved: very small splinter of lower part of
Fabric:similarto 1. verticalplaque. Traces of black and red, respectively,
Preserved: very small portion of upper part of on volutes and calyx.
verticalplaque. Back side broken away. Profile:not specifiableat present.
Profile: not specifiableat present. Decoration: two upward-curvingvolutes, as in
Decoration: lotus flower,here only partiallyvisible 14, 15, and 19. Only partially visible, divided by
but probably of the same type described in detail a downward-droopingcalyx, more visible here than
in 14. Recognizable (from left): upper part of pistil on 14.
BIBLIOGRAPHY
LUISA
VioLA
via F. Baracca, n. 10/A
85013 Genzano di Lucania (Potenza)
Italy
PLATE 94
,.ri j
| | I | _ _ I I * 2 aj ~' ~ r
a. b. 2
An ,~~~~~~~~~~~V
16~~~~~~~~~~~~~~~~
M 1 Ad- | i- | ] | -1 18
Ct'~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
/4~~~~~~~~~~~~~~~~~~c
CO
10 -~ ~ ~ ~
'14~~~~~~~~.
~~~~~~~~?~~~~~ ;Wpi a
If Lw I Iz __I 2 0
as N
e4^ara
arm
0~~. . .
"4~~~~~~~~~~~~~~~~~~~~4
2~~~~~~~~~~~~~~~~~~~~~k~~~~~~~~~4