Association for the Study of Marble and Other Stones used in Antiquity



7th International Conference


             Thassos, Greece 15 – 20 September 2003



FOR INFORMATION ON THE full proceedings see HOME page







ThasSian Marble A Connection between ThasSos and Thessaloniki



Theodossia Stefanidou-Tiveriou



Aristotle University of Thessaloniki, Department of Archaeology, Aristotle University, GR – 541 24, Thessaloniki, Greece,







The manufacture of marble products in Thessalonike was directly connected with Thassian marble. Thassian marble was perennially and widely used in the city, especially in the Roman period. The question thus arises: in what form did the marble come to Thessalonike from Thassos, and did the marble-carving workshops on Thassos itself exert a morphological influence on the Thessalonian products?

I intend to investigate the connections between Thessalonike and Thassos, selecting one specific category of marble products that is abundantly represented in both places: the marble sarcophagus. If we consider: i) the marble, ii) the various types of sarcophagi and their decoration, iii) their dimensions and proportions, and iv) the unfinished products, we reach the following conclusions.

A type of sarcophagus that was especially common in the 2nd and 3rd centuries ad was widely used in Thassos and Thessalonike. The marble was worked at the quarries of the island into a rectangular cist with a pedimented lid, a shape that was very familiar from the similar, widespread Proconnesian products, which were exported all over the Roman Empire in an unfinished state, influencing the local workshops.

Those destined for Thessalonike were distinctly different from those intended for the local Thassian market. The Alyki quarries, mainly, received special orders from the Macedonian capital for sarcophagi with dimensions based on the Roman foot of 29.42 cm, and frequently had certain special features, such as a coherent base. Those for the local market were based on the local foot of 32.53 cm. There were similarities, however, in the ratio of height to depth, with two basic types: one with the height of the cist less than the depth and with a low lid; and the other with the height equal to, or greater than, the depth and with a correspondingly high lid. Neither group was completely standardised.

The Thassian products were sent to Thessalonike in a very roughly worked state. They were fashioned further in the workshop and thus acquired a more distinctive profile, as the type with a border was established. Thassos, by contrast, was usually content with the simple Proconnesian type of cist, but was also influenced to a limited extent by Thessalonian models. So, although it obtained its material from Thassos, the Macedonian capital was instrumental in determining the form of the final product.








Gerald A. Hess



Department of Art History, The Pennsylvania State University, 224 Arts II Bldg., University Park, Pennsylvania 16802 USA,






Hadrian’s estate at Tivoli (A.D. 125-134) is palatial and formal as well as revealing his own need for privacy and love of contrast in art and architecture. Masterpieces, several signed, exist among the many statues and reliefs of white marble and in coloured stone. With some irony, the villa’s contents may well have brought repose and yet pain. The old and young centaurs in dark stone, signed by Aphrodisian artists Aristeas and Papias, suggest a proud reliance on teamwork and amazing mastery as if a single artist was the creator. The centaurs served as foci to the garden room of Hadrian’s ‘Academy.’ Conceptually and physically each figure is planned to also draw attention to their detailed work, and the names of the two artists. The statues are more than eclectic and exaggerated figures. Haunting contrasts exist between the young and the old centaur and intensify emotion of the latter, making this the most effective Roman portrayal of this subject. A close parallel for the older centaur exists in the Louvre in white marble, possibly island marble, but it is a reduction of this theme, while revealing and preserving the tormentor, the Eros who rides on his back. Although scholars argue that the Tivoli pair may each have had an Eros as rider, the interplay between the young and old centaur reveals their plight and utter loss of the elder. Choice of material used for the sculptures is often overlooked and bronze is frequently credited as the ‘lost’ original and prototype for these works in white and coloured stone. Dark stone for the Roman sculpture has been summarily credited as ‘dark like bronze’ and chosen because of this bronze-like appeal. Close inspection reveals amazing attention to detail, appropriate to careful carving in marble of a variety of grain size, and fine treatment (polishing) of the surface to enhance the hue of the dark stone (Bigio morato); Aphrodisian sculptors delighted in a variety of marble and coloured stone. A monumental centaur in ‘Asian’ white marble provides us with a remnant of an aggressive figure for a major garden estate in Rome, Horti Lamiani. No single source or predecessor provides the features of the Tivoli centaurs but the most convincing monument to serve as a catalyst to their artists are found in Thasian reliefs of the mid 4th century B.C. on Samothrace and the centaur found in 1974 on Thasos (Estudes Thasien XV no. 23).  The Hellenistic original often claimed as model for the Tivoli centaurs never existed but rather we should seek to be informed by their artists’ selection and blending together of art and content as early as the 4th century B.C.









E. J. Walters



Department of Art History, The Pennsylvania State University, 224 Arts II Bdg., University Park, Pennsylvania 16803 USA   email:







            It is surprising that the Roman portrait identified as Julius Caesar on Thasos has been overlooked in discussions of the Roman imperial cult in the east. Scholars credit Augustus for devising and realizing a complex system of honor, which is recognized as the imperial cult. Yet, Julius Caesar fostered positive relationships in the east and laid a path for his nephew and adopted son Augustus, which the latter could reject, build upon, or tailor. The island of Thasos had an interesting position with the Romans in the first century B.C.  As credit to her enterprising citizens, valuable trade and constancy towards Rome, Sulla in 80 B.C. (Etudes Thasiens V no.174) granted to Thasos a special and favored status with regard to Rome. This special status definitely would be forfeit with the republican occupation in 42 B.C.  At Thasos the portrait of Julius Caesar at the earliest could date to his final year, 44 B.C. Although damaged, it seems unlikely that it would survive in any fragment with his opponents here in 42 B.C. Inscriptions preserved from Thasos honor members of Augustus’ family. A dedication of 16-13 B.C. (IGR I.835) draws attention to the Augustan women: his wife Liva as goddess, his daughter Julia and another Julia (II his granddaughter). They could be part of a family monument.  In particular, the demos, city-people of Thasos, refer to Augustus’ daughter Julia as benefactress through the efforts of her ancestors---this monument gives thanks then for reinstating the honors given to Thasos by Julia’s ancestors. Would the Thasian Julius Caesar be part of this memorial?  This question is an open one, but the study of sculpture of Roman date in Athens, Thasos and of course, Rome offers support, visual evidence for the Thasian Caesar to date in the later first century B.C. If this monument pre-exists the Augustan family memorial, then Thasos honored Julius Caesar as general with Roman imagery, possibly Julius’ chosen image as military leader in the old Roman sense. Of interest is the use of the oak wreath, corona civica, on the Thasian Julius Caesar: it is not a Julio-Claudian promotion and a mechanism with which to date this Thasian head to the 40’s A.D. as has been thought. Was Roman interest in Thasos rooted in Sulla, better asserted by Julius Caesar and finally maintained by Augustus, possibly through his family’s intervention? A competition of sorts may have resulted in policy that fortunately benefited Thasos. The extant solemn image of Julius Caesar could prove Thasian constancy to Rome.









Georgina E. Borromeo1, John J. Herrmann, Jr.2, and Norman Herz3



1 Museum of Art, Rhode Island School of Design, 224 Benefit Street, Providence, RI 02903,

2 Museum of Fine Arts, 465 Huntington Ave., Boston, MA 02115,

3 Department of Geology, University of Georgia, Athens, GA 30602,







In preceding ASMOSIA conferences, John Herrmann, Richard Newman, and Robert Tykot have shown that dolomitic marble from the island of Thasos was used by sculptural workshops virtually everywhere in the Mediterranean basin during Roman Imperial times. Nonetheless their art historical analyses have shown that the identification of Thasian marble can at times lead to recognizing the work of sculptors actually from Thasos or its province of Macedonia.

Another such case may be provided by a puzzling portrait head of the Emperor Hadrian in the Rhode Island School of Design (accession number 59.050).  The head had been in at least two English collections before arriving in Providence in 1959.  The English provenance suggests that the head originally came from Italy, where most Grand Tour souvenirs were collected, but this cannot be considered certain.  The Providence head is close to refined types of portraits of Hadrian usually created and preserved in Italy. It is perhaps best considered a variant of the Capitoline Imperatori 32 type, with the addition of anomalous independent features. The odd combination of unusual features has led sholars to argue that the portrait is a forgery, or reworked. A subjective impression that the marble was also strange played a role in denigrating the piece. Brunilde S. Ridgway has hypothesized a provincial origin in Asia Minor, but she was not able to reach a definite conclusion about the portrait head.

The marble can now be definitively shown to be dolomitic marble from Cape Vathy on Thasos on the basis of x-ray diffraction and isotopic analysis. While the head differs significantly from the statue of Hadrian on Thasos, it has a number of features in common with portraits on the Macedonian mainland.  The Providence Hadrian is suavely realistic in general, but it presents abbreviations and abruptly linear features that recall first and second century portraits in the museums of Thessaloniki and Polygiros. Locks of hair are stylized as simple pointed ovals split by broad furrows, the back of the hair is unfinished, and the upper lip is defined by a pair of sharply geometric arcs.  These provincialisms suggest that a sculptor from Macedonia, where Thasian dolomite was the native sculptural material, was responsible for the work.  The sculptor may have been working in Italy, but it seems more likely that he was attempting to follow Italian standards in his Macedonian homeland.









J. Clayton Fant1, S. Cancelliere2 L. Lazzarini3, M. P. Martinez4, B. Turi5



1 Department of Classical Studies, Anthropology and Archaeology, University of Akron, Akron OH 44325-1910,

2 L.A.M.A, Dipartimento di Storia dell’Architettura, Istituto Universitario di Architettura di Venezia, San Polo 2554, I-30125 Venice, Italy

3 L.A.M.A, Dipartimento di Storia dell’Architettura, Istituto Universitario di Architettura di Venezia, San Polo 2554, I-30125 Venice, Italy

4 Dipartimento di Scienze della Terra, Universita di Roma, Piazzale Aldo Moro, 5, 00185 Rome, Italy

5 Dipartimento di Scienze della Terra, Universita di Roma, Piazzale Aldo Moro, 5, 00185 Rome, Italy







Samples were taken from various parts of the outdoor triclinium, at ground level facing west. This masonry structure, unpushlished, is associated with the better-known glass mosaic nymphaeum (Rediscovering Pompeii (Rome 1992), no. 194 p. 270ff.). In its first phase, the triclinium was nestled between walls painted in opulent finto marmo. The painting style suggests 4th, and since the extensive renovation of the triclinium cuts into the painted walls it must postdate the painting. The renovation must be after the earthquake of AD 62, and hence the wall painting must belong to the earliest phase of 4th Style at Pompeii. It is striking to find finto marmo used in such close, and unabashed, association with genuine marble; the fact that the tricliniium uses only white marbles and the wall painting represents polychrome ones eases the clash.

To the naked eye, three or possibly four different kinds of white marble are used to face parts of this triclinium. Results analyzed by L. Lazzarini and B. Turi and associates will allow definite identification of the marbles and exploration of the issues of taste and market availability that went into their selection and combination.









John J. Herrmann, Jr.1 and Robert Tykot2



1 Museum of Fine Arts, 465 Huntington Ave., Boston, MA 02115, USA;

2 Department of Anthropology, University of South Florida, 4202 East Fowler Ave., SOC107, Tampa, FL 33620-8100, USA;







One of the most admired forms of architectural ornament in late Roman and Early Byzantine times is decoration with fine-toothed acanthus leaves. Such leaves are carved making use of rows of drill holes to suggest finely serrated leaves. The technique has a long history in Roman architectural ornament, beginning in middle Imperial Italy, where oak leaves or shaggy acanthus leaves were given sparkle with drill holes. In its later form, the rows of drill holes are more regularly patterned, more decorative, and less realistic. Capitals with fine-toothed acanthus have been attributed to Proconnesus/Constantinople, and pilaster capitals with fine-toothed acanthus were produced – presumably using local marble - in Aphrodisias from the second through the fourth centuries. Kramer has attributed pilaster capitals with fine-toothed acanthus to the Dokimeion quarries.

Capitals with fine-toothed acanthus leaves were sampled in Perge, in the Museum of Fine Arts, Boston, and in two private collections. The grain and color of the marble was observed visually. The stable carbon and oxygen isotope ratios of the samples were determined by mass spectrometry and the results compared with published databases. The results confirm that the quarries of Dokimeion requently provided the material for pilaster capitals of the third and fourth centuries with fine-toothed acanthus leaves. A contemporary pilaster capital with a fine-toothed flower on its abacus, however, proved to be Proconnesian marble. The carvers of Dokimeion evidently did not have a monopoly on the fine-toothed technique, but there may well have been a preference for their marble and their workmanship in pilaster capitals whose main leaves were fine-toothed. The Dokimeion and Proconnesian pilaster capitals are types that were exported from Asia Minor to Italy.









J. Pollini



University of Southern California, Dept. of Art History, Von Kleinsmid Center 351, L.A., CA 90089-0047: Tel: (310) 838-4363







In 1988 a high quality marble portrait head of Augustus was discovered 1.5 m above the ancient beach of Herculaneum.  This still unpublished portrait is important for several reasons: the unusual manner in which it was worked for insertion into another marble element, the way it combines two of Augustus’ portrait types, and what it tells us about the last hours of Herculaneum.

The angle of the neck to the partially flattened base of the tenon is extremely acute, suggesting that the head was not set into an upright statue body. Considered here are other possibilities for display, especially the adding of the head to a marble shield bust. Typologically, the head does not replicate a single type, but combines elements of two of Augustus’ types (Type IV= Louvre MA 1280; Type V = Prima Porta).  Although the head can be securely dated between 27 B.C. (creation of Type V) and A.D. 79 (eruption of Vesuvius), stylistic considerations place it ca. 20-1 B.C. New studies of the stratified layers of volcanic ashes and debris and of numerous skeletal remains from a vaulted area on the beach-front permit us to determine the precise time of the pyroclastic surge that sent the head and other heavy objects air-borne in the direction of the beach. This portrait most likely came from the “Area Sacra” across from its findspot, since this precinct was renovated in the Augustan period, with one of the temples being dedicated to Venus, the gentilicial goddess of the Augustus House.









A. Jones



American Museum of Natural History, Anthropology Department, Central Park West at 79th Street, New York, NY  10024,







The Arthur M. Sackler Museum of the Harvard University Art Museums recently acquired an under life-size Roman marble statue of a young boy (1999.231).  The statue is known to have been in the collection of the Earls of Pembroke at Wilton House, England, by 1731 and may have arrived in England as much as a century earlier.  It was among the first of thousands of sculptures in the classical style that filled England’s private estates by the end of the 18th century. 

Examination and technical analyses revealed the sculpture to be reconstructed from eleven fragments of at least 3 differing marble types using iron pins and clamps, lead sleeves, tree resin-based adhesives, lime mortar, and in the single clearly modern repair, a stainless steel pin.  All but the one modern repair appear to have remained undisturbed since the time of the statue’s arrival at Wilton House.  These repairs illustrate 17th century Italian restoration techniques described by the Roman sculptor-restorer Orfeo Boselli.

Analysis of the marble fragments by visual inspection, Carbon-Oxygen stable isotope analysis, petrographic thin-section, and x-ray fluorescence spectroscopy confirmed that the greater part of the statue was carved from a single piece of marble, from the neck to the base, and excluding the head, arms, and right lower leg.  The head, arms, and leg are all seemingly of different marble types and not original to the body proper.  The head is probably of Parian marble, while the body marble is as likely from East Greek quarries (Marmara or Denizli) as Classical-period Carrara quarries.  At least 3 and as many as 6 marbles were used.  Forthcoming analysis by ICP-MS (induction coupled plasma mass spectrometry) should further clarify the source quarries for the different marble types.

Stylistically, the body has been dated to the late 1st  – early 2nd century AD, while the head is thought to be of the late 2nd – early 3rd century AD.  Though modest, the statue is of particular interest for the historical record it contains in the techniques and materials of its restorations.  It may also be an example of an “ancient” statue that was effectively created in its first restoration from available, disparate excavated marble fragments. Statues similarly composite in nature supplied a Northern European antiquarian market in the 18th century for Roman statues, that was often less concerned with a statue’s degree of restoration than that it was complete.







 Patricia A. Butz

Department of Art History, Savannah College of Art and Design, P.O. Box 2072, Savannah, GA 31402-2072 USA





The base of the colossal statue of Apollo, dedicated by the Naxians on the island of Delos in the early sixth century B.C., carries two inscriptions, ID 4 and ID 49. The first is concurrent with the installation of the sculpture; the second is dated to the fourth century B.C. and is associated with the circumstances of its restoration. The first, unlike the second, is not a standard, formulaic dedication and despite its brevity has long attracted the attention of scholars for its peculiar wording about the nature of the stone used for the monument. It has commonly been translated, “I am of the same stone, statue and base oldest known for the Archaic period, and addresses the variety of meanings that have been offered for it. The proper understanding of “same stone” will be addressed and, consequently, the significance such an inscription places on the nature of the stone and the manufacture of the monument as being the preeminent pieces of information imparted to the ancient viewer. Indeed, the assertion “same stone,” made in the first person, demands the examination of spectator relations in respect to the work and its material content across time, including our own understanding today of the marble and its provenance.

The issue of spectator relations is especially relevant since a second inscription was added later, marking the preservation of the monument as a whole by means of its simple rededication. The paper considers this second inscription, its textual and palaeographic relationship to the original, and the historical context for the destruction of the monument that prompted the repair. In conclusion the paper will show how both inscriptions give strong evidence for the importance of the inscribed base and, thanks to ID 4, the early establishment of the principle of Greek free-standing sculpture as always consisting of a unified duality, statue and base.






Investigations on marbles and stones used in Augustean monuments of Western alpine provinces



Alessandro Betori1, Maurizio Gomez Serito2, Patrizio Pensabene3



[1] Soprintendenza per i Beni Archeologici del Lazio, Roma, Italy,

2 Politecnino di Torino, Torino, Italy,

3 Dip. Scienze dell’Antichità, Università di Roma “La Sapienza”, Roma, Italy,







Also in these provinces the Augustan policy of Romanisation produces urbanistic and architectural developments based on architectural and decorative patterns coming from Rome. Particularly (ciò vale per) monumental types as city-walls gates and celebrative arches, because they manifest immediately the new power: celebrative and propagandistic messages are induced not only through figurative scenes or statues but also through architectural decoration and (l’uso di particolari) building materials. We can observe very soon decorative forms that imitate patterns of Rome (Augustan Arches of Aosta, around 25 B.C., and Susa, 9/8 B.C.) and dialectic relations between (l’impiego di) Lunense marble and local marbles and stones which are similar to it (Foresto marble), or, in any case, can be employed as substitutive (Chianocco stone). We investigated in particular (the?) Praetoria Gate in Aosta, not only to identify the marble used in entablature, but also to study the coloured local stones of good quality employed on its façade and in Augustan Arch too. Local stones were used also for the architectural decoration of important private buildings as the Augustan Villa of Almese, at the frontier between Italia and Alpes Cottiae district, whose planimetrical structure is based it same on middle-italic models.







Fulvia Bianchi1, Matthias Bruno2, Andrea Coletta3



1 Via Monte Pertica, 21, 00195 Rome, Italy,

2 Via del Pellegrino, 130, 00186 Rome, Italy,

3 Via Flaminia Vecchia, 785, 00191 Rome, Italy,







The different imagines of Colosseum (coins and reliefs) give us a more complete idea of what is still visible today of the original interior aspect of the greatest amphitheatre of Roman Empire, destroyed in the centuries by earthquakes and spoliation. The study of a great number of architectural elements of the cavea and the upper portico kept in the storerooms offers the opportunity to reconsider the original interior aspect of the monument and to clarify aspects of building programs and activities and ancient restorations as distinctive feature of publica magnificentia.

The research is based on a systematic archaeological, quantitative and qualitative investigation of seat- blocks, column bases, shafts and capitals, which has permitted to determine macroscopically the utilisation of different marble qualities. Luna was used only for the seat – blocks of the cavea, proconnesian, pentelic and again Luna marble for capitals and bases, while proconnesian, karystian and different granite qualities for column shafts of the upper porticus.

Several restorations and re-building activities are testified by different typological groups of architectonic elements, which have permitted to reconsider their original distribution in the upper portico of the cavea in relation to their working techniques too. In fact the several holes present on the upper surface of all capitals are due to their different functionality and to the probable presence of marble architrave despite of the large intercolumniation width.







O. Palagia1, Y. Maniatis2, E. Dotsika2, D. Kavoussanaki2



1 Department of Archaeology, Athens University, 157 84 Athens,

2 Laboratory of Archaeometry, “Demokritos”, 153 10 Agia Paraskevi Attikis,







The pedimental sculptures of the temple known as The Hieron in the sanctuary of the Great Gods on Samothrace were excavated by the Austrians in the 19th and the Americans in the 20th century. The fragments are divided between Vienna and Samothrace. Phyllis Lehmann, The Hieron (1969) argued that the north pediment, facing the sea, carried the nurturing of Aetion, founder of the mysteries of Samothrace, whereas the rear pediment was decorated with busts of the Great Gods. She identified the marble as Thasian and dated the sculptures to the second century B.C. Her reconstruction has been criticized for the inclusion of the central akroterion as a principal figure in the north pediment. As a result of the discovery of further fragments both in Vienna and Samothrace and thanks to scientific analysis of the marble provenance, all of Lehmann’s conclusions are here challenged.

A sample taken from one of the statues in Vienna was analyzed using stable isotopes and maximum grain size. It was found to fall in the overlapping area between Paros and Proconnesos. The use of Proconnesian marble on Samothrace, however, is undocumented, whereas Parian is encountered in other sculptures, e.g. Nike of Samothrace. The scene in the north pediment is here interpreted as the presentation of the infant Dionysos to the nymphs, and a reclining statue in Samothrace Museum is attributed to the south pediment. The pediments are tentatively dated to the turn between the fourth and third centuries.




Provenance investigation of some marble sarcophagi from Arles with STABLE ISOTOPE AND maximum grain size analysEs



V. Gaggadis-Robin1, C. Sintès2, D. Kavoussanaki3, E. Dotsika3, Y. Maniatis3



1 Centre Camille Jullian - CNRS - MMSH - 5, rue Château de l'Horloge - BP 647 - F- 13094 Aix-en-Provence - Cedex 02 - Tél: 00 33 4 42 52 42 71 - Fax: 00 30 4 42 52 43 75 - mail:

2 Musée de l'Arles Antique - Presque île du Cirque romain - F - 132000 Arles - tél : 00 33 4 90 18 88 88

3 Laboratory of Archaeometry, NCSR Demokritos, 153 10 Aghia Paraskevi, Attiki, Greece, e-mail:







This paper presents the results of scientific analyses of marble samples from five coffins and of one lid of sarcophagi of the Museum of Arles in South France. They can be dated in the second, or in the third century A. D. The aim of this investigation is to identify the origin of some imported sarcophagi and to clarify if the local workshops carved marbles blocs in local or imported materials. The results are expected to give valuable information for the catalogue of these objects, which is in preparation.

The up to now results indicate that, the lid of the famous attic Phaidra’s sarcophagus is made in pentelic marble, as well as the coffin, and not in Carrara marble, as it was suggested by a scholar. The coffin of the little girl Chrysogone has been produced by an Asia Minor workshop and the analysis shows that its marble is Proconnesian. Its form and decoration (inscription, twisted columns and Medusa’s heads) agree with a Proconnesian provenance.

The analysis included some sarcophagi whose style and iconography cannot suffice to clarify their provenance. The analytical results for the Cornelia Iacaena’s coffin place the origin of its marble to the overlapping field between Proconnessus and Paros. However, its décor, the inscription and the garlands with Herakles knot carried by ram heads, brings it near again to Proconnesian workshops.

Concerning the local workshops in Arles, up to now it was considered that they carved only local limestones. Nevertheless this scientific investigation proves that they used also imported marbles. The material of Attia Esyche’s complete sarcophagus is Proconnesian marble for the coffin and Proconnesian or Paros for the lid. The low relieves carved on it: eagles carried garlands with Medusa’s heads in the centre, the unskilful flying Erotes holding the inscription, show that the sculptor is local and used an imported marble.

The uncommon scene of the principal view of the coffin with Eros and Psyche, the heavy execution of the bodies and the geometric aspect of the garlands suggest a product of a local workshop. That one worked on imported marble blocs, from Asia Minor, too. The analysis puts the marble just outside Proconnesian field, which with a high probability this marble is from Procennessus.








J. Trilling



University of Vienna (after July 1, 2003:  39 President Avenue, Providence, RI 02906)







My subject is the use of colored marble slabs in the nave of Hagia Sophia/ Ayasofya in Istanbul, built by the Emperor Justinian I between 532 and 537.  The decoration consists of at least eight varieties of marble in the modern sense, plus alabaster (“onyx” to the Byzantines) and porphyry, applied with extraordinary lavishness and attention to the intrinsic patterns of the stone.  In no other work of ancient or medieval art are technical aspects of marble – where it comes from, how it is quarried, what it looks like in its natural state – so closely bound up with artistic issues of creation and perception.

The Byzantines approached marble in a different way from their Roman predecessors.  Their great contribution was book-matching, a technique in which a rectangular panel is sliced in half parallel to its surface.  The two thinner panels that result are “unfolded” and set edge to edge, like the facing pages of a book when it is opened.  When certain kinds of marble are book-matched, their veining creates dramatic symmetrical patterns in two or more colors.  The decoration of Hagia Sophia is the supreme expression of this technique. 

Like other examples from the same period, known only from descriptions, the marble panels in Hagia Sophia were regarded not just as rare and precious building materials but as pictures with recognizable subject matter.  We know this from the verse ekphrasis (rhetorical description) of the church by Paul the Silentiary, composed in 562 when the building was restored after an earthquake.  Paul presents the interior of Hagia Sophia as an image of the world, not symbolically or generically but in specifics:  the dome of heaven, a grass-covered hill, a stream with flowers growing on its banks, and “wheatfields and sheltering woods, playful flocks of sheep and gnarled olive trees, spreading vines… and the blue-grey sea…”  The dome of heaven is obviously the dome of the church itself, and the wavy blue-grey marble of the original floor, some of which is still in place, must have been the sea.  Between these extremes the correspondence are harder to pin down, especially since the original decoration of the upper part of the church has disappeared.  Nevertheless, it is almost certain that some of the features enumerated by the poet correspond to the marble panels in the nave.  Even with the text to help us it is far from clear which marbles evoke which scenes.  In another passage, Paul the Silentiary speaks of “the pale onyx (i.e. alabaster) with glint of precious metal,” but to modern eyes there is nothing metallic about it, whereas the panels of Carystian marble, about which he makes no such observation, seem at close range to be flecked with silver.  An essential step toward understanding how the Byzantines looked at marble is recognizing that they did not see it the way we do.







Lisa Cooke1 and Ian Thomas2



1 Oxford University Museum of Natural History, Parks Road, OXFORD. OX1 3PW,

2 Ian Thomas, National Stone Centre, Porter Lane, WIRKSWORTH. Derbyshire. DE4 4LS,







The early 19th century collection of decorative stones and minerals put together in Rome by Faustino Corsi consists mainly of antique Roman and Italian coloured marbles. It is a very individual collection and has some surprises; one of the greatest is the inclusion stones from England. Coloured limestones, fluorspars and other veined minerals were the personal gift of William Spencer Cavendish, 6th Duke of Devonshire, from his estates in Derbyshire. This paper will discuss the mines and quarries from which the stones were obtained as well the circumstances of the gift. The Duke also took a great interest in coloured marbles and used many fine examples of Derbyshire and antique Roman stones in the internal decorative scheme of his great house at Chatsworth.








J. A. Harrell



Department of Earth, Ecological and Environmental Sciences, University of Toledo, 2801 West Bancroft St., Toledo, Ohio 43606-3390, USA (







A previously unknown quarry was recently discovered in Wadi Abu Bokari (a.k.a., Bakari or El-Bakriya; 250 15.15' N, 330 45.51' E) in Egypt's southern Eastern Desert. This site is associated with a gold mine that dates primarily to the ancient Egyptian New Kingdom, and the Ptolemaic and Roman periods. Based on tool marks and pottery, the quarry may be either Ptolemaic or early Imperial Roman in age. Additional fieldwork in summer 2003 will hopefully refine the dating. The quarry stone, which is a grayish medium-grained (< 5 mm) granodiorite, is not recognized as one of the ornamental stones used by the Romans. It has a superficial resemblance to the tonalite gneiss (granito del foro) from the famous Roman quarry at Mons Claudianus in Egypt. If used by the Romans, objects carved from the granodiorite may thus have been mistaken for the gneiss by scholars. Images of polished slabs of the granodiorite may be seen on the author's web site at The quarry consists of fifteen open-cut excavations that are scattered over an area of half a square kilometer. The largest of these is 25 m across with the others ranging between 4 and 13 m. Within the quarry are four well-cut blocks that appear to be pedestals or trapezophori for supporting either stone basins or table tops. These are 80 to 90 cm high with a square (84 to 88 cm on a side), vertical-sided (14 to 21 cm high) base and four sides that taper gracefully to a flat, square top (47 to 50 cm on a side). Other worked blocks, all rectangular, are found within the quarry as are also lines of pointillé pits and primitive wedge holes.





Characterization of Calcarenite Quarries in East Crete: Sourcing Ashlar Blocks at Gournia



S. Pike and J. Soles



Lynchburg College, Environmental Science Program

1501 Lakeside Drive, Lynchburg, VA  24501, USA







Large ashlar blocks found at several Late Minoan palace sites in eastern Crete are composed of lithic material known locally as ammoudhopetra or beachrock.  This building material is geologically identified as biocalcarenite, a sedimentary rock comprised primarily of calcareous clastic grains.  Biocalcarenite quarries located near the neopalatial sites of Zakros, Mochlos, Mallia and Palaikastro are reasonably presumed to have supplied the ashlar blocks needed for their respective local building programs.  The Late Minoan I (LMI) modification of the administrative center at Gournia also incorporated several large ashlar blocks of biocalcarenite.  Because there are no exposures of this rock in the immediate vicinity of Gournia samples were collected from the known ancient ammoudhopetra quarries in eastern Crete to determine the likely source of the Gournia ashlar blocks.  Petrographic techniques were employed to create a database to distinguish between the various biocalcarenite quarries.  A petrographic analysis of the Gournia ashlar blocks was than compared to the database.  The results indicate that the biocalcarenite quarry at Mochlos supplied the ashlar blocks for the refurbishing of the administrative center at Gournia.








Adel Kelany



Supreme Council of Antiquities

Aswan, Egypt (







The Aswan quarries are the source of red granite used for ornamental and building stones from the Old Kingdom through the Roman period. The famous Unfinished Obelisk Quarry, located on the east bank of the Nile River and covering an area of 510 m x 340 m, and has recently undergone archaeological excavation by the Supreme Council of Antiquities.  This quarry reveals evidence of exploitation specific to the New Kingdom and the Roman period, and for the former, this relates to the extraction of red granite for obelisks and colossal statues.

The preliminary results of these excavations have produced fresh insights into hard-stone production techniques and the logistics of obelisk and colossal statue extraction that until recently have been poorly understood.  In particular, there is clear evidence for the comprehensive use fire-setting for trimming blocks with the use of mudbricks to concentrate and control the fire.  This paper also presents a general discussion of the stone tools and other artefacts that relate to both the New Kingdom and Roman period extractions with some preliminary results from this study.  Important epigraphic evidence is also described, including rock drawings of ostrichs, fish, boats and obelisks plus many mason's lines. And finally, a newly discovered inscription is discussed that mentions the dispatching of two obelisks from the quarry to Karnak Temple (in modern Luxor) during the reign of king Tuthmosis III of the New Kingdom's 18th Dynasty.









E. Bloxam1, P. Storemyr2, T. Heldal3



1 Institute of Archaeology, University College London,

2 Expert-Center für Denkmalpflege, CH-8005 Zürich,

3 Geological Survey of Norway, N-7491 Trondheim,







The Egyptian Old Kingdom was an epoch in which the procurement and transportation of hard stone from sources over a wide geographical area was practised par excellence.  The incorporation of hard stones in royal pyramid complexes, such as basalt for temple paving and anorthosite gneiss for royal statuary, marked an expansion in hard stone consumption to include stone sources outside the Nile Valley.   The social context in which these expeditions were undertaken is on the whole quite poorly understood, this is mainly due to the absence of relevant textual sources. Generally, theories about the social organisation of quarry expeditions in the Old Kingdom have erred towards low-levels of organisation and production being practised in rather an ad hoc manner. Counter positions have come from extrapolating the social organisation of quarry expeditions from Middle and New Kingdom (2nd millennium BC onwards) epigraphic sources. These accounts have expounded ideas of quarry expeditions as highly organised enterprises structured along rigid hierarchical lines and involving large numbers of people into several thousands.

This paper presents a re-evaluation of these ideas based on recent archaeological survey and excavation of two Old Kingdom hard stone quarries outside the Nile Valley in the Western Desert: the Widan el-Faras basalt quarry in the Northern Faiyum Desert, Lower Egypt and Chephren’s Quarry (anorthosite gneiss) in Upper Egypt. Analysis and interpretation of production methods, settlement data and ceramic evidence from both these quarries will be presented in a comparative framework.  The paper concludes by proposing that the social organisation of Old Kingdom quarry expeditions, outside the Nile Valley, were small-scale operations involving specialists, well organised through non-hierarchical kinship ties and mobilised for specific projects. Furthermore, the study demonstrates how fresh insights into the social context of Old Kingdom quarry expeditions can be advanced via a comparative research methodology, given the absence of reliable epigraphic sources.








T. Endo, S. Nishimoto



Research Fellow of the Japan Society for the Promotion of Science

Waseda University, Department of Architecture

2-13-3-101 Tsuji Minami-ku Saitama-shi Saitama-ken, 336-0026 JAPAN







The limestone quarry at Dibabiya is located 35 km south of Luxor on the eastern bank of the Nile, opposite Gebelein. There are two rock-cut galleries in the mountain. The authors conducted a survey in the summer of 2001 to seek out important traces and inscriptions that were left by ancient workmen and that have architectural significance. The present state of the quarry and some of the survey’s results will be reported.

The northern gallery had a stela of the scribe Huy, though completely destroyed now, which showed that stone blocks from this quarry were used for the construction of the memorial temple of Seti I at Qurna. A number of parallel lines drawn in red ink are visible on the ceiling. Between these lines were found hieratic inscriptions dated from “abd 3 šmw sw 27” (the third month of summer, day 27) to “abd 1 3ht sw 23” (the first month of inundation season, day 23), as well as several marks made in various places by workmen. The interval between the red lines is approximately 20 cm, and the parallel lines are divided by short lines at almost equal intervals of ca. 80 cm.

In contrast, the southern gallery had a stela noting that the king Smendes sent 3,000 men to this quarry to produce stone blocks for the repair of the Luxor Temple built by Thutmose III. Numerous red lines and some faint inscriptions are preserved on the ceiling. Unlike the lines in the northern gallery, these lines are quite straight, and the alignment of mason's marks within the cells is characteristic. The cells of the grid formed by the red lines are ca. 80 cm x 20 cm, resembling those in the northern gallery and the quarry at Qurna.

These traces found at the quarries could be some kind of administrative record for marking the progress of quarrying works, and not all of them have been published in detail yet. Although additional investigations will be needed in the future, this information is expected to shed new light on the work organization for excavating stone blocks at quarries in New Kingdom Egypt.

The project is partly supported by a grant from the Japan Society for the Promotion of Science.





Pharaonic Limestone Quarries in Wadi Nakhla/Egypt



D. Klemm1, R. Klemm2



1Department für Umwelt-und Geowissenschaften der Universität, Buzallee 32

D-86911 Dießen, München, Germany,

2Institut für Ägyptologie der Universität, Buzallee 32

2 D-86911 Dießen, München, Germany,







Wadi Nakhla, east of Deir el-Barsha in middle Egypt is a relative small wadi within the Eocene limestone escarpment between Beni Hassan in the north until about the plain of Amarna in the south. Until today the best localization map of this outstanding quarry district was published in the fundamental work of Newberry (1895), only slightly improved by these authors (Klemm and Klemm 1993).

Stratigraphically the entire units used for ancient tomb constructions or quarry activities are part of the Beni Hassan member of the Minia formation at the boarder of Lower to Middle Eocene of the Palaeogene period of the Tertiary system. These units are overlain from Sheikh Ibadan on toward the north by the nummulites-limestones of the Samalut formation of the Middle Eocene Mokattam group.

One of the most prominently mined wadis of this region is Wadi Nakhla with huge gallery quarries up to 250 m deeply driven into its flanks.

The Beni Hassan member in Wadi Nakhla is formed of a rhythmic layering of cycles built up by more or less dense packstone-wackestone strata, passing into coral beds, weathering in a type of sponge structure and interstitially filled with packstone –wackestone biosparite (a fine grained secondary calcified original sedimented mikrite). All together 17 such cycles could be separated in this Wadi Nakhla sequence from which only about 3 attracted the interest of the ancient, due to their relative thickness and resistivity to deterioration. One layer was especially intensively mined or used for tomb construction during various periods of Egyptian history: Layer 12, according to our numbering from the base of the wadi toward the top of quarry activities in Wadi Nakhla.

Quarrying and tomb constructions in Wadi Nakhla started at least in Old Kingdom with a main emphasis in Middle Kingdom, a reopening of gallery quarries in New Kingdom and a maximum in Nectanibo times.








Hadjidakis P.1, Matarangas D.2, Varti- Matarangas M.2



1Delos Museum, 846 00 – Mykonos,

2IGME, 70 Messogheion str. Athens 115 27







Delos must always raise some astonishment when one compares its size to its history.  Although a small, rocky island, no more than 5 Km long and 1.300 m. wide, for ancient Greeks it was the most sacred place, because Apollo and Artemis, two of the most important deities of the Greek pantheon, were born there. It is estimated that at the peak of its glory, around 90 BC, about 30.000 people lived there.  Rich merchants, bankers, and ship-owners from all over the world settled there, attracting many builders, artists and craftsmen who build for them luxurious houses, using mainly locally quarried stones. The small island became soon the maximum emporium totius orbis terrarum - the greatest commercial centre of the whole world.

A couple of years ago was undertaken a Geological study of the island aiming mainly to trace the origin of the building material, in order to help towards a better understanding of the monuments structure and finally to assist their restoration and conservation.

Delos, an island belonging geologically to the Attico-Cycladic complex, consists mainly of Miocene granitoid rocks, with common screens of metamorphosed country rock (meta-pelite, meta-psammitic, marble, amphibolite). In its northern part there are foliated rocks of granodiorite and from their mineralogy and geochemistry they appear to be part of the Miocene plutonic assemblage of the Cyclades. The main rock type of the central part is megacrystic granodiorite, while on the south biotite granodiorite predominates. Several shear zones hundreds of metres wide extend across the island and separate these rock types. Most of these shear zones consist of thin screens of metasedimentary rocks and parallel sheets of strained plutonic rocks. The Skardanas Shear Zone, in the north, separates flat-lying foliated granodiorite from the main megacrystic granodiorite of central Delos. This shear zone lacks metasedimentary rocks and consists of thin sub-vertical sheets of variably deformed tonalite, granodiorite and megacrystic granodiorite, with numerous late granite sheets oblique to the foliation, similar to those that cut the northern foliated granodiorite. The Kynthos Shear Zone is separated from the Skardanas Shear Zone to the north and the Fourni Shear Zone to the south by extensive megacrystic granodiorite. The Fourni Shear Zone in the southern part of the island contains highly strained rocks typically dipping 30o NW. This shear zone includes highly strained originally net-veined quartz diorite and tonalite, intercalated with less-deformed granodiorite, some containing both deformed and undeformed quartz diorite enclaves.

Eluvial mantle and alluvium and poros deposits are found in several places in Delos.

Several sites of small scale surface quarrying and also many ancient quarries were discovered during this study, indicating large-scale operations. Great quantities of marble, granite and porolithos extracted from these quarries were used for the construction of monuments, public buildings and private houses. In the ancient granite quarries close to the Fourni area, still remain unfinished columns and pilasters providing a lot of information about the means and tools they used to cut the stone.








K. Koller1, P. De Paepe2, L. Moens3



1 Inst. for Studies of Ancient Culture, Austrian Academy of Sciences, Baeckerstrasse 13 PF 8, A-1010 Vienna/Austria,

2 Dept. of Geology and Soil Science, Ghent University, Krijgslaan 281 S8, B-9000 Ghent/Belgium,

3 Dept. of Analytical Chemistry, Ghent University, Proeftuinstraat 86, B-9000 Ghent/Belgium,







Since 1998 non-dolomitic white marble from quarries in the surroundings of Ephesos and samples from selected monuments of the ancient city as well as sculpture in the depots and the museum are studied with the aim of determining their origin. The interdisciplinary project initiated by M. Aurenhammer, Austrian Archaeological Institute, is conducted by L. Moens and P. DePaepe from Ghent University in collaboration with the Institute for Studies of Ancient Culture and the Efes müzesi in Selçuk.

Apart from the marble quarries near the Mausoleum at Belevi, said to be located by the shepherd Pixadoros, which is mentioned by Vitruvius as he describes the use of marble as a building material for the Archaic Temple of Artemis Ephesia, there are many more quarries around Ephesos, most of these located among the Küçük Menderes, the antique river Kaystros. Various traces of antique marble extraction were found in the quarries, the best preserved at mountain heights, where the exploitation obviously stopped with the decrease of the city.

A first analysis (stable C and O isotope ratios) of samples collected from sculpture and building materials shows that about half of these are within the range of the datasets from the Ephesian marble quarries which is an expected result regarding the use of local marble for building purposes. Moreover an accumulation of archaelogically dated samples in different quarries, especially of building materials from the archaic and classical Greek Artemision and also from monuments of the Roman period allows a chronological separation of quarries and exploitation periods. The analysis of samples taken from sculpture however shows that the local marble was less important for sculpture in Ephesos but – as far as possible to say until now - marble imported from quarries like Aphrodisias or Dokimeion was applied.








Philip C. LaPorta



City University of New York, The Graduate Center,365 Fifth AvenueNew York, New York 10016-4309, U.S.A.







Prehistoric chert quarries in the Wallkill River Valley of northwestern New Jersey are distributed in the northeast-southwest trending Kittatinny Supergroup carbonate succession.  The Cambrian to Middle Ordovician carbonate sequence accumulated over a structural topographic high in shoreline, near shoreline, and shallow subtidal environments.  Cherts in the carbonate units occur as linings along unconformities and stylolite surfaces and as replacements of algal stromatolites, paleokarst features, porous oolitic sequences and evaporitic sequences.  Recently, chert has also been identified in association with bentonitic ashes and dilation breccias in zones of thrust faulting. 

The structural style of the Wallkill River Valley is modeled after the flat-ramp theory (Rich-type duplex) of thrust fault development.  The Kittatinny Supergroup was initially deformed by Ordovician-age, fault-related folds of the Taconian orogeny, which were consequently redeformed by Pennsylvanian-age thrust faults (thrust ramps and flats) of the Alleghanian orogeny.  Mapping of the Wallkill Valley quarries have elucidated two quarry varieties; one type located in the Alleghanian thrust ramp section and the other located in the thrust flat section.   Thrust ramp quarries are aligned along the northwest-dipping (30◦ to 60◦ dips) forelimbs of Taconic fault-related folds deformed by the Alleghanian thrust ramp section.  Thrust flat quarries occur in the shallow forelimbs (10◦ to 17◦ dips) of Taconic folds deformed by normal fault and back thrust zones located in the Alleghanian thrust flat section.  The structural deformation of the chert-bearing Kittatinny carbonates provides fundamental constraints on the development of the two types of prehistoric quarries in the Wallkill Valley.  Level of quarry development is structurally pre-determined by the inclination of the chert-bearing beds or by the quarries proximity to a Taconic fold-hinge.  Mapping has shown that full-scale quarry movements are predominant in the thrust ramp quarries, where strata are inclined between 30◦ and 65◦.  Chert-bearing strata inclined between 65◦ and vertical may also escalate from quarry motion to quarry movement.  Bedrock quarries developed in the thrust-flat section, in shallowly-inclined strata, are less likely to develop into motions or movements.  Much of this is due to the limitations on access from above through the accentuation of joint surfaces. 

Mapping of specific quarries, within these structural templates, has elucidated four principal zones of mining activity, the zones of extraction, ore milling, ore processing and ore refinement.  At the zone of extraction, joint-bounded blocks are levered from the outcrop surface.  Gangue or country rock are freed from the chert ore in the zone of ore milling.  In the zone of ore processing, chert blocks are dressed along joint and cleavage surfaces and sorted empirically by their mechanical properties.  Ore refinement occurs above the quarry face on a stable platform where bifaces, cores and flakes (half-products) are generated for distribution.  Additionally, a sophisticated mining technology fashioned from Silurian-age metaquartzites and metaconglomerates have been discovered in the quarries.  The mining instruments are associated with a distinct class of mine-tailings and the continuum of ore milling and processing describes an elaborate chain of operation present within a fully developed prehistoric quarry. 








T. Heldal1, P. Storemyr2, A. Salem3, E. Bloxam4, I. Shaw5, R. Lee6



1 Geological Survey of Norway, N-7491 Trondheim,,

2 Expert-Center für Denkmalpflege, CH-8005 Zürich,,

3 Egyptian Geological Survey and Mining Authorities, Aswan Office, Egypt,

4 Institute of Archaeology, University College London, London WC1H 0PY, UK,,

5 Dept. of Archaeology, University of Liverpool, Liverpool L69 3BX, UK,,

6 8 Elm Tree Road, Birmingham B30 2BN, UK,







Chephren’s Quarry is one of the world's oldest hard-stone quarries (anorthositic to amphibolitic gneiss), from which the famous life-size statues of Chephren and a large number of Predynastic stone vessels was made. The quarry area covers nearly 100 km2 of flat, hyper-arid desert, some 60 km west of Lake Nasser (River Nile) in the extreme south of Egypt. Lately, it has been discovered that this unregistered archaeological site of great importance is seriously threatened by modern development – the canal and irrigation works related to the South Valley Development Project. An initiative to a Norwegian-British-Egyptian project aiming to provide accurate maps and documentation of the Chephren's Quarry site was taken, and fieldwork was carried out in January 2003.

In total, five ancient stone quarry areas, an ancient track and a mine were mapped. This included approximately 700 ancient extraction sites (including 40 larger quarries) and 2-300 features related to the ancient quarrying infrastructure. The mapping was carried out with GPS linked to pocket computers, and the maps were compiled on laptops during the fieldwork using ArcPad and ArcView software. For each registration, a number of features were recorded and included in a map table structure, allowing compilation of different thematic maps, covering geological, archaeological and protection-relevant information.

The paper presents some of the map themes, and gives examples of how geological features can be applied in the archaeological interpretation of the site when combined with information about the quarries. Furthermore, GIS analyses demonstrate spatial differences in the use of tools, the distribution of statue and vessel quarries and the type of infrastructure. This indicates a development from vessel quarrying in the marginal parts of the Chephren gneiss deposits towards a central part dominated by statue quarries. GIS analyses falso allow interpretation of the quantity of the ancient stone exploitation. Finally, this study concludes by suggesting a model for the use of geographic information systems as a documentation and decision-aiding tool for the preservation, risk-assessment and management of ancient quarries, based on the experiences made.








P. Storemyr1, T. Heldal2



1 Expert-Center fuer Denkmalpflege, CH-8005 Zurich,

2 Geological Survey of Norway, N-7491 Trondheim,







Ancient quarries are often unique archaeological sites, which can give invaluable information about ancient technologies, trade relations and the daily life in antiquity. Such sites not only consist of stone extraction areas, but frequently also associated infrastructure like settlements, roads and tracks, stone working areas, smithies etc.

Like many other types of archaeological sites, ancient quarries are very vulnerable to damage and destruction. Such sites are often unregistered and frequently difficult to recognise in the landscape, especially for the layperson. The traditional threat has been new quarrying operations undertaken without regard to ancient traces. Moreover, many are situated in areas, which until recently were remote, but now have come under pressure from modern development projects and advancing infrastructure. Looting and vandalism is another problem, as well as natural hazards like landslides and – in desert areas – flash floods.

This paper will show development trends using examples from Egypt and Norway, two very different countries, but surprisingly having similar problems with regard to destruction of ancient quarries. Moreover, possible strategies for documentation, protection, management and presentation are briefly presented. Since ASMOSIA unites the international expertise on ancient quarries, it is also asked how this expertise can be used in the service of rescuing the sites on which the association relies.








M. Wurch-Kozelj, T. Kozelj



Ecole Franches Archéologique, Thassos, BP.35, 64 004,







Rectangular sarcophagi from the Roman period can be found in Thasos. Some of them are unfinished in the quarry, others are left just as they were ready to have been transported, many of them were found in archeological exacavations in differents states of work (unfinished through to fully complete sculpted, or ripped open and emptied­, broken) and a few are still in their intended places.

An examination of these rectangular sarcophagi demonstrates out the different phases of work, and shows explicity forward the reasons for keeping rough surfaces or/and uncarved parts of sarcophagus…









Tony Kozelj and Manuela Wurch-Kozelj



Ecole Franches Archéologique, Thassos, BP.35, 64 004, e-mail:,,







The exhibition “The Harbour of Ancient Thassos” (Το Λιμάνι της Αρχαίας Θάσου), during the “European Days of Cultural Heritage”, 4-5 October 1997 presented the harbour of Thasos: the commercial relations and differents laws in use in Antiquity. The wellknown and extensively published block of orthostate (Museum of Thasos, inv. 284), with its three parts of epigraphy, concerning the winetrade, was so well exposed that the details of its front surface, where the epigraphies were engraved, were completely visible.

Observation showed, not just the surfaces, but the different thicknesses in correlation with the three parts of epigraphy, which indicate the re-use of the block; keeping the lower part of the initial epigraphy, erasing the upper part (cutting down the initial surface of the block), engraving another epigraphy on this new second level surface; again erasing the middle part of the block to engrave a new epigraphy (on a third level surface).

The re-examination of the epigraphies, on the different levels of the block’s surface, the disposition of the letters…, allows new conclusions about the chronology of the writing.








M. Bruno



Via del Pellegrino, 130, 00186 Rome, Italy,







A few years ago over 1200 fragments of ancient sculptures of different periods, sizes, dimensions and marble qualities were systematically catalogued. Many of these fragments are joints used for the production of one complete assembled statue, apparently made of one single marble block, or for ancient sculpture restorations. The provenance of these fragments is mostly unknown, because during the modern excavation activities they were not inventoried and their finding spots were not exactly indicated.

Several fragments were discovered during the excavations of the ancient theatre of Ostia in 1912 - 1914,  where, in one of the taberne of the substructure, a small sculptor atelier, probably of late antiquity, was identified and many marble fragments of ancient statues, architectural fragments  and inscriptions were discovered stockpiled inside. The statue fragments here invented show clearly that they were destined to be re-used for new sculptural purposes or for restorations of damaged statues and that they were certainly get from different ancient ostian public monuments or private domus abandoned in late antiquity.

The systematic study of these sculptural fragments and joints throw new light on sculptural technique and ancient restorations as well as on late antique spoliation methods as shown by the stockpiled marble fragments in some tabernae of the theatre and the surrounding area.







G. Kokkorou-Alevras1, E. Poupaki2, A. Chatziconstantinou3, A. Efstathopoulos4



1 Department of Archaeology and History of Art, University of Athens, University Campus, Athens 157 84, Greece,

2 33, Athanassiou Diakou Str., Ellinikon, Athens 167 77, Greece,

3 4, Terpsihoris Street, Paleon Faleron, Athens 17562, Greece,

4 4, Ioanninon Street, Galatsi, Athens 111 46, Greece,







Since 2002, a research team of the Department of Archaeology and Art History of the Athens University, in collaboration with the geologist Mr. A. Chatziconstantinou, is working on the editing of a corpus of ancient Greek quarries. The main objective is the construction of a digital database that will include all evidence data of the research on ancient quarries. The project is divided into four phases, which can be described as follows:

Phase A. Recording, positioning and digitization of the already known quarries of antiquity (Greece, Asia Minor and Italian Peninsula), as well as identification of new ones.

Phase B. Accurate identification of the extracted material via the latest methods of analysis.

Phase C. Editing of a stone transaction network during antiquity.

Phase D. Final results: new information on the commercial, financial and political affairs among the different centers of the ancient world.

Up to now, the recording of 200 quarries in Greece and Cyprus has been classified, including reference material, such as: bibliography, published photographs, topographical plans and petrologic descriptions. Furthermore, the integration of geological information and stone analysis results, as well as the positioning of each site by means of GPS (Global Positioning System), are planned.

This paper discusses the methodological issues and problems that occurred during the design of this ancient quarry database, as well as its current status. Such issues are: multi-chronological classification; utilization of bibliographical sources and categorization based on quarrying traces or in situ findings. Finally, a live operation of the database will be demonstrated during this conference.





POlychrome hellenistic sculpture in Delos.

Research on surface treatments of ancient marble sculpture.

 Part II.



B. Bourgeois1, Ph. Jockey2



1 Centre de Recherche et de Restauration des Musées de France, 6, rue des Pyramides, F- 75041 Paris cedex 01

2 Université d’Aix-Marseille I, Maison Méditerranéenne des Sciences de l’Homme, 5, rue du château de l’horloge, F- 13094 Aix-en-Provence cedex







The paper deals with the ongoing research project on the polychromy of Hellenistic marble sculpture in Delos, done in collaboration between the French School in Athens, the Research and Conservation Center of French Museums and the CNRS. It will present the new results obtained since the last ASMOSIA VI conference, over the past three years. Approximately 90 objects have been examined by now, and the use of a video-microscope has allowed for a very acute, detailed and thorough investigation of the polychrome remains. Portable XRF analysis has also been performed, thanks to collaboration with A. Karydas, Demokritos Laboratory, and Haricleia Brekoulaki, Wiener Laboratory. By this mean, the characterisation of pigments, preparation layers, as well as gold leaf, has been achieved in a completely non-invasive, non-destructive way. The techniques and tastes involved in the painting and gilding of Delian marble statuary will be addressed, with particular attention being placed on important features such as: in most cases, evidence of lead white as a ground layer; a relatively limited range of pigments, (in contrast to wall painting practices for instance), now clearly identified for most of them, counterbalanced by a well attested and mastered technique of mixing pigments in order to obtain subtle pictural effects; a systematic combination of painting and gilding for the final kosmesis and chrysôsis of the sculptures, even in the case of small-scale minor production.








H. Brecoulaki1, A. Karydas2, P. Jockey3, B. Bourgeois4



1 University of Sorbonne (Paris I)

2 Institute of Nuclear Physics, NCSR “Demokritos”, 153 10 Aghia Paraskevi, Attiki, Greece

3 Université de Provence, Ecole française d’Athènes),

4 Centre de Recherche et de Restauration des Musées de France







The analytical characterization of immovable monuments such as marble sculptures and paintings requires at first the application of a non-destructive technique with instrumentation that can be adjusted in-situ, close to the area of interest. Furthermore, the technique should provide a fast, multielemental and sensitive analysis. The above requirements can be fulfilled to a great extent by the use of X-ray Fluorescence (XRF) analysis. The experience gained by the use of XRF technique for the characterization of pigments on the Hellenistic marble sculptures of Delos and on the paintings of ancient Macedonia funerary monuments will be reviewed, emphasising the specific analytical questions/problems that emerged for each case regarding the study of the pigments and techniques of application. Therefore, a comparative evaluation of the specific uses of painting materials in each context will be attempted. The role of other analytical techniques that can be also transported on site to complement or support the XRF data will be also discussed.









A. Mentzos1, V. Barbin2, J. J. Herrmann, Jr.3



1 Department of History and Archaeology, University of Thessaloniki, 54006 Thessaloniki, Greece;

2 Université de Reims Champagne-Ardenne, Laboratoire des Sciences de la Terre, Centre de Recherches Agronomiques, 2 esplanade Roland Garros, F-51100 Reims;

3 Museum of Fine Arts, 465 Huntington Ave., Boston, MA 02115, USA;







A few samples of high-quality architectural ornament in Macedonia were analyzed to supplement the results of previous testing. Small chip samples were taken, ratios of stable isotopes of carbon and oxygen were determined, and the samples were studied with cathodoluminescence.  A Late Roman sima decorated with palmettes at the Rotunda, Thessaloniki and an ornate Early Byzantine architrave from the Rotunda Museum both proved to be from Aliki, Thasos. These tests enrich the picture of ongoing creativity in the realm of architectural decoration by Macedonian carvers working with Thasian calcitic marble.

It has long been recognized on a purely optical basis that the early Byzantine churches of Phthiotic Thebes, modern Nea Anchialos, Magnesia, southern Thessaly, made use of architectural elements carved of Proconnesian marble and executed in the Constantinopolitan style.  Optical identifications of very large grained greyish marble also suggested to the authors that some of the marble at Nea Anchialos came from the calcitic quarries of Thasos. Isotopic and cathodoluminescence analysis confirmed the presence of marble from Aliki, and other Thasian quarries. The presence of Proconnesian marble was also verified. Thasian and Proconnesian marble was used for ionic impost capitals, corinthianizing colonnetes, and chancel barriers. Stylistic and typological similarities show that the same workshops worked on marble from both sources.  If workmen from Constantinople produced these architectural sculptures, they were present at or near the building site in Thebes.  Designs not yet attested at Turkish sites also indicate that Phthiotic Thebes was a significant creative center.









A. Calia1, M.T Giannotta1, L. Lazzarini2, G. Quarta1



1 CNR – IBAM –Istituto Beni Archeologici e Monumentali – Lecce, c/o Villa Tresca - Campus Universitario - Via Monteroni – 73100 Lecce (Italy),,

2 LABORATORIO DI ANALISI DEI MATERIALI ANTICHI, DSA-IUAV, San Polo 2468 Palazzo Badoer - I-30125 Venezia (Italy),







In the sea off Torre Sgarrata, situated on the Ionian coast South of Taranto (Puglia – South Italy) between 1965 and 1967 the scholar Throckmorton discovered the wreck of a Roman ship, which sank between the end of the II and the beginning of the III century AD, and recovered the cargo. This consisted of 18 sarcophagi and 17 blocks (both of white marble), and 6 slabs of alabaster. Archaeologists identified the quarries of Aphrodisia (Ward-Perkins, 1965) and Thassos (Parker, 1992) as the source of the marbles.

This paper presents the mineralogical-petrographic and geo-chemical studies that were carried out in order to characterise the white marbles and determine their provenance; another contribution (A. Alessio et al.), published in the present volume, takes up the typological, stylistic and metrological study of all the items in the cargo.

The samples taken from the sarcophagi (n.14) and from the blocks (n.15) were subjected to isotopic C13 and O18 analyses using a mass spectrometer, and mineralogical-petrographic analyses by XRD and polarised light optical microscope on thin section. The data obtained made it possible to establish that the marble artefacts belong to the well known to two distinct facies of the island of thassos: one of dolomitic composition (9 samples) from cape vathy, and one calcitic (20 samples) from Aliki’. It also emerged that the differing provenance of the marbles used for the sarcophagi reflects typological differerences: while both types are of rectangular shape externally, those of dolomitic marble have a rectangular cavity, and those of coarse-grained calcite have an oval cavity.









Donato Attanasio1, Susan Kane2, Rosario Platania3 and Paolo Rocchi1



1ISM-CNR, P.O.Box 00016, Monterotondo Staz. Roma, Italy,

2Dept. of Art, Oberlin College, Oberlin, Ohio 44074, USA,

3ISM-CNR, P.O.Box 00016, Monterotondo Staz., Roma, Italy,







Preliminary characterization of 319 architectural and sculptural marbles sampled at Cyrene in the past two years has detected five Thasian dolomitic sculptures, which went unnoticed or were assigned differently in previous studies. This finding has stimulated a more thorough investigation of the presence and use of Thasian marbles, both calcitic and dolomitic, at Cyrene as a contribution to more detailed knowledge of the distribution of this marble variety within the Mediterranean basin. Provenance determinations were carried out by discriminant function analysis of a properly chosen set of EPR and petrographic variables, according to a method already published. The overall outcome indicates, as expected, that the great majority of Cyrene marbles originate from the Paros and Pentelicon quarries (48% and 30%, respectively), with three other quarrying sites following at great distance (Proconnesus 9%, Thasos 6.3%, and Carrara 6.1%). Although, at this preliminary stage, the use of marbles from other sites cannot be excluded with certainty, their presence seems to be unlikely and, in any case, quite sporadic.

Twenty Thasian samples have been identified at Cyrene—five dolomitic and 15 calcitic, these latter include two doubtful specimens compatible also with a Naxian provenance. The five dolomitic sculptures are two statues from the Extramural Sanctuary of Demeter and Persephone, one of the five muses now located in the house of Giasone Magno (the other four being Pentelic), a head of Athena and a statue of Aphrodite from the Cyrene museum.

The majority of the 15 calcitic assignments are the 8 samples taken from marble balustrade or fence elements in the two early Byzantine churches of Al Athrun, whereas the other architectural elements, mainly columns or capitals, are predominantly Proconnesian. This finding confirms the well-known opinion that in late antiquity both quarries were involved, with high degree of specialization, in the prefabrication of architectural components to be exported and completed at their final destination.

The other 7 Thasian samples include three additional architectural and funerary elements and four statues, one of which is an archaic kouros, indicating that the use of Thasian marbles at Cyrene, although limited, spans the entire history of the city.








L. Lazzarini, S. Cancelliere



Laboratorio di Analisi dei Materiali Antichi, Dip. di Storia dell’Architettura, I.U.A.V., S.Polo 2468, 30125 Venezia, Italy.







Marmor thessalicum and Lapis Atracius were names of geographic origin given by the Romans to a beautiful green breccia named verde antico during the Italian Renaissance. This stone was first introduced in Rome in Hadrian’s times for Colums, facing slabs, tubs, etc., soon becoming the most important green stone of Roman antiquity. Later in the Byzantine period it was also worked for sarcophagi, iconostasis, and baptismal fonts. The large use made by the Romans and Byzantines, and the Medieval re-use of spolia, produced an almost ubiquitous distributions in all the Mediterranean countries. Both ancient and modern quarries of verde antico are situated on the northern and southern slopes of Mount Mopsion, an area that in antiquity was under control of the small town of Atrax. Now the village closest to the quarries is Chasabali, in the Omorphocorion region, province of Larisa, the capital of Thessaly. The verde antico formation belongs to a large ophiolitic complex of Upper Jurassic age. Its various facies have been fully analysed minero-petrographycally and chemically, in order to distinguish them from the similar verde di Tino (green marble from the Marlas area on the island of Tinos, Cyclades), also used in Roman times. Marmor Thessalicum is composed of black-to-green antigorite and white-calcite clasts in a greenish matrix formed by a mixture of these two minerals. Accessory minerals are magnetite, chromite, tremolite and asbestos. It may be classified petrographically as an ophicarbonate breccia, a rock formed by serpentinisation of ultramafic rocks, and by mixing with limestone fragments and solutions, that have undergone an average-grade metamorphism in an ocean floor environment. The typical brecciated fabric and some trace elements allow clear differentiation of this very important stone from the verde di Tino. 








T. Cramer1, K. Germann1, W.–D. Heilmeyer2, V. Kästner2



1 Technische Universität Berlin, FG Lagerstättenforschung - Sekr. BH 4, Ernst-Reuter-Platz 1, 10587 Berlin, Germany,

2 Staatliche Museen zu Berlin, Stiftung Preussischer Kulturbesitz, Antikensammlung, Bodestrasse 1-3, 10178 Berlin, Germany







With the purpose of provenance determination, marble objects of the Berlin Collection of Classical Antiquities were examined in the Institute of Applied Geosciences of the Technical University Berlin: beside the Pergamon Altar mainly Archaic and Hellenistic objects originating from the Meander Valley and its surrounding areas (Milet, Priene, Didyma, Myus, Magnesia) in Asia Minor. After joint field surveys of archaeologists and geoscientists an extensive comparative database of Western Anatolian (Priene, Herakleia, ”Milet” and other quarries near the Bafa-Lake, Euromos, Stratonikea, Milas, Aphrodisias, Ephesos etc.) and Greek marble quarries is available. A reliable provenance determination of the usually quite homogeneous white marbles can be carried out only by using a combination of different properties. Features such as colour, banding, and patterns as well as grain size and orientation, grain boundaries, smell during grinding, and the mineral phases were determined qualitatively and partly also quantitatively. Microphotos of the whole thinsections allow a fast overview of the main petrographic features. Calcite, dolomite, and quartz were determined semi-quantitatively by means of XRD. The mainly carbonate-bound chemical components such as Mg, Sr, Fe, Mn, and the REE, with their distribution patterns determined by ICP-MS and ICP-OES, as well as the stable oxygen and carbon isotopes, proved to be useful for the provenance determination. Cathodomicrofacies and EPR spectra of some samples were kindly determined by Klaus Ramseyer, Danielle Decrouez, and Donato Attanasio.

On this basis, the marbles of a set of museum objects were assigned to their original quarries. Our previously expressed assumption that the marble of the Pergamon Altar was already quarried from 180 BC on the island of Marmara and brought to the Acropolis of Pergamon, could now be verified. Prokonnesian origin can also be assigned to marbles of the Demeter and Market Temples in Pergamon, as well as to the Eumenes Hall in Athens, the Harbour Thermes in Ephesos, and the Athena Temple in Troy. Isotopic data of the Pergamon Altar and from samples collected on Marmara Island, confirm the necessity to establish an extension of the isotopic field of this important marble provider and to introduce a second field with δ18OPDB values between –8 and – 12 (δ13CPDB + 2 - +3). In Hellenistic Pergamon also marbles from Lesbos, Thasos, Hymettos, and probably Ephesos were used for representative buildings. In Pergamon, where the main local building material is an andesitic rock, no indication was found for the use of marble from small lenses in the nearby Kozak massif.

In the Meander Valley region, a large number of local marble sources were available, widely used at many places. Nevertheless, also here some marble import took place. In Priene, e.g., with its considerably large marble resources, parts of the Athena Temple (4th c. BC) were built from the same marble used for the Heraion on Samos Island (6th c. BC), which was quarried on Phourni Island.






Matthias Bruno1, Carlo Gorgoni2, Paolo Pallante2



1 130, Via del Pellegrino, I-00186 Rome, Italy (

2 Department of Earth Sciences, University of Modena and Reggio Emilia 19, Via S. Eufemia, I-41100 Modena, Italy ( (







Following a research program on the identification of white marbles used in the most important monuments of ancient Rome, the results of the study of the Baths of Caracalla (the Thermae Antoninianae, built at the beginning of the III century A.D.) are presented. These results are important since the Severan Age, especially under Caracalla, was also a period of impressive building policy. In fact, the new dynasty entrusted the propaganda of their own image to imposing monumentality. The enormous Baths of Caracalla were decorated with a very large number of coloured stones, only in part remaining due to spoliation. The column shafts, in particular, were made of Pavonazzetto, Giallo Antico, and a variety of granites. White marble was also widely used, and is the main object of the present investigation. More than 30 samples were taken from friezes, architraves, cornices, capitals, column bases, etc., and submitted to careful archaeometric (i.e. mineropetrographic, chemical, and isotopic) analyses aimed at the definition of their provenance. The analytical results showed that Proconnesian and Pentelic marbles were mainly used, in comparable amounts, for the various architectural-decorative elements. In fact, in that period Proconnesian marble was largely dominant in public architecture, as testified for example in the Arch of Septimius Severus (in combination with Carrara marble) and in the Severan reconstruction phases of the Temple of Bellona and the Porticus of Octavia, here together with considerable reuse of Pentelic marble from the previous Domitianean phase. As for the latter marble, the present investigation confirms that it was used almost homogeneously in all Imperial Ages, initially only for friezes or entire monuments with special ideological-propagandistic significance (such as the Arch of Titus). In later Imperial times, it was more extensively employed, together with Proconnesian marble, also for general building elements such as column bases and capitals. According to the archaeometric results, in the Baths of Caracalla the architectural elements in Pentelic marble, with reference in particular to a series of figured Corinthian capitals, were possibly prepared using marble blocks expressly imported from Attica, thus confirming the importance of this great Imperial building program. More in general, these results were also useful for updating the diagram depicting the diffusion and use of different white marbles in temples and other public buildings of Imperial Rome.








M. Fischer



Tel Aviv University, Department of Classics, Ramat Aviv 69978, Israel,







The survey of marble occurring in Ancient Israel during the Roman period, which was concluded by the author in 1998 has revealed that this material was imported from various quarries mainly into three cities: the Mediterranean harbors Caesarea and Ascalon, and the inland city Beth Shean- Scythopolis. At that time the picture obtained has shown that the main bulk of import should be dated to the Roman rule in Palestine, the 2nd and 3rd centuries CE and that the main marble supplier was Proconnesus as detected both by marble analysis and historical sources. The source of marble is being determined by using a multimethod approach including stable isotope analysis, X-ray diffraction (XRD) and electron paramagnetic resonance (EPR). Some discoveries made during the past years enable us to shed a new light both on the origin and diffusion, and mainly the chronological distinction of the use of imported marble into Ancient Israel. Some of these discoveries refer to the Persian and the Hellenistic periods considered so far to have been outside the frame of marble circulation. Three relevant case studies are presented here. The first one was uncovered in the Mediterranean harbor Apollonia, north of Tel Aviv-Jaffa, where it could be related to the Persian layer (4th century BCE). It represents a Totenmahlrelief as known from Attic funerary art of the 4th century BCE. Both visual examination and laboratory analysis have pointed to the Pentelic origin of the marble. A second example represents the fragment of a statuette from the Hellenistic destruction layer (end of 2nd century BCE) of the Mediterranean anchorage place Yavneh-Yam, ca. 15 kms south of Tel Aviv-Jaffa. Its marble is of Parian origin. A third example is the protome of Minerva-Athena from Roman period Beth Shean-Scythopolis, identified as Thasian marble. The large amount of Proconnesian marble for architecture and statuary of Roman Palestine has already been emphasized in earlier studies.

Corroborating the data offered by the archaeological, artistic and laboratory analysis of the items presented here, an attempt is made at a chronological distinction for the use of marble of different origins. The Pentelic origin of an Attic artifact of the Persian period, which is equivalent to Greek Classic, is beyond any question. Parian marble in the Hellenistic Middle East is mentioned by contemporary sources and the same is done for Proconnesian marble in Roman times. Thasian marble is an archaeological fact for Roman Palestine statuary.








A. van den Hoek1, J. J. Herrmann, Jr.2, R. Newman3, E. Talamo4



1 Havard Divinity School, 45 Francis Ave., Cambridge, MA 02138;

2 Museum of Fine Arts, 465 Huntington Ave., Boston, MA 02115;

3 Museum of Fine Arts, 465 Huntington Ave., Boston, MA 02115;

4 Sovraintendenza ai Musei, Gallerie, Monumenti e Scavi del Comune di Roma, Piazza Campitelli, 7, 00186 Rome; fax 3906 575 4207







An opportunity to explore the range of uses for Thasian dolomite in Rome was presented by the new branch of the Capitoline Museums in the decommissioned Montemartini powerplant. All marble sculptures that appeared to be the coarse-grained, pure white marble of Cape Vathy were sampled, and the samples were analyzed with x-ray diffraction in the Museum of Fine Arts, Boston to determine whether they were dolomite or calcite.  The possibility that the similar dolomite from Coin, near Malaga, Spain was discounted because of the absence of evidence for Spanish marble exportation.

Sixteen dolomite sculptures could be identified, only a small percentage of the marble sculptures in the Montemartini Museum. All were figure sculptures, and all dated from the Roman Imperial period.  In some cases, stylistic similarities between dolomite sculptures suggest workshop connections. An Augustan military trophy could be connected with a dolomite cuirass carved in middle-Hellenistic Pergamon. A colossal Athena of the Velletri type is only one of several gigantic dolomite replicas of this figure; others are in Yale, and the Louvre (untested). Technical preferences are observed; two graceful statues of Muses make use of dolomite for the draped bodies and another marble (probably Parian) for the heads.

Other identifications of dolomitic sculptures presumably from Thasos have been made since the last ASMOSIA conference in 2000.  Of special interest is another Somnus/Hypnos (“Eros sleeping on his torch”) of the type prefabricated on Thasos. 








F. Gabellone1, M.T. Giannotta1, A. Monte1, A. Alessio2



1 CNR – IBAM –Istituto Beni Archeologici e Monumentali – Lecce, c/o Villa Tresca - Campus Universitario - Via Monteroni – 73100 Lecce, Italy,,

2 Soprintendenza Archeologica ai Beni Archeologici della Puglia – Taranto, via Duomo, 33 – 74100 Taranto, Italy,,







The ancient sea routes are marked by the wrecks of ships, which sank along the coasts of the Mediterranean. Among the wrecks transporting marbles (navis lapidariae) from the Roman era are those of Torre Chianca, San Pietro in Bevagna and Torre Sgarrata, discovered off the Ionian coast of the Salento peninsula in southern Italy. These are the object of multidisciplinary research conducted by IBAM and the Soprintendenza with the aim of identifying the origin of the transported marbles, which is essential to the study of the marble trade in antiquity.

Only the Torre Sgarrata wreck has been recovered - by Throckmorton between 1965 and 1967. The presence of a coin bearing the head of the emperor Commodus (180-192 AD) among the materials found has made it possible to date the wreck to the end of the II – beginning of the III century AD. The cargo consisted of 18 sarcophagi and 17 blocks, all of white marble, as well as six alabaster slabs. The marbles were attributed by Ward-Perkins (1965) to the quarries of Aphrodisia and by Parker (1992) to the quarries of Thasos.

The salvaged materials were taken to Taranto where currently 15 sarcophagi and 14 blocks, all in white marble, remain. Each of these has been catalogued with a view to classifying them in metrological and typological terms, and sorting them into homogeneous groups.

The metrological study shows that the Roman foot is the most likely unit of measurement.

The typological analysis makes it possible to recognise two types of sarcophagus:

1-     rectangular sarcophagi with rectangular cavity (six items)

2-     rectangular sarcophagi with apse cavity (nine items).

The blocks, of rectangular shape, are semi-worked products; some of them were intended for use as architectural elements or as lids for the sarcophagi.

The archaeometric investigation, presented in a separate paper in this conference, has made it possible to recognise two types of marble, one dolomitic and the other calcitic, quarried from two different quarries on the island of Thasos, located respectively at Cape Vathy and Aliki. They correspond exactly to the two types of sarcophagus cited above, while only three of the blocks are of dolomitic marble.

An analysis of how the individual items of both types and marbles were stowed aboard the ship supports the hypothesis that they were all loaded at the same thasian port.









F. Van Keuren1, L. P. Gromet2



1 Lamar Dodd School of Art, University of Georgia, Athens, GA  30601-4102, U.S.A.,

2 Department of Geological Sciences, Box 1846, Brown University, Providence, RI  02912, U.S.A.,







Analyses were conducted of the marbles of three sarcophagi at the Museum of Art, the Rhode Island School of Design.  Two of the sarcophagi are reported to come from Rome, and the fragment of a third sarcophagus has no known provenance.  Isotopic analyses of the lid and chest of the first sarcophagus, the front portions of a western sarcophagus with three-sided decoration, indicate they are most likely made of Carrara marble (lid: δ13C = 2.24‰ and δ18O = –2.05‰; chest: δ13C = 2.33‰ and δ18O = –1.84‰).  The statistically close isotopic ratios and the similar appearance of the lid and chest support the hypothesis that they were carved from the same block—a finding consistent with the unified iconography of the two parts. On the sarcophagus lid, Apollo and Artemis shoot down at the children of Niobe, who are depicted on the chest.   

Isotopic analyses of the lid and chest of the second sarcophagus, of eastern design with continuous decoration on all four sides, reveal that they are most likely made of marble from Dokimeion (lid: δ13C = 1.89‰ and δ18O = –5.04‰; and chest: δ13C = 1.46‰ and δ18O = –5.16‰).  The statistically distinct carbon isotope ratios and a different coloration of the lid and chest indicate that they were probably carved from separate blocks.  The Asiatic roof shape of the lid matches its Asiatic marble.  But the theme of Achilles dragging Hektor's body on the front of the chest is characteristic of Attic sarcophagi, and suggests the manufacture of this part of the sarcophagus in Rome, where the sarcophagus was found and where Attic models were available. 

The isotopic analysis of the third sarcophagus fragment was inconclusive (δ13C = 3.54‰ and δ18O = –4.86‰).  However, the marble was determined to be dolomitic and hence Thasian from Cape Vathy.  Its high magnesium content was revealed by a lack of reaction with dilute hydrochloric acid, and confirmed through electron microprobe analysis. This fragment is one of four sarcophagi that depict the Lydian Queen Omphale, who is decked out in the lion's skin or weaponry she has confiscated from her slave Herakles. The other three examples are made of Prokonnesian and Dokimeian marble.  Altogether, the marble provenances of the four sarcophagi demonstrate the myth’s currency in the Aegean world where Omphale’s kingdom was located.








Y. Maniatis1, P. Sotirakopoulou2, K. Polikreti1, E. Dotsika1, I. Tzavidopoulos1



1 Laboratory of Archaeometry, Institute of Materials Science, NCSR “Demokritos”, 153 10 Aghia Paraskevi, Attiki, Greece, e-mail:

2 Museum of Cycladic Art, 4 Neofytou Douka Str., 106 74 Athens, Greece, e-mail:







The Keros hoard comprises of a large series of Bronze Age idol fragments, which are assumed to be coming from the prehistoric settlement in the Cycladic island of Keros, situated SE of Naxos. Eighty of these fragments are housed in the Museum of Cycladic Art and the rest elsewhere. The fragments are now under study in the context of their publication. However, the study of the “Keros Hoard” material presents many difficulties, due to the fact that it is a product of illicit excavations and that the site from which it has reputedly come seems to have suffered a geological landslide. Therefore, the scientific determination of the provenance of the marbles is expected to be of great value for both the interpretation of the find and the certification of the authenticity of at least some of the objects comprising it.

In order to approach the provenance of the marble in the most secure way, a series of scientific techniques were used. At first, all the objects at the Museum of Cycladic Art were examined macroscopically with the help of a light torch, a magnifying glass, and a millimetre scale. Based on this first examination, twelve different types of marble were identified initially, on the basis of colour, grain size and translucence. The overwhelming majority of the marbles ranges in colour from white to whitish and buff, quite often with tints of yellow and grey, but grey marbles are also present. These are generally fine-grained, with a grain size ranging from 1 to 1,5 mm, but those with a very fine grain, of about 0,5 mm, and the moderate- or coarse-grained, with a grain size of about 2 or even 3 mm, are not missing. Most of them are of moderate translucence, but there are also marbles with little or exceptional translucence, as well as marble fragments which due to the great erosion of the surface appear completely opaque.

Following the initial in-situ examination at the Museum, 22 pieces, representative of the different types of marble observed, were sampled and analysed with EPR Spectroscopy and Stable Isotopes. As the amount of sample obtained was very small we followed a stepwise procedure, the EPR analysis, which does not destroy the sample, was performed first and then the Stable Isotope analysis, by diluting the same sample.

The analyses results are now treated statistically and interpreted but the first conclusions can be drawn. All samples seem to come from Naxos. The white or whitish marbles, with a moderate grain size (about 1.5 mm) and relatively small translucence come from northern Naxos (Apollonas). The very fine-grained whitish marbles, with tints of tints of grey or yellow and rather great translucence may come from south or SE Naxos or Keros, with at least one sample coming definitely from Keros. Of particular interest is a rather large group of white and generally fine-grained marbles (grain size about 1 mm) of small to moderate translucence, which may represent a single source or workshop located at some so far unknown site in northern Naxos, most probably near Apollonas but not at the well known quarry area. The area near the village of Apiranthos is also a candidate region for the provenance of these idols.








Y. Maniatis1, S. Papadopoulos2, E. Dotsika1, D. Kavoussanaki1, I. Tzavidopoulos1



1 Laboratory of Archaeometry, Institute of Materials Science, NCSR “Demokritos”, 153 10 Aghia Paraskevi, Attiki, Greece, e-mail:

2 18th Ephoreia of Prehistoric and Classical Antiquities, 17 Erythrou Stavrou Str., 651 10 Kavala, Greece, e-mail:







Fragments of 11 marble vessels were found during the rescue excavation at the archaeological site of Limenaria, in Thassos. Most of these artifacts are dated to the end of Middle Neolithic Period and two to the Early Bronze Age. They are all fragments imitating utilitarian pottery vases like plates and bowls. The objects are generally thin-walled and nicely polished and the marble is in many cases translucent. They were found inside storage pits, which may indicate an attempt by the neolithic people to protect their valuable objects.  

It is the first time that analysis for the identification of provenance of neolithic marble objects is attempted, and results are actually very important for the movement of people and exchange networks in those early times. For the characterization of marble and identification of provenance a series of scientific techniques were used such as: optical microscopy, stable isotope analysis and EPR spectroscopy. 

The examination under the microscope showed that the majority of objects are made of an extremely fine grain marble with maximum grain size 0.4-0.8 mm. A couple of objects are made in a coarse marble with maximum grain size 2.5-3.0 mm. The colour of marble is whitish and having some argillaceous vanes in some cases.

The up to now results from the stable isotope analysis indicate that the marble is not of a thassian origin)! The coarser ones come from North Naxos and all the fine grain ones come from South East Naxos or Keros. This result is very interesting and peculiar, as Thassos is full of marble and importing marble from far away must have a serious reason, which needs to be discussed and explained.








M. Unterwurzacher1, H. Stadler2, C. Franzen1, P. Mirwald1



1 Institute of Mineralogy and Petrography; University of Innsbruck, Innrain 52, 6020 Innsbruck, Austria; E-Mail:

2 Institute of Pre and Early History, Mediaeval and Modern Archaeology; University of Innsbruck, Innrain 52, 6020 Innsbruck, Austria; E-Mail:







The existence of the Roman settlement at Schroettelhofer field near Oberdrauburg, Carinthia, is known since the 19th Century. Due to its geographical location at the intersection of two thoroughfares between Italy and Noricum and Pannonia the site was of particular regional significance.  Since 1995, this settlement area has been the subject of archaeological studies in the frame of an international cooperation between the universities of Innsbruck, Munich and Laibach.

Building structures of wood and stone were excavated. These are indicative for a Roman villa, dated from the 2nd to the 4th Century AD and are unique compared to other Roman stations in that region. The inventory of stoneware and array of furnishings, consisting of precious metal objects and a variety of marble objects such as mouldings, dolphin heads and bowls underline the importance of this site.

The marble artefacts were object of a specific study. Marbles were widely used in Roman times. However, the provenance of these objects, particularly in the Roman provinces, is often unclear. Information on that is of considerable importance e.g. for reconstruction of historical contextes. Important alpine marbles come from Laas / South Tyrol and from Gummern / Carinthia. Often and in particular in the Middle Ages they were used instead of Carrara marble. The basic mineralogical – petrographical properties of both marbles have been subject of a number of studies (e.g. CORTECCI, G.; DINELLI, E.; D'AMICO, C.; TURI, B. (2000)* and MÜLLER und SCHWAIGHOFER (1999)**).

Under this aspect the marble objects excavated in Oberdrauburg were studied. Grain size evaluation, cathodoluminiscence properties and chemical analysis provide a good characterisation of the materials. Particularly useful proved the determination of C and O isotopes and Mn content. A Carrara provenance of the objects can be excluded on the grounds of the data obtained. A regional comparison among south alpine marbles (South Tyrol - Western Carinthia) speaks in favour of a provenance from the Gummern Marble complex in Carinthia.








R. H. Tykot, M. Brandaglia, M. D. Glascock, R. J. Speakman



Department of Anthropology, University of South Florida, 4202 E. Fowler Ave., SOC 107,Tampa, FL 33620 USA.







The sourcing of obsidian artifacts, first applied in the Mediterranean four decades ago, is now resulting in a detailed understanding of the sequence of behaviors responsible for the acquisition, transport, manufacture, and use of obsidian in this region. At one point limited by the small number of artifacts analyzed from a single site and for a single cultural period, the development in recent years of non- or minimally destructive, and less expensive analytical techniques has permitted the analysis of significant numbers of artifacts so that statistically meaningful patterns of obsidian source utilization may be determined. In addition, while earlier provenance studies in Italy usually attributed obsidian artifacts to one of four island sources - Lipari, Palmarola, Pantelleria or Sardinia - it is now possible to determine discrete subsources on each island.

In this study, part of a large-scale project supported by the National Science Foundation, the precise geological provenance of a large assemblage of obsidian artifacts recovered from an Early Neolithic site on Isola del Giglio in the Tuscan archipelago was determined. The results provide new insights into the socioeconomic role of obsidian trade during the chronological phase in which agriculture, village settlements, and ceramics first appear in the central Mediterranean region. Following the high-precision measurement of density and a visual estimation of source, both techniques that have been shown to be useful in souce identification, the Isola del Giglio artifacts were analyzed by laser ablation ICP mass spectrometry. LA-ICP-MS was selected for elemental analysis of the obsidian artifacts since it provides quantitative results for a large number of trace elements in a virtually non-destructive manner.

Because Isola del Giglio is strategically located along north-south trade routes on the west coast of Italy, the results obtained provide an important comparison with data available from other Early Neolithic sites in Tuscan, Liguria, Corsica and Sardinia. The similarities and differences in source utilization among these areas are interpreted in terms of geographic location, transport methods, and exchange mechanisms, as well as the actual function served by obsidian artifacts in particular lithic assemblages. The geological, archaeological, and analytical approaches employed in this study may profitably be applied to obsidian studies in other parts of the world.








S. Meloni1, C. Lugliè2, M. Oddone1, L. Giordani3



1 Istituto C.N.R. per l’Energetica e le Interfasi – Sezione di Pavia -, Dipartimento di Chimica Generale, Università di Pavia, Viale Taramelli12, Pavia (Italy),,

2 Dipartimento di Scienze Archeologiche e Storico-Artistiche, Università di Cagliari, Cittadella dei Musei, Cagliari (Italy),

3 Dipartimento di Chimica Generale, Università di Pavia, Viale Taramelli 12, Pavia (Italy)







Obsidian is a volcanic glass, which can be handcrafted and reduced to thin and sharp sheets and used as cutter or arrow’s head. These tools have been utilized since prehistoric times and their diffusion spread over large areas starting from the sites where obsidian flows are located.

The identification of provenance of the prime matter of obsidian artifacts is considered significant to track the pathways of trading in those times.

One of the sites where obsidian flows are located in the Mediterranean basin is at the volcanic complex of Monte Arci in Sardinia (Italy). Artifacts made with Monte Arci obsidian were found in neolithic settlements in northern Italy and south France.  

Recent investigations have shown that at least three obsidian flows can be identified in the Monte Arci area. It was considered worthwhile to characterize these sources by their minor and trace element content and eventually to find discriminating parameters useful to provenance the prime matter of obsidian artifacts and to identify the obsidian flow exploited.

In the present work obsidian samples from the three Monte Arci flows (SA, SB and SC) were submitted to instrumental neutron activation analysis for the determination of a number of minor and trace elements. In addition some obsidian artifacts from two archaeological sites in Sardinia, dated back to early or middle neolithic period, namely Su Paris de sa Turre (Santa Caterina di Pittinuri, OR) and Coddu is Abionis (Terralba, OR), were also investigated.

Element contents are presented and discussed for their precision and accuracy.

Several approaches were adopted to characterize and discriminate the 3 Monte Arci obsidian flows: rare-earth element patterns, the quantitative evaluation of the Eu negative anomaly, linear discriminant analysis and cluster analysis. The 3 sources can be fully discriminated and applying the same procedures including the obsidian artifacts data, it is possible to advance a provenance assignment showing the flow from where obsidian had been quarried for artifacts manufacturing.






ON stones used as tesseræ of roman mosaics (Lombardy - Italy)



R. Bugini1, L. Folli2



1 Istituto CNR Conservazione Beni Culturali Sez. “Gino Bozza” – p. Leonardo da Vinci 32, 20133 Milano, Italy

2 Viale Rimembranze 37, 20075 Lodi, Italy







A lot of stone tesseræ coming from roman mosaics were analysed in order to identify the lithotypes and the quarry sites. About 50 samples were collected from different archæological sites in Lombardy.

The selected sites are: two “domus” in Milan (via Amedei and Chiesa Rossa); remains in the church of Garlate (Lecco); one “domus” in Calvatone (Cremona); three buildings in Brescia (Palazzo Martinengo, Collegio Arici and Ortaglia); two villas near the lake Garda (Desenzano and Sirmione). These sites are mainly referred to 1st century AD.

The samples were investigated by optical microscopy on thin section and X-ray diffraction on powder, compared with selected samples coming from ancient quarry sites in Lombardy and Venetia.

Most of stone tesseræ are white or black with other colours (pink, red, yellow) recurring in few cases only.

The white tesseræ are made by oolitic limestone belonging to San Vigilio formation (Lower Jurassic) and quarried near the eastern shore of lake Garda, by micritic limestone belonging to Biancone formation (Lower Cretaceous) and quarried near Verona (monti Lessini). Mosaic tesseræ made by white marbles with different texture and grain size are found in few cases only. Noteworthy it is a pavement (opus signinum) from a “domus” (via Amedei, Milan) containing elongated white pieces of aragonite in a mortar made by lime and crushed brick.

The black tesseræ are made by black limestone belonging to Upper Triassic formations, but the the identification of provenance is uncertain; famous quarry sites of black limestone are located near Varenna (lake Como), val Seriana (Bergamo), Riva di Solto (lake Iseo) and val Degagna (lake Garda).

The red tesseræ are made by nodular limestone belonging to Rosso Ammonitico formation (Upper/middle Jurassic) and quarried near Verona (Monti Lessini). Pink and yellow limestone used to make coloured tesseræ have the same geographic provenance.

In conclusion San Vigilio and Biancone white limestone and Triassic black limestones were used as tesseræ in mosaics of different archæological sites far from the quarry location too: these stones therefore were specifically devoted to mosaic making.








P. Degryse 1,Ph. Muchez 1, E. Trogh 1, M. Waelkens 2



1 Fysico-chemische Geologie, K.U.Leuven, Celestijnenlaan 200C, B-3001 Heverlee, Belgium,

2 Dept. Archeologie, K.U.Leuven, M. Theresiastraat 21, B-3000 Leuven, Belgium.







Different types of building stones have been macroscopically, petrographically and geochemically characterised at the ancient city of Sagalassos (SW Turkey). The provenance of most of the building stones may be related to local lithological units, though some stone types were clearly imported from considerable distance. The local natural building stones include limestones, conglomerates, breccias and sand- to siltstone of different qualities, originating from the Lycean nappes and flysch deposits, both in the immediate vicinity of the city as well as further away on its territory. Travertine and volcanic building stones were brought to the city from a distance of a few kilometres, respectively from the area of Başköy and that of Gölçük. Marbles were transported in small quantities from as far as the Docimian quarries, 250 km to the north of Sagalassos. The provenance of some building stones, including brown-grey limestone and greyish granite, could not yet be identified. Throughout the building history of the city, local beige and pink good quality limestone remained the most important building stone. High quality white limestone from the territory of the city represents a small but important fraction of the total amount of building stones used.

Only few buildings at Sagalassos can be attributed to the Hellenistic period. Petrographic and stable isotope analysis of the building blocks from these structures and from limestones exposed in outcrops and quarries on the territory have shown that the limestone originated from quarries in the immediate vicinity of the monumental centre of Sagalassos. For the imperial period, three periods of intense building activity can be identified: the Julio-Claudian period (Augustus to Claudius: 27 BC to 54 AD), the Flavian (?)-Trajanic period (54 AD to 117 AD) and the period from Hadrian to the Severi (117 AD to 235 AD). The use of white limestone forms a clear trend from the Trajanic period (98-117 AD) onwards. The petrographic and stable isotope data indicate that the limestone used, originated both from nearby quarries as well as from quarries much further away on the territory of Sagalassos. The selection of building stone went hand in hand with the appreciation for structural strength and suitability for carving complex architectural decoration, together with the desire to obtain a polychrome architecture.









Ø. J. Jansen1, T. Heldal2, R. B. Pedersen3, S. H. H. Kaland4



1 University of Bergen, Bergen Museum, Natural History Collections, N-5020 Bergen,

2 Geological Survey of Norway, N-7491 Trondheim,

3 University of Bergen, Dept. of Earth Science, N-5007 Bergen,

4 University of Bergen, Bergen Museum, Cultural History Collections, N-5020 Bergen,







The earliest stone buildings in the city of Bergen, mainly churches and monasteries, were erected in the 12th and 13th centuries. A number of rural stone churches were also built in the surrounding area. Ashlars and finer stoneworks in these buildings were made almost exclusively of soapstone. Until recently, however, little has been known about the medieval soapstone quarrying in the region. New research begins to shed light on this part of Norway's early industrial history by locating the quarries and linking them to the rock material used for particular buildings.

Soapstone was used in Norway since long before the arrival of Christianity. Since the first millennium BC, the use of soapstone, mainly for cooking vessels, was periodically common in most parts of the country, partly due to the fact that soapstone has a high heat-storage capacity. The production of such vessels reached a peak in the Viking Period (800-1050 AD) and remained common also in the following centuries. Soft greenschist was another important material, used for baking slabs.

Some of these quarries, especially those located close to the sea, were turned into building-stone quarries in the 12th and 13th centuries. The derivation of building stone can be distinguished from the extraction of vessels by studying the quarrying methods.

The mineralogy and chemistry of soapstone can vary significantly on both local and regional scale, whereby it is difficult to identify the provenance of particular material on the basis of standard mineralogical and geochemical analyses. The present paper will discuss criteria for soapstone characterization, and will suggest a provenance identification scheme based on a range of criteria, including whole rock geochemistry, isotope geochemistry, mineral composition, textural characteristics, quarry characteristics, and data from historical sources. These criteria are presently used for the mapping and interpretation of the Medieval building-stone quarrying industry in the surroundings of Bergen.








K. Laskaridis 1, V. Perdikatsis2



1 IGME – Lithos Laboratory 1st km Markopoulou Av. 190 02 Peanea Attiki, Tel: 210 – 6642821, FAX number: 210 – 6642822,

2 Technical University of Crete, Department of Mineral Resources Engineering, 73100 CHANIA / Greece,Tel. 08210 37418,







References to the “white whole-grain” marble of the Thassos Island date back to the 1st century BC, when stones were quarried at the Alyki peninsula (at SE of the island), the cape Fanari (at the NE) and the cape Vathy. Even from that time, marble from Thassos was exported to Samothraki and other neighboring islands, the coast of Minor Asia, the South of Greece and Rome.

In antiquity, the quarries of Thassos belonged to two major categories:

a)            Those that were operated with the purpose of extracting material for the construction of temples and the creation of art pieces (statues, monuments etc).

b)            Those that were operated in order to extract blocks for export purposes.

The aim of the present work is to make apparent the reasons for which the Thassos marble has exhibited a time-enduring behavior, on the basis of the evaluation of the aesthetic and physico-mechanical properties of the stones. For this purpose the relative properties have been measured and evaluated for two different types of white marble produced on the island.

The stone properties can be classified in those that characterize the material (such as the mineralogical composition, the particle size distribution, the apparent specific density, the coefficient of open porosity, the absorption coefficient etc) and those that are used to document the suitability of the material for specific uses (e.g. the effect of cyclic temperature variations, the modulus of elasticity, the compressive strength, the flexural strength, the impact, the abrasion resistance, etc).

The relative properties have been measured according to the specific DIN for natural stones, at “LITHOS” the accredited laboratory of IGME.

The aesthetic and the physico-mechanical characteristics - as well as their mineral composition which determines their ability to resist “time” and mechanical stresses - render the various types of the white marble of Thassos an ornamental stone with a wide application spectrum and confirm their ability to endure time.  









B. Tsikouras1, K. Mihopoulos2, K. Hatzipanagiotou3, N. Ninis4



1 University of Patras, Department of Geology, Section of Earth Materials, GR-265 00 Patra, Greece,

2 University of Patras, Department of Geology, Section of Earth Materials, GR-265 00 Patra, Greece.

3 University of Patras, Department of Geology, Section of Earth Materials, GR-265 00 Patra, Greece,

4 Committee for the Preservation of the Epidavros Monuments, GR-210 52 Ligourio, Greece.







The success of a restoration project depends significantly on finding and choosing appropriate stones, aesthetically and mechanically compatible with the authentic ones. Here are investigated three “porolithoi” that have been used extensively in the Epidaurean Asklepieion – one in the foundation of buildings and (“foundation porolithos”) and two in their upperstructure (“Aegina” and “Kenchrae” stones) – together with six new stones proposed and used as were suggested. The names of the new stones relate to their provenance and quarrying localities: Alfopetra, Vryses, Aggistri, Cephalonia, Cyprus and Zakynthos. The first two were proposed as a substitute for the “foundation porolithos”, the next two for the “Aegina” stone and the last two for the “Kenchreae” stone. The latter is a porous oosparitic limestone. All the other investigated rocks are biomicritic and biosparitic limestones. SEM analysis defines the petrographic characteristics of the stones including porosity. Qualitative analysis with EPMA revealed that all stones are composed of pure calcite. Calcite crystals show variable degree of dissolution in all examined samples. Physicomechanical properties of the rocks coupled with the petrographic and SEM observations suggest that the “Aggistri” and “Cephalonia” stones are the most compatible to the “Aegina” stone while the “Vryses”, “Cyprus” and “Zakynthos” are compatible to the “Kenchreae porolithos”.











I. Vardoulakis1, I.-O. Georgopoulos1, A. Pachigianni2



1 Department of Mechanics, Faculty of Applied Mathematics and Physics, National Technical University of Athens.,

2 Committee for the Preservation of the Epidaurus Monuments (ΕΣΜΕ), Archaeological Museum of Epidaurus,







In this paper results are presented from mechanical and physical properties’ tests, performed on the stone material of the ancient foundation members of the Tholos of Epidaurus, as well as the artificial stone mortars applied on the above members, during the restoration works at the monument. These experiments aim at a rational evaluation of the poro-mechanical compatibility of natural stones and mortars. The experimental results from the uniaxial compression tests were finally used for the calibration of appropriate elasto-plastic models for a compacting, cohesive-frictional and dilatant material. At the same time, tests measuring permeability, absorption and porosity were conducted. Within the limits of this study, the above physical properties were set as the representative indicators of the durability characteristics of the examined stones and mortars. In an attempt to utilize the theory in the direction of improving our knowledge concerning the restoration mortars’ behavior, the outcome of the physical properties’ tests was juxtaposed to the empirical knowledge, concerning similar to the artificial stone mortars materials, such as conventional concrete.









I. Papayianni, M. Stefanidou



Department of Civil Engineering, Aristotele University of Thessaloniki, 54124 Thessaloniki, Greece,







Apart from marble and limestone many other stones from local deposits such as gneiss, serpentinite, and schist have been widely used for the construction of monuments mainly because of their abundance in the erea. In the case of the remains of the city walls in ancient Dioklitianoupoli, in Northern Greece, serpentinites are the stones of the masonry, which suffer from severe degradation due to the deterioration problems of serpentinite under the local climatic conditions.

An adequate number of samples were taken and studied systematically in order to make diagnosis of the symptoms and interrelate the characteristics of the sernentinite with its respond to the environmental and functional conditions.

The characteristics studied are:

-               Minerological composition (by polarized microscope and XRD)

-               Mechanical strength and modulus of elasticity

-               Porosity and pore size distribution

-               Soundness

-               Water elevation by suction

In parallel, new cut samples taken from the near deposits were tested before and after treatment with consolidants in order to decide upon the replacement of serpentinite where it is necessary.









E. Leka



Universite Paris I (Pantheon-Sorbonne) and Thrakis 14, 171 21 Nea Smyrni Athens Greece







This paper examines the problems involved in the study of sculptures restored in antiquity as well as some possible ways of resolving them. Comparative discussion of examples of archaic Greek sculpture from mainland Greece, the Aegean islands and Eastern Greece restored at different periods throughout antiquity will be employed to explore the main aspects of the subject. At first, the study will focus on ancient Greek preservation methods and restoration techniques and the materials used for this purpose. Subsequently observations on how these techniques and materials changed from one period to the other or from one region to the other will be made. Finally some light will be cast on a. the relevant written sources, b. the general character of the sculpture restoration in antiquity, c. the reasons and d. specialized craftsmen responsible for it.









I.-O. Georgopoulos1,  I. Vardoulakis1, J. Labuz2



1Department of Mechanics, Faculty of Applied Mathematics and Physics, National Technical University of Athens,

2Department of Civil Engineering, University of Minnesota, USA,







In this paper results of triaxial compression tests performed for various confining pressures on Dionysos Marble specimens are presented. In order to study the effect of specimen shape two types of cylindrical specimens have been tested with height to diameter ratios of 1:1 and 2:1, respectively. The mechanical properties of this marble in compression are evaluated and their dependency on confinement is critically examined. The mechanical behavior of the considered marble is being modeled within the frame of Elastoplasticity theory. The experimental results support a cohesion softening, friction hardening constitutive model for a quasi-brittle, dilatant solid with definite reverse plasticity effect.











Michael Greenhalgh



The Sir William Dobell Professor of Art History, School of Humanities, The Australian National University,Acton, Canberra 0200 AUSTRALIA,







 The paper, drawn partly from research in the French military archives, examines accounts of fortifications in the Eastern Mediterranean and suggests two reasons for the otherwise puzzling dearth of columns in seaboard, ship-accessible antique cities:

1.            the widespread use of antiquities (especially column shafts) in the construction and decoration of pre-gunpowder castles - one explanation for "disappearing  columns";

2.            the long-lived vogue for stone and marble projectiles slung from machines and then from gunpowder-driven  cannon (especially in the Ottoman Empire, and for both shore fortifications and on ships), which was responsible for the continuing and large-scale destruction of antiquities, especially columns, well  into the 19th century. Projectiles were needed in very large quantities, and column-shafts provided part-made sources, often in convenient sizes. (By contrast, the burning of large quantities of marble for lime in this area is a late-19th-century practice.) Such a "practical" re-use of the past, for which plenty of the evidence survives throughout the region in fortresses and moats, overlaps with the coming of the archaeologists and the sanctification of the classical past, for it probably continued at least into the 1860s.  






Provenance Studies of Lapis Lazuli Nondestructive Prompt Gamma Acivation Analysis (PGAA)



J. Zöldföldi1, 2, Zs. Kasztovszky3



1 Landesdenkmalamt Baden-Württemberg, Berliner Straße 12, D-73728 Esslingen a.N., Germany

2 Department of Geochemistry, Institute of Geoscience, University of Tübingen, Wilhelmsrt. 56, D-72074 Tübingen, Germany, E-mail:

3Institute of Isotope and Surface Chemistry, Chemical Research Centre of the Hungarian Academy of Sciences; H-1525 Budapest, PO. Box 77, Hungary, E-mail: 







Archaeological objects made of lapis lazuli are widely distributed in the ancient East and some date back as early as the second half of the fourth millennium B.C. in Central Asia. The mineralogy of lapis lazuli has been given considerable attention in the past. Moreover, the new non-destructive analytical techniques of historical artefacts have gained more attention in archaeometry research. Knowledge of the elemental composition, including major and trace elements may provide clues concerning the provenance and raw materials. PGAA is one of the new candidates to answer these questions.

PGAA is based on the identification of gamma photons originated in (n,g) nuclear reaction. The chemical elements are identified in according to the energies of their characteristic gamma-ray peaks, while the quantitative analysis is based on the exact determination of the peak intensities. The detected gamma-ray intensity is directly proportional to the mass of a given element, the analytical sensitivity for that element and the measurement time.

Lapis lazuli is an opaque semi-precious stone consisting mainly of a blue mineral, an alumo-silicate of the complex feldspathoid sodalite group. Lazurite can be described as „an isomorphic combination of haüynite and sodalite“. The chemical formula for haüyne can be written as: (NaCa)4-8(Al6Si6O24)(SO4,S)1-2 and for sodalite as: Na8(Al6Si6-24)Cl2, for lazurite essentially, as (NaCa)4(AlSiO4)3(SO4SCl), but with considerable variation in the amounts of SO4, S and Cl.

In this project we succeeded to collect lapis lazuli samples from the most relevant quarries in the world. Rock samples from Afghanistan, from Lake Baikal, from Chile and from Ural Mountains have been investigated. With PGAA we were able to detect the major components, H, Na, Ca, Al, Si, S, Cl, K, and the accessory elements Mg, Fe, Mn. In addition, the trace elements of B, Sc, Cr, Co, Sm and Gd were also idetified.

According to some characteristic element ratios the samples from Afghanistan and Baikal are more or less overlapping, while the samples from Ural and Chile are definitely different from the others.






Raman Microspectrometry and PIXE Investigation

of Maya Green Stones from Calakmul, Mexico



T.-H. Chen, R. Garcia-Moreno, M. Menu



Centre de Recherche et de Restauration des Musées de France, CNRS-UMR 171, 6 RUE DES PYRAMIDES, 75041 Paris CEDEX01, France,







The Maya culture, developed in the southeast of Mexico, Guatemala, Belize, and the extremes of Honduras and El Salvador, has always related green colour to the most sacred values. Therefore, green stone artefacts are of most importance in ceremonial archaeological contexts, principally in royal tombs and construction offerings. Its approach has generally concerned formal and symbolic aspects, leaving the identification of quarries and dating rather unknown.  The lack of mineralogical research on these antique objects in the past has favoured the generalisation of the term Maya jade to design them.  By having an approach to their chemical characterisation, we might have a better idea of the variety and selection criteria of green stone exploitation by ancient Mayas.

The present study involves several samples of green stone, presenting different aspects, from controlled excavations in Calakmul (Mexico), one of the most important ancient Maya capitals (400 B.C.-1200 A.D.) The samples have been structurally and elementarily analysed in the C2RMF by means of non-destructive techniques: Raman microspectrometry and external beam PIXE (Particle Induced X-ray Emission). The aims of this work are to characterise the green stones, to correlate their chemical composition to Raman spectra, and to explore their possible provenance. Most of the samples are jadeitite, in which the dominant mineral is jadeite (NaAlSi2O6), with presence of albite, omphacite, mica, diopside as well as other minor minerals. According to the Raman spectra, the strongest Si-O-Si vibrational band of the jadeites is in the wavenumber range between 660 and 710 cm-1. With the variety of impurities, i.e. different concentrations of Fe, Mg or Ca in the samples, the wavenumber shifts and variation of half-width of these vibrational bands were observed. Principally, the band shifts towards a lower wavenumber and the half-width increases with augmentation of other mineral contents. Besides, comparison of major, minor and trace element concentrations obtained by PIXE with those in literature might represent an advance in characterising their most probable provenance: the Motagua Valley in Guatemala.

PIXE and Raman microspectrometry, two complementary, non-destructive techniques allow characterisation of the greenstone objects without sampling. Elemental analysis by PIXE might reveal the provenance of the jades. Finally, the correlation between chemical composition and the Raman spectra of jadeitites could permit estimation of jadeite content by simply using Raman microspectrometry which is more available, and actually represents an increasingly employed technique in archaeometry field under non-destructive consideration. 






An attempt for Greek marble discrimination based on trace- and isotope- analyses combined with mineralogical and petrographical analysis.



K. Kritsotakis1, V. Perdikatsis2, K. Laskaridis3



[1] Institut fuer Geowissenschasften, Saar str. 21, Johanes Gutenberg Universitaet, 55099 Mainz, Deutschland

2 Technical University of Crete, Mineral Resources Department, 73100 Chania, Greece,

3 IGME - Lithos Laboratory, 1st km Markopoulou Aven, 190 02 Peanea, Attiki







The archaeological and commercial interest for the characterization of origin of building stones and particularly marbles from concrete marble quarries is big.

With this aim several studies have been carried out in the past, that referred to the petrographical characterization and mainly in trace element analysis.

In the present work a farther attempt is made for the discrimination of the origin of marbles, based on their petrographical – mineralogical characterization and analysis of major and trace elements as well as isotope analysis.

In particular, 9 major elements   (Al., Fe, Mn, Mg, K, P), 27 trace elements: (Li, Be, B, Sc, B, Cr, Co, Ni, Cu, Zn, Ga, Ge, Rb, Sr, Y, Zr, Nb, Sn Sb, Cs, Ba, Hf, Pb, Bi, Th, U), 14 EUT: (La, Ce, Pr, Nd, Sm, Eu, Tb, Gd, Dy, Ho, Er, Tm, Yb, Lu) and the ratio of isotopes Pb-, Sr- und Rb-:   (204/206Pb, 207/206Pb, 208/206Pb, 207/204Pb, 208/204); (85/87Rb); (84/86Sr, 87/86Sr, 88/86Sr87/86Sr) have been measured.

11 different Greek marbles have been analyzed from various marble regions, among them from the island of Thassos.

The chemical analyses were carried out by ICP – MS (PQ3 S Quadrupol MS), the mineralogical analysis by XRD, and the petrographical study by optical microscopy with thin sections. 

The study of the analytical results shows that the discrimination of marbles is possible based on the trace elements in combination with the isotope analysis. This discrimination is improved by comparison to the mineralogical and petrographical analysis.

For the statistical confirmation for this discrimination more samples will be analyzed, from each type of marble and quarry, as well as from more places from Greece and from the wider Mediterranean area.








F. Biricotti, M. Severi



Dipartimento di Fisica – Università di Roma “La Sapienza”, Piazzale Aldo Moro 2, 00185  Roma ,  Italy

tel. 0039 6 49913508   fax 0039 6 4463158,






A method has been developed for the measurement of some quantities that characterize the internal structure of white marble, in particular those of archaeological interest. This method is non-destructive and is based on the measurement of light diffusion into white marble, considered as turbid media. The measured parameters offer a signature for each sample. The instrumentation is transportable and self-contained, allowing its use in museums, quarries, and archaeological sites. The measurements have no cost beyond that of the instrumentation, and their analysis is almost immediate. Possible information about the quarries of origin can be obtained by comparing sample measurements with a database of specimens of known origin. Both the methodology and the instrumentation are briefly described.

A database has been built designed expressly to include any information suitable to describe quarry samples and artifacts of archaeological and artistic interest made with white marble. In addition to quantitative data that characterize their internal structure, the database also contains some pertinent images and the available humanistic and technical information. The database is managed by an application that also allows to select the recorded data in a large number of ways and to extract and to arrange in table their properties that are considered most meaningful. These tables can be exported into other commercial programs to perform statistical analyses. The structure of the application can be modified in order to make it suitable to the recording and the analysis of data of other types.










Beatrice Moroni1, Ilaria Borgia2, Maurizio Petrelli1



1 Department of Earth Sciences, University of Perugia, Piazza Università, 06100 Perugia, Italy

2 Department of Chemistry, University of Perugia, Via Elce di Sotto 8, 06123 Perugia, Italy







When dealing with geochemical approach in the archaeometry of chert a main factor to be taken into account is the complexity of the chemical pattern, which is the result of the combination of chemical signatures, inherited during the processes of transport, deposition, diagenesis, and alteration.

While variations in chemistry attributable to the source rocks are most likely to be represented by the Rare Earth Elements (REEs), which tend to be transferred into chert as part of the original sediment and are quite stable during diagenesis and weathering, variations attributable to diagenesis and weathering are well evidenced by alkali metals (Na, K, Mg, Ca) and related trace elements (Rb, Sr, Ba) which are easily mobilized during these processes. Therefore detailed geochemical analysis is necessary in order to obtain informations on the origin and provenance of chert.

The requirement of detailed geochemical characterization of chert tools may be in contrast with the basic demand of preserving them as much as possible. This leads to the necessity of improving non-destructive techniques of chemical analysis.

The aim of this work is to study the influence of weathering on the chemical characteristics of chert tools in order to find out the key factors for provenance studies and to test the validity of non-destructive analytical techniques in the characterization of chert.

Prehistoric chert tools from the National Archaeological Museum of Perugia, coming from different geologic formations and weathering conditions, were analyzed by portable X-ray fluorescence spectroscopy (PXRF) and laser ablation mass spectrometry (LAMS) in order to register any chemical variation from the inner, sound portion of material to the external, weathered surface.

Comparison between the results of PXRF and LAMS, and with the results of conventional X-ray fluorescence spectroscopy (XRF) performed on powdered sound core of the same tools gave the opportunity to test the potential of each method in respect to bulk, conventional analytical methods in the resolution of the archaeometric problem of provenance.





Classifying Maltese prehistoric limestone megaliths by means of geochemical data



JoAnn Cassar



Institute for Masonry and Construction Research, University of Malta, Msida MSD06, Malta








Two stone types, referred to as franka, a more durable type of stone, and soll, which is much less durable, occur within the Lower Globigerina Limestone member of the Maltese Islands. It is impossible to distinguish visually between them in their unweathered state. The weathering mechanisms of the two types of limestone have been amply studied and are known to differ.

Globigerina Limestone is one of two local materials used to build the Maltese prehistoric megalithic “temples”, which are the earliest free-standing stone monuments in the world. Weathering rates and forms of the megaliths vary greatly, some stones proving to be very durable, whilst others are in an advanced state of deterioration.

Extensive research over a number of years has led to the development, and publication, of a numerical methodology to identify samples of durable franka and less durable soll samples on the basis of geochemical data. It has been determined that Cluster Analysis, using the variables SiO2, Al2O3, TiO2, K2O and Fe2O3, can be successfully used to predict the weathering behaviour of Globigerina Limestone. After confirmation of the efficacy of this methodology on quarry and borehole samples, it was applied to twenty-two samples of Globigerina Limestone taken from megaliths of the prehistoric site of Ħaġar Qim. Thus, these megaliths have been classified, on the sole basis of geochemical data, as either consisting of the more durable franka or the more easily weathered soll limestone. Using previously established geochemical limits; only 2 out of the 22 analysed samples remain unclassified. The percentage of samples classified as either franka or soll types by using at least 4 out of the 5 predetermined geochemical variables is thus 91%. However, having 9% of unclassified blocks does not necessarily mean that it is the method, which is at fault. Even when working on quarry and borehole samples, some samples were in fact not classified as either franka or soll. One possible reason for this is geological, as here one is dealing with the natural environment, where changes occur gradually and irregularly over extended periods. Therefore, it would be reasonable to expect that besides having end members, it is also possible to find “intermediate” situations which represent “transition periods” in the surrounding environment, in the mode of sedimentation or in the input of clastic material.

The methodology used and the results obtained have strong implications for the conservation management of these megalithic sites. Determining whether the different rates and forms of deterioration of the Globigerina megaliths are due to intrinsic properties or else solely due to external conditions, besides being of academic interest, can be a potent instrument for planning the future conservation of the site. This could, for example, help site managers decide whether to build a shelter over the whole site or else solely over individual blocks, or whether to move other less durable, decorated megaliths indoors, as had in fact been done in the past.  Information derived from the use of this method could also tip the balance if and when re-burial is being considered. It can also be used to decide whether the carved Globigerina Limestone blocks, now housed indoors, can be returned to their original, unsheltered, position outdoors.







OF ancient marble ARTEFACTS



Carlo Gorgoni, Paolo Pallante



Department of Earth Sciences, University of Modena and Reggio Emilia, 19, Via S. Eufemia, I-41100 Modena, Italy ( (







The study of superficial isotopic alterations of ancient marble (and limestone) artefacts is important for the issues of: a) correct definition of provenance; b) authentication; c) reconstructing the conditions of the original location. In this study a large amount of literature data for statues, quarry faces, etc. was collected and assessed. The data was then implemented with the results of analyses conducted on some Roman sarcophagi of the Modena Archaeological Museum and on other sculptures. Since the inconclusiveness associated with this type of investigation often derives from uncertainty about the previous history of the artefacts under study, a careful sequential isotopic analysis was also carried out (at increasing depths below the exposed surface) on three limestone tesserae of a late Roman mosaic from Aquileia, the relevant characteristics of which were well known. In addition to demonstrating a close correlation between porosity (and permeability) of carbonate materials and their surface modifications, with obvious implications for fluid dynamics, in cases of sufficient porosity isotopic shifting was effectively identified, decreasing evenly from the surface to the unaltered underlying material. After generalising the interpretative model, extending it to a much wider scale (geological-formational), the critical examination of all the data from readings and literature permitted a discussion of the process of isotopic alteration on the basis of four distinct evolutionary (shifting) trends on the C-O correlation graph: two main negative ones and two minor positive ones, with all their (numerous) possible combinations. In addition to the primary characteristics of the stone materials and the alteration time, the observed complexity was correlated with a series of potentially influential conditions and processes: i) aerial exposure, marine immersion, burying, resulting in the artefact interacting with very diverse fluids; ii) climate and other environmental features; iii) alteration mechanisms (neogenetic dissolution-reprecipitation of carbonates) of purely chemical nature or with biological mediation, both direct and indirect; iv) natural contamination. Other relevant factors were also considered and discussed including, for example, the possibility of intrinsic natural isotopic dishomogeneity (primary) of the artefact, any mineralogical transformations (in particular calcitization or rarely dolomitization), and the processing that the artefact might have been subject to (impregnation with various substances, mechanical or chemical cleaning, etc.).




Gypsum as a building material at the Minoan site of Knossos



K. Kouzeli and E. Zgouleta



Stone Conservation Centre, 79 Pireos Str., 105 53, Athens, Greece, tel.: ++30 210 3231467, fax: ++30 210 3312814, e-mail:







In antiquity the use of gypsum as a building material was very limited because of its properties and its rarity. It was bound for interior and semi-exterior use, serving also decorative purposes.

The aim of this study is the characterization of the types of gypsum used in the Minoan Palace and the archaeological site of Knossos in Crete as well as an attempt to interpret the deterioration mechanisms responsible for the various deterioration patterns present at the monuments.

Part of the project concerns conservation treatments since preventive conservation, which is the most suitable alternative in the case of gypsum, cannot be applied to the whole extent of the site.








Philippe Blanc1, Annie Blanc1, Danielle Decrouez2, Pierre-Alain Proz2, Karl Ramseyer3



110 rue de Monceau, F-75008 Paris, France;

2Muséum d'histoire naturelle, 1 route de Malagnou, CH-1211 Genève 6, Suisse; ,

3Institut für Geologie, Universität Bern, Baltzerstrasse 1, CH-3012 Bern, Schweiz; karl.







Since the late eighties, numerous studies documented the successful application of cathodoluminescence to help decipher the provenance and quarrying sites of white marble used in antiquity as building stones or sculptures. All these studies were destructive, i.e., they used material which was extracted from the object under investigation. For well some statues extraction of material might not be possible. For these cases a portable cathodoluminescence detector was designed (Blanc, sous presse) which allows cathodoluminescence observations on the surface of archaeological objects. The portable cathodoluminescence cell uses the electron gun and the vacuum-system of a commercial cold cathode instrument and a digital camera. The cell is a cylinder with 44 mm outer-diameter and a view area of 20 mm. The upper side is closed by an optical port and the lower side rests on the object surface by means of a flexible seal which adapts to the surface roughness and assures a vacuum-tight connection.

Good results were obtained from clean, flat surfaces, but the results for pitted rough or slightly convex surfaces were also adequate. This new portable equipment was tested on artifacts from the Museum of Art and History of Geneva (Chamay et al., 2000) previously studied with a hot cathodoluminescence microscope (Ramseyer et al., 1989). First results revealed that this new equipment is capable of reproducing the cathodomicrofacies, i.e., colour, intensity and distribution of the luminescence, previously described from extracted material. Thus, this new approach has the great potential as a non-destructive and non-damaging method to aid the provenance determination of white marble.


Blanc, P. (sous presse). Démonstration d'une cellule portable de cathodoluminescence sur surface de marbre poli. Table ronde "Les roches décoratives dans l'architecture antique et du Haut Moyen Âge" Autun, 18-19 Nov 1999.

Chamay, J., D. Decrouez, V. Barbin & K. Ramseyer (2000). Deuxième complément au catalogue des sculptures en pierre du Musée d'Art et d'Histoire. Genava, Bulletin du Musée d'art et d'histoire, n.s. 48: 135-141.

Ramseyer, K., J. Fischer, A. Matter, P. Eberhardt & J. Geiss (1989). A cathodoluminescence microscope for low intensity luminescence. Journal Sedimentary Petrology 59: 619-622.





Fractal Analysis (FA) and Quantitative Fabric Analysis (QFA) Data Base of West Anatolian white marbles



J. Zöldföldi1, 2, Balázs Székely2



1 Landesdenkmalamt Baden-Württemberg, Berliner Str. 12 D-73728 Esslingen a.N., Germany

2 Institut für Geowissenschaften; Universität Tübingen, Sigwartstr. 10, D-72076 Tübingen, Germany &







Provenance analysis of archaeological artefacts requires reference databases to compare the measured values of the archaeological objects to those of the possible raw material sources. These reference data sets are available for the majority of the measuring techniques commonly applied in provenance analysis. However, some newly developed techniques, like image-processing-based geometric thin-section analysis (Quantitative Fabric Analysis, QFA) lack appropriate reference data.

To help fill this gap we have initiated the creation of reference data for the most common QFA techniques, as well as for Fractal Analysis (FA) computation for white marbles. To this end, primarily having the provenance studies of Trojan objects in mind, about 50 white marble samples from West Anatolian occurrences, mostly from the tectonic units Sakarya Zone and Menderes Massif/Tauride block (separated by the Izmir-Ankara-Erzincan suture) have been analysed. Two groups of parameters were derived on digitally enhanced images of marble thin sections: (i) measured parameters: perimeter, area, convex hull parameters (perimeter and area), orientation of the long axis and (ii) derived parameters: axial difference, perimeter/area ratio, shape factor.

The geometric properties of the individual grains were measured in an automated manner.. Average sample specific values were calculated from the measured data together with the statistical parameters of the distribution. Based on these results samples of the two aforementioned main tectonic units can be separated. Among others, the average perimeter/area ratio is found to be useful for the separation. Along the axes of combination of selected fractal-distribution-related parameters characteristic gap forms in the abstract space of parameters. We interpret the separation in the geometric parameters of the as a possible consequence of the different thermotectonic evolution of the hosting rocks, which may have influenced the grain geometry pattern.

The data points belonging to the occurrences from the Menderes Massif and Tauride block form well defined clusters, although more overlapping in the case of the Sakarya Zone tectonic unit is typical.







F. Antonelli1, S. Cancelliere1, L. Lazzarini1, A. Solano2



1 Laboratorio di Analisi dei Materiali Antichi - Dip. di Storia dell’Architettura, IUAV-Università degli Studi, Venezia

2 Museo Archeologico di Nicotera







Granitoid rocks are among the most used stone materials for columns and pillars during the Roman period, when the main lithotypes employed in Rome and the Mediterranean Provinces were exploited in Egypt (Aswan and Eastern Desert), in Asia Minor (Troas and Mysia) and in Italy (Calabria, and the islands of Elba, Giglio, Sardinia). Some of these rocks show very similar macroscopic features so that their naked-eye distinction is often very difficult. This is also the case of the "Granito del Foro” (or Marmor Claudianum), from the Gebel Fatireh, in the Eastern Desert, and the “Granito di Nicotera” (Calabria), both exploited from the 1st century AD onwards. These are two granitoid rocks featuring a grain size of 1-1,4 cm, and a white-greyish colour with black patches. In order to obtain a certain microscopic and geochemical discrimination of these rocks, and to provide the first-known petrographic classification of the "Granito di Nicotera", a mineralogical-petrographic study of several samples from ancient and modern fronts of the Nicotera's quarry was undertaken.

The results are presented here and compared with the corresponding literature data known of the Marmor Claudianum. They indicate that an easy microscopic differentiation between the "Granito del Foro” and "Granito di Nicotera" is possible due to the significant presence of horneblende in the first and of muscovite in the second.







L. Lazzarini1, M. Luni2, B. Turi3,



1 Laboratorio di Analisi dei Materiali Antichi - Dipartimento di Storia dell’Architettura, IUAV università degli studi, S.Polo 2468/B, 30125 Venezia, Italy.

2 Istituto di Archeologia e Storia dell’ArteAntica, Università di Urbino, via del Balestriere 2, 61029 Urbino, Italy.

3 Dipartimento di Scienze della Terra, Università degli Studi “La Sapienza”, Piazzale A. Moro 5, 00185 Roma, Italy.







Several past and recent excavation campaigns conducted by Italian, English and American teams of archaeologists have brought to light an impressive number of Greek archaic sculptures and architectural elements made of white crystalline marbles. Since Lybia has no marble sources, it is obvious that all these marbles were imported from other Mediterranean regions. To determine such sources, the marble of more than 30 artifacts was sampled and subjected to petrographic examination on thin section and to C & O isotopic analysis with a mass-spectrometer.

The majority of results obtained indicate that the great kouroi, korai akroliths and architectural elements (mainly sphinx, acroteria, were made out of marble from the open-pit quarries of Lakkoi; one small torso of a kouros was carved out of the famous Parian lychnites from Stephani; other two kouroi were of Naxian and Pentelic marble respectively.

These results are the utmost importance to archaeologist since they clearly prove not only a preferred relationship between Cyrene and the island of Paros, but also contacts with the quarries of Naxos and Attica. The identification of a kouros of Pentelic marble is particularly interesting as it is one of the very few sculptures made with this marble of archaic times.











Institut Francais D’Etudes Anatoliennes (IFEA), Nuru Ziya Sok. 22, Palais De France, 80072, Beyoglu, Istanbul.







The prehistoric ground stone implements like ground-edge axes have been widely studied in Western Europe for the last decades especially in terms of typology, technology and their provenance for the raw material. However, ground stone production sites are not so common in Eastern Europe. One of the three newly discovered production sites in southeastern Thrace is Yartarla near Sarköy, Tekirdag (N 40°41'05" E 027°01'50"). The site, which was also a settlement between the Late Chalcolithic and Early Bronze Ages, provide us some details about the raw material preference for the production of especially polished stone axes, adzes and chisels as well as hammer stones. In order to understand the distribution patterns of these tools starting from the production site, petrographical analysis were made both in Yartarla and other prehistoric sites in the region. These studies show us that most of the tools were manufactured from metabasite, which has a limited outcrop occurrence in Ganos Mountains. As metamorphic rocks suitable for axe production have not documented in other parts of Anatolia until now, this research may be considered important. In this study, the quarry, the production site and finished tool use relation are discussed with a brief site-catchment analysis concentrated mainly on Yartarla.


Key words: Prehistoric ground stones, polished stone axes, southeastern Thrace, Tekirdag, geoarchaeology, metamorphic rocks, petrographical studies.








Nicoletta Marinoni1, Alessandro Pavese1, 2, Luca Trombino2, 3



1 Dipartimento di Scienze della Terra "Ardito Desio", Università degli Studi di Milano, Via Botticelli 23, 20133, Milano (I), tel 0039-02-50315775, fax 0039-02-50315597, E-mail,,

2 Istituto per la Dinamica dei Processi Ambientali, Sede di Milano, Via Mangiagalli 34, 20133 Milano

3 Dipartimento di Scienze della Terra "Ardito Desio", Università degli Studi di Milano, Via Mangiagalli 34, 20133, Milano (I), tel 0039-02-50315533, fax 0039-02-50315494, E-mail,







Digital image processing (DIP) is a computerised analysis of digital images. It has been widely used in several disciplines, including medicine, biology and material science. Relatively, only few applications have been performed in Earth Science thus DIP appears a suitable tool for defining rock texture. Nowadays grain morphology and fabric of sedimentary rocks are usually performed in a manual way; in particular the grain size distribution is detected by mechanical sieving. This traditional method is not suitable since only a rough measure of the width of the particles is carried out (the relative error is greater than 10%). Several measurements are also necessary to obtain accurate and reproducible information about rocks; indeed the quality of the observation largely depends on the interpretative skills of the operator. This provides a motivation for automating the measurement processes, which can be carried out by image analysis methods.

In this study the grain size distribution of sedimentary rocks, in particular sandstone, is carried out by means of DIP methods and petrographic analysis. An evaluation of the rock mineralogical composition is also attained.

Thin sections have been prepared and successively captured using a digital camera connected to the microscope. By the application of DIP to microscopic images rock texture has been investigated. The different optical features of minerals and rock fragments seem to be a useful tool to distinguish among particles according to their mineralogical nature.

The results are compared to those obtained by mechanical sieving: a strong correlation between DIP and manual measurements is achieved. Digital image technique appears faster and more accurate than the conventional method. Indeed by the application of DIP method on thin sections an objective and quantified grain size observation is carried out. However DIP yields more information about rock texture than the manual method and can be considered a better alternative for grain size distribution measurement in comparison with the traditional mechanical sieving.








Donato Attanasio1, Norman Herz2 and Susan Kane3



1ISM-CNR, P.O.Box 00016, Monterotondo Staz., Roma, Italy,

2Dept. of Geology & Center for Archaeological.Science, Univ. of Georgia, Athens, Georgia 30602, USA

3Dept. of Art, Oberlin College, Oberlin, Ohio 44074, USA,







The first large scale investigation of the white marbles imported and utilised at Cyrene for over one thousand years was carried out in the 1980’s using isotopic analysis as the main provenancing tool. Recent work, based on more systematic sampling and the use of considerably improved analytical and data processing techniques, is providing a more detailed picture of the marble types and distribution.

More than 300 marble samples, including 175 sculptural marbles, collected at Cyrene in the past two years are now being analysed. The long-term goal is to develop a unified classification model for all Cyrenaican marbles using a multivariate approach that combines several provenancing techniques. Isotopic, EPR and petrographic analyses of the 22 Roman or Hellenistic sculptures sampled at the Extramural Sanctuary of Demeter have been used for preliminary set up and tuning of the method.

At this stage, assignments are based on two different, independent databases—one isotopic and the other EPR/petrographic. This means that a truly multivariate approach is not yet possible, although the provenances obtained using the two different methods can be critically compared to obtain a final assignment.  The EPR/petrographic database includes only 9 possible sources: Afyon, Carrara, Naxos, Paros/Stefani, Paros/Chorodaki, Pentelicon, Proconnesus, Thasos/Aliki (calcitic) and Thasos/Cape Vathy (dolomitic), while the isotopic database is considerably more comprehensive.

The final results indicate that 12 (55%) of the statues are Parian (4 identified as lichnites and 8 as Chorodaki marbles), 7 (32%) are Pentelic marble and 2 (9%) are Thasian dolomitic, while one of the samples could not be assigned with confidence. The isotopic and EPR/petrographic methods appear to be usefully complementary and disagreements can almost always be easily and unequivocally solved.

Preliminary attempts to extend the assignment procedure to other samples indicate that the prevailing presence of Parian and Pentelic marbles shown by the sculptures of the Demeter sanctuary is a distinctive feature of most Cyrenaican samples.







Donato Attanasio1, Giuseppe Mesolella,2 Patrizio Pensabene,2 Rosario Platania3, Paolo Rocchi1



1ISM-CNR, P.O.Box 00016, Monterotondo Staz., Roma, Italy,

2 Dip. di Scienze dell’Antichità, Università di Roma “La Sapienza”, Roma, Italy

3 IMIP-CNR, P.O.Box 00016, Monterotondo Staz., Roma, Italy,







The first approach to Villa Adriana transmits undoubtedly an idea of wealth and opulence stemming from the pervasive presence of precious marbles and other expensive building materials. Close archaeological inspection, however, discloses a partly different perspective demonstrating that careful use of the resources and economical concerns were also important. Although solid marble elements such as trabeations, columns, capitals are obviously quite common, the technique of marble veneering is also frequently employed not only for walls and pavements, but also for covering calcareous architraves, capitals and columns; stucco work, in imitation of marble, is also relatively frequent.

The quality of workmanship is inhomogeneous, much coarser working being employed in the case of less important or less visible architectural elements. The Imperial Triclinium and the building with Three Exedrae are convincing examples, among others, of the above statements.

These building peculiarities are strictly interconnected with the choice, distribution and availability of marble varieties and detailed provenancing work is expected to provide support to the above considerations, sheding new light on the building process of this unique Roman monument.

Systematic work has been undertaken collecting, up to present, 130 marble samples whose provenance was determined using a method based on the combination of EPR spectroscopy and petrographic data. Eight different marble sites, considered to be possible provenance sources, were included into the database. They are: Afyon, Altintaş, Carrara, Hymettus, Paros, Pentelicon, Proconnesus and Thasos. The large majority of the samples, over 70%, originates from Carrara, with Pentelicon distant second (26 samples, 20%) and a few Proconnesian and Thasian dolomitic samples (8 and 2, respectively).

More detailed analysis indicates that different marble varieties were often used at Villa Adriana depending on the type of item considered. The columns are largely made of Pentelic marble, with a few Proconnesian examples, while all the 19 capitals that were sampled appear to come from Carrara.  The attempt to economize on the use of marble does not seem to concern Pentelic, but refers, almost exclusively to buildings using Luna marble for the decoration. The possible implications of this rather unexpected result will be discussed.








N. L. Ninis1, S. Kourkoulis2, V. Bakolas3



1 Civil Engineer, Committee for the Preservation of the Epidauros Monuments, GR-210 52, Ligourio, Greece

2 National Technical University of Athens, Department of Mechanics, Lab. of Testing and Materials, 157 73 Zografou Campus, Theocaris Bld. Hellas,

3 National Technical University of Athens, Department of Mechanics, Lab. of Testing and Materials, 157 73 Zografou Campus, Theocaris Bld. Hellas.







The mechanical characteristics of a conchyliates stone are examined here. A similar material was used by ancient Greeks for the construction of a number of celebrated temples. The most characteristic example is the Temple of Zeus at Olympia.

The investigated stone is peculiar in the sense of having a microstructure that compares to the dimensions of the conventional laboratory specimens and therefore interacts with the observed behaviour in unconfined compression. Special emphasis is, therefore, placed on the scale effect, which is expected to have a marked effect.

Three different specimen sizes were used in the study. The full stress-strain curve of the material was obtained for each size, under unconfined compression loading. It is concluded that although the peak stress of the material if relatively low (of the order of a few MPa) its energy absorption capacity is remarkably high due to the fact that after the peak stress the curve proceeds smoothly with a very low inclination reaching strain levels of the order of 2% to 4%, depending on the level of pre-stressing, before final collapse [1].

The influence of specimen size on the experimental behaviour is proved to be remarkable, at least for the classes of specimens used. On this basis some tentative conclusions are drawn on how the material would be expected to behave in real life situations. The observed behaviour is simulated numerically.

Another important aspect examined is the strain rate dependency of unconfined compression strength in porous stones with cellular microstructure as the one investigated here, which, for laboratory size specimens at least, could be classified as particulate materials. It is concluded that the results of conventional quasi-dynamic tests are completely different compared to the respective ones obtained under quasi-static loading, which simulates much better the real loading conditions of the structural elements.








U. Muss1, A. Bammer2, L. Moens3, P. de Pape4, J. de Donder, K. Koller5, M. Aurenhammer5



1 Österreichisches Archäologisches Institut, Franz Klein-Gasse 1, A- 1190 Wien, Austria,

3 Dept. of Analytical Chemistry, Ghent University, Proeftuinstraat 86, B-9000 Ghent/Belgium,

4 Dept. of Geology and Soil Science, Ghent University, Krijgslaan 281 S8, B-9000 Ghent/Belgium,  

5 Institute for Studies of Ancient Culture, Austrian Academy of Sciences Baeckerstrasse 13 PF 8A, 1010 Vienna, Austria,

6 Osterr. Archaologisches  Institut Franz Klein-G 1, Wien A-1190, Austria, tel. +43 1 4277 27140, fax.+43 1 4277 9271,








The use of marble was a rather late phenomenon at Ephesus. The first known constructions were built in a soft, marly yellow limestone, which comes from Igdelitepe, a quarry close to the sea. The early sanctuaries at the Artemision site are built in this material. This limestone was easy to work and its transportation by sea was easily accomplished.

The first use of marble at Ephesus is noted at The Artemision. The famous temple of Artemis was built in marble at different times. The archaic temple of Croesus was constructed around the middle of the 6th century B.C. and again after a destruction by fire after 356 B.C. During the last years it was possible to take samples both from the architecture and the sculptural decoration of this archaic Dipteros.

However the first marble in the Artemision was used for an even older building lying west of the Temple, for the so called Hekatompedos, a building which has never been completed but the beginning of which can be dated around 600 B.C. Here a foot of a wall in marble has been preserved in situ, a sample of which has also been studied recently.

Concerning the huge courtyard-alatar - also lying west of the temple- with building phases from the 6/5th and 4th centiury B.C. ist has been possible to take samples not only from the remaining foundations which belong to the earliest remains in the altar area, but also from a lot of pieces of architecture and even sculpture which belonged to the 4th century building.

The investigations by isoptopic analysis show that the marbles for all archaic buildings can be separated from those marbles, which belong to structures of the 4th century B.C.








P. Lapuente1, B. Turi2, Ph. Blanc3



1 Petrología y Geoquímica. Dpto. Ciencias de la Tierra. Univ.Zaragoza, 50.009 Zaragoza (Spain),

2 Dpt. Scienze della Terra. Univ. “La Sapienza”, Piazzale Aldo Moro, 5. 00185 Roma (Italy),

3 Dépt. Stratigraphie, ESA 7073. Service MEB, case 104, Univ. Pierre et Marie Curie, 4 place Jussieu, 75252 PARIS Cedex 05 (France,







This paper reports the results of a study designed to investigate the marble and coloured stones provenance used in the theatre of Caesaraugusta. A combination of techniques using petrography, cathodoluminiscence and isotopic analyses has provided basic information for understanding the stone sources.

The building was first constructed in the Tiberian period with later additions and modifications during the Flavian dynasty. Its complex internal structure was built entirely from concrete (opus caementicium). The seating benches were raised on a series of concentric concrete vaults.

This monumental stage setting, accidentally discovered in 1972, has a total surface of 7000 m2 and a capacity for 6000 spectators. A vigorous programme of archaeological excavations has been underway throughout the last few years, and pieces from all the historical periods of Zaragoza have been collected. Many remains have now been restored and a sophisticated polycarbonate roof has been installed to preserve the theatre and allow this magnificent testament to nearly two millennia to be viewed.

Most of the building stones were obtained from local sources, especially gypsum alabaster which was used not only for masonry but also for decorative architecture (capitals, columns, cornices, bases). Local limestones shape the arc of proedria, forming the edge to the semicircular orchestra. White marbles and Rosso Antico were used for architectural ornaments and small decorative mouldings. An enormous range of different marbles and coloured stones, in large and small slabs covered the orchestra. These came from quarries as far afield as the Saint Béat district, in the French Pyrenees and from quarries on the coast of the Tarraconensis province (Brocattello near Dertosa, Buixcarró close to Saetabi and Santa Tecla in Tarraco). In addition Numidian, Turquish, Greek and Carrara marbles have also been identified. In white sculpture, an exquisite head of a princess carved on Paros lychnites and a colossal bust of Diana Hunter moulded on Saint Béat marble must be pointed out.









Lorenzo Lazzarini1, Fani Athanasiou2,



1 Laboratorio di Analisi dei Materiali Antichi - Dipartimento di Storia dell’Architettura, IUAV università degli studi, S.Polo 2468/B, 30125 Venezia, Italy.

2 Archaeological Museum of Thessaloniki, Odos Andronikou 6, 54621 Thessaloniki, Greece.







The recent study of the great opus sectile of the Oktagon of Galerius’ Palace at Thessalonica has shown the predominating, characteristic, presence of a beautiful pinkish-gray breccia identified by direct comparison with stone collections of the XIX c. as “Breccia Policroma della Vittoria”, a stone hitherto of unknown origin.

Further research and direct observation of modern buildings of Thessalonica has allowed the localisation of the ancient (and modern) quarries of the breccia at Akrini, a village close to the ancient Aenei, province of Kozani.

Petrographic analyses of thin sections prepared from samples collected from the Oktagon and the quarries have shown identical fabric and composition thus confirming the proposed provenance of the breccia.

To-date, the presence of this stone seems limited to Late-Imperial or Early-Christian contexts of Thessalonica, Philippi and Aquileia.








Lorenzo Lazzarini1, Bruno Turi2,



1 Laboratorio di Analisi dei Materiali Antichi - Dipartimento di Storia dell’Architettura, IUAV università degli studi, S.Polo 2468/B, 30125 Venezia, Italy.

2 Dipartimento di Scienze della Terra, Università degli Studi “La Sapienza”, Piazzale A.Moro 5, 00185 Roma, Italy.







Recent fieldwork conducted on the island of Chios (Greece) has led to the discovery and preliminary study (Lazzarini 2001) of the ancient quarries of a new “nero antico” (or “bigio morato”) in the low hills on the peninsula of Margaritis (also known as Marmaro) close to the northern town of Kato Kardamila.

The size of the quarries is small-to-medium, with the largest on the western part of the hills showing a front some 60 m long and 5-10 m high. The considerable amount of stone extracted indicates that this is probably the most quarried black stone of antiquity. It was used locally at least from the IV c. BC for funerary stelae and pedestals, for statuary, and for columns and blocks in Roman times, as testified in the lapidary of the local archaeological museum.

To characterise the stone and differentiate it from other similar black limestones petrographic studies of thin sections and C & O isotopic analyses were performed. This “nero antico” may be classified as a wackestone (Dunham) or dismicrite (Folk) characterised by 5-10% of bioclasts of gastropods, bivalves and ostracods. The presence of locally-concentrated peloids as well as of diffused carbonaceous particles and limonitised pyrite has also been observed.

The isotopic analyses of 8 samples gave the following intervals:

δ 18O(PDB)= - 1.77 / +3.37; δ 13C(PDB)= - 0.02 / +1.71.

These data and the petrographic features point to a calcareous mud deposited in a rather closed lagoon and allow a quite clear distinction of this “bigio morato” from the many others used in antiquity. These results have been tested on some sculptures of Roman Age.








L. Lazzarini



Laboratorio di Analisi dei Materiali Antichi - Dipartimento di Storia dell’Architettura, IUAV università degli studi, S.Polo 2468/B, 30125 Venezia, Italy.







The paper presents the map of distribution in the Mediterranean area of the most important stones used by the Romans and later re-used in Christian and Islamic monuments.

The maps have been constructed with reference to a database including more than 6000 records of stone occurrence in about 400 ancient sites. They allow an immediate, visual appreciation of the provinces of the Empire reached by each lithotype, the concentration areas and the re-distribution of stones in the Middle Ages.








Laura Giordani1, Massimo Oddone2, Sandro Meloni2



1Dipartimento di Chimica Generale, Università di Pavia, Viale Taramelli 12, Pavia (Italy),

2Istituto C.N.R. per l’Energetica e le Interfasi – Sezione di Pavia - Dipartimento di Chimica Generale, Università di Pavia, Viale Taramelli12, Pavia (Italy),,







Many investigations have been reported in the literature on the provenance of raw materials used to manufacture artifacts of archaeological or historical interest. In many cases chemical studies involve the evaluation of the elemental composition, specifically trace elements. Provenance assignments are usually carried out by a statistical comparison with databases on artifacts of known origin or on prime matter sources known in the antiquity.

The University of Pavia (northern Italy) has promoted a research program on the conservation status of the Certosa di Pavia monument (Chartreuse of Pavia). An investigation was started on the evaluation of the decay of the marble façade of the monument and on the possibility to carry out provenance studies even on materials deteriorated for a very long exposure (up to 5 centuries) to atmospheric agents.

White marble samples have been collected from the monument façade, mostly ornaments and statues. Historical records assign the used marbles to the Apuan area (Carrara) or to the western Alps area (Valdossola). The decay status of the samples was investigated by FT-Raman spectroscopy and X-ray diffraction. Relevant quantities of gypsum (CaSO4.2H2O) have been detected mixed up with the original calcite matrix.

Trace element determination was carried out by instrumental neutron activation analysis using the TRIGA Mark II reactor of the University of Pavia. The abundances of a number of elements, together with their precision and accuracy, is presented and discussed.

Meaningful increases of the content of some elements such as Ni, Zn, Co and Cr, known to derive from deposition of atmospheric particulate, have been observed with respect to the elemental database on the composition of white marble from the Apuan area. This confirms the role played by human activities in the deterioration of cultural heritage. In addition an enrichment of light rare-earth elements has been observed with respect to the heavy ones.

If the usual multivariate analysis procedures for provenance assignment are applied to the matrix data of examined samples and   to the database on the white marble quarries of the Mediterranean area, some discrepancies occur and provenance assignment becomes questionable. As a consequence, more investigations are required on the limits of application of standard provenance assignment procedures to spoiled samples and on the depth and time dependence of pollution penetration.









M. Bruno



Via del Pellegrino, 130, 00186 Rome, Italy,







Recently several parallelpipedal marble blocks and quarry-rounded column shafts were recovered from the bed of the Fossa Traiana in the area of the excavation of Ostia antica. The most represented marble quality is the green cipollino from Eubea, attested by eight column shafts of different size, followed by five shapeless parian marble blocks, three fragmentary column shafts of chian marble, one block and one column of proconnesian marble, one block of frigian marble and one huge column shaft of africano from Teos in Asia Minor.

Numeral inscriptions are visible on karystian column shafts, frigian and parian marble blocks, while only two parian blocks bear consular inscriptions referable to the II century A.D. These rough  quarry blocks belong undoubtedly to the stockpile of the marble yard of Portus and to the same group of items discovered in the ’50, along the left river bank of Isola Sacra, and in the ’70, in the river bed of the “Canale di Fiumicino”.

The new findings attest that many other ancient quarry items are still present in the river bed of the Fossa Traiana, but also that in the marble yard of Portus rough shafts were refined and  restored as well attested by the chian marble columns reinforced with lead-coated iron cramps.








B. Moroni1, P. Lapuente2



1 Department of Earth Sciences, University of Perugia, Piazza Università, 06100 Perugia, Italy

2 Department of Earth Sciences, University of Zaragoza, 50.009 Zaragoza, Spain







In Central Italy chert, one of the most abundant, utilized, and enduring raw materials in prehistory, is a typical feature of many Meso-Cenozoic geologic formations where it originated from substitution of carbonate sediments by silica during diagenesis, that is the number of processes that transform loose sediment in sedimentary rock.

Owing to this origin chert can contain relict structures of the original sedimentary rock, and have a chemical composition strongly conditioned by a series of factors including chemistry of the original sediment, mode of diagenesis and type and extent of chemical selection by natural weathering. Therefore study of textural and geochemical characteristics of chert is a basic approach in the characterization of cherts from different geologic formations and, hence, a very useful tool in provenance archaeological studies of chert artifacts.

This study was carried out on a selection of chert tools of Palaeolithic to Neolithic industries from some localities in Umbria with final aim to recognize the geologic formations and, possibly, the sites of provenance of the constituent material.

The tools were preliminarily observed at macroscopic scale and the constituent pieces analyzed for colour, size and shape. Then selected tools underwent stratigraphic-micropalaeontologic study in thin section and chemical characterization by X-ray fluorescence spectroscopy coupled with principal component statistical analysis. In addition cathodoluminescence was performed on a few samples of less silicified material as a support to textural analysis in deciphering relict structures.

Integrated modeling of the results and comparison with reference material taken in outcrop from rock fronts and detrital formations led to the identification of the geologic formations and sites of provenance of almost all material for a better understanding of the procurement procedures and migration patterns in Central Italy.








P. Tsompos, N. Epitropou, Ch. Skilodimou



I.G.M.E., Messoghion str.70, 11527 Athens, Greece,

FAX number: 2107779467, Phone number: 2107758830,








The marble quarries of Aliki Peninsula, in the southeastern part of Thasos island, were the main marble sources for the exploitation of white marble during the Classical, Hellenistic, Byzantine and Roman periods, in North Greece.

Geologically, the white to bluish-white colored, well – bedded thin – partly, medium – grained marbles of Aliki quaries, belong to the metamorphic Alpine system of Profitis Elias marbles, which also consist of schists, gneisses, and amphibolites.  

Tectonically, it constitutes a tectonic horst, extended in NE/SW direction. The greater area is bounded by a double system of vertical normal faults attaining NE/SW and NW/SE directions.

Geomorphological the Peninsula is a monocline structure. The morphology of this structure has been altered due to mining activities and their debris were placed along the monocline at the northwestern part of the Peninsula. The main elevation of those debris cones is 30 meters above sea level. Beach rock formations have been observed on the excavation’s place at the southern seaside of the Peninsula.

The use of remote sensing data (air photographs in scale 1: 2000) and GIS techniques revealed that the strike and the dip of Aliki marbles are at 50o NE/SW and 20o SE directions respectively. Based on the same data and taking into account the tectonism and geomorphology of the study area, the average visual thickness and the volume of the excavated marbles as well as the volume of the derived debris deposits were estimated.    






K. Koller


Institute for Studies of Ancient Culture, Austrian Academy of Sciences, Baeckerstrasse 13 PF 8A, 1010 Vienna, Austria,






The Terrace House 2 in Ephesus was excavated from 1962 to 1985 by the Austrian archaeologist H. Vetters. Because of the excellent condition of its lower floors, its fine decoration comprising marble panels, mural paintings and mosaics and its important finds (sculptures, small artifacts, household goods) the Terrace House 2 is one of the most important residential monuments of imperial times in the Eastern Mediterranean region. According to the recent research four building phases could be differentiated from about 50 AD onwards; the seven housing units of the Terrace House 2, including their decorations and inventories, were destroyed by an earthquake around 262 AD.

One aspect of the decoration of the walls of Terrace House 2 are the marble décors – so far neglected in the research – which were increasingly selected since the 2nd century AD. The wall-systems which were partly structured by pilasters, e.g. in the marble hall in housing unit 6 or also in a now reconstructed banquet hall in the first floor of the housing unit 4, and had high-quality capitals; their forms confirm the dating. Within the marble décors of Terrace house 2 about 30 different kinds of marble and coloured stones were used: local white and grey marbles from quarries in the surroundings of Ephesos but also numerous coloured marbles and stones imported from the provinces of Asia minor, Greece and the Greek islands, Egypt and North-Africa. The diversity of the local and imported marble and coloured stone not only reflects the luxury of the furnishing of the urban elite in their houses but also casts a light on the trade routes and the prosperous economy of the city.





The opus sectile pavements from the baptistery in the Xanthos Cathedral (Lycia, Turkey)


M.-P. Raynaud1 and J.-P. Sodini2


1 Centre de recherche sur la mosaïque antique Henri Stern, ENS, 45, rue d'Ulm, 75005 Paris

2 Universite Paris 1, Institut Universitaire de France, 3 rue Michelet 75005 Paris




The baptistery is located at the northeastern side of the eastern basilica excavated at Xanthos, probably the Cathedral of this city.

When it was erected, in early Christian times, the baptistery was a tetraconch covered by a dome decorated with mosaics and having its walls lavishly furbished with a marble veneer. In the center, stood a huge baptismal font, separated in two halves by a narrow path allowing the clergy to walk on it, set across the staircases and the font itself: every part of it was covered with white marble and slate. The opus sectile pavement of the room was also made of white marble slabs lined by narrow slate strips.

In the XIth century, the baptistery was reoccupied and transformed in the catholicon of a small monastery: a deeper projecting apse was added in the east with a small synthronon and a sanctuary set in it, closed by a templon. The font was filled up and closed by huge white marble slabs. The opus sectile pavement was then carefully refurbished, following middle Byzantine decorative trends. From place to place, new complex compositions, like rosaces, were introduced, reusing coloured marble pieces which were roughly recut on the spot in tiny geometrical elements: triangles, squares, octagons and other motives, using colour contrasts, enhanced the former pavement.

Many of these reused coloured marbles have been identified and give an insight of the marbles available from spolia of roman Xanthos. The origin of white marbles is as usual difficult to trace. Proconnesian grey marble can be recognized, to our opinion. But there are also many other white marbles. One variety may come from the near-by quarry of Mylasa.








P. Morbidelli1, P. Tucci1, E. Azzaro2, P. Pensabene3



1 Dipartimento di Scienze della Terra, Università di Roma “La Sapienza”, P.le Aldo Moro, 5 I-00185 Roma Italy;,

2 Dipartimento di Chimica e Fisica della Terra, Università di Palermo, Via Archirafi, 36 I-90123 Palermo;

3 Dipartimento di Scienze dell'Antichità, Università di Roma "La Sapienza", P.le A. Moro 5, I-00185 Roma, Italy







The abundant finding of millstones in all the most important archaeological sites of the Mediterranean basin lead the researchers of Earth Science to co-operate with archaeologists with the aim to identify the most probable source sites of the stone materials used to realised these manufactures.

The present work wants to furnish a contribution, through mineralogical-petrographic-chemical and isotopic data, to solve this problem. Seventeen millstones (also called hourglass or Pompeian-style for their shape) founded in the archaeological site of Ostia Antica (Roma, Italy) in the bakeries located along Via dei Molini (10 samples) and along Via Semita dei Cippi (7 samples). Petrographic analyses on thin sections, X-ray diffractometry (XRD), chemical (XRF; major and trace elements), and Sr isotopic analyses were carried out on all samples. The utilised materials is a phonolitic lava (HK serie with 69.48<I.A.<82.71) characterised by the paragenesis of large, heuedral leucite phenocrysts, pyroxene and plagioclase grains in a groundmass formed by minute grains of plagioclase, K-feldspar, pyroxene leucite and oxides. 87Sr/86Sr isotopic ratios (min. value: 0.709744±20, max. value 0.711048±28) fall in the field of the Districts of Vulsini, Vico and Alban Hills (Roman Comagmatic Province). The last two Districts can be rejected (Vico: higher P.I., occasional presence of sanidine as phenocrysts, occurrence of biotite in the groundmass; Alban Hills: lower SiO2 content, rare plagioclase and scarce occurrence in the area of these lithotype). As for Vulsini District (Orvieto: ancient quarry indicated by Peacock, 1986; peripheral sectors of the Eastern Vulsini District), known to be an important production area of the millstones, the strong petrographic similarities between our samples and Vulsini ones are not confirmed (for all the samples) by the trend of some significative elements (La vs Ce; Sr vs Rb; Rb vs Cr).




Stone materials of the roman villas around lake Garda (Italy) (Poster)



E. Roffia1, L. Folli2, R. Bugini3



1 Soprintendenza Archeologica della Lombardia – via De Amicis 11, 20 Milano, Italy

2 Viale Rimembranze 37, 20075 Lodi, Italy

3 Istituto CNR Conservazione Beni Culturali Sez. “Gino Bozza” – p. Leonardo da Vinci 32, 20133 Milano, Italy







Different Roman villas are located on the western shore of lake Garda (Lombardy – Italy): imposing building characters and sumptuous decorations are the most important of the Roman architecture in Northern Italy.

Three villas located in Sirmione, Desenzano and Toscolano were examined in order to investigate the use of stone materials in buildings and decorations. These villas were built starting from the end of 1st century BC to the beginning of 1st century AD and were inhabited for a long time. Desenzano and Toscolano villas lasted until the first half of the 5th century AD with important enlargements and rebuilding, Sirmione villa on the contrary was destroyed in the middle Imperial Age.

The study of stone materials was performed by petrographic analyses on a great number of samples coming from building structures, decorative works, mosaics, wall veneerings, sculptures etc.

The most important sources of building materials are indentified in the hill region between lake Garda and Verona, but wall veneering in Toscolano shows a great variety of coloured marbles coming from quarry sites in Egypt, Greece and Anatolia.

In particular, building structures in Sirmione villa are made by a limestone outcropping in the same area of the villa (Scaglia lombarda formation - Eocene/Upper Cretaceous). The other stoneworks as columns and pilasters are made by a white limestone called “pietra Gallina” (Eocene) coming from monti Lessini (Verona) and by a yellowish limestone called “pietra di Vicenza” coming from Monti Berici (Padua - Oligocene); vaults are made by travertine, a very porous stone probably coming from recent deposits near lake Iseo (Brescia).

The wall veneerings and the other stoneworks of Desenzano villa are made by nodular limestone (Rosso Ammonitico formation - Upper/middle Jurassic) from monti Lessini (Verona), oolitic limestone from San Vigilio (eastern shore of lake Garda - Lower Jurassic) and micrite limestone called “Scaglia” from monti Lessini (Verona – Upper Cretaceous). The mosaics and opera sectilia for flooring are made by red and white nodular limestones (Rosso Ammonitico), white oolitic limestones (San Vigilio), black limestones of uncertain provenance. The sculptures are made by white marbles different in texture and grain size, but the reduced dimensions of sculpture fragments didn’t allowed further analyses.

Toscolano villa is today almost covered by heterogeneous buildings; the most important findings are fragments of wall veneering, opera sectilia, mouldings etc. Eleven coloured marbles were classified: Africano, Breccia corallina, Breccia di Settebasi, Cipollino, Giallo antico, Palombino, Pavonazzetto, Porfido rosso antico, Porfido serpentino verde, Portasanta, Rosso antico, in addition to different kinds of white and grey marbles, different black limestones and sandstones.








R. Tykot1, J. J. Herrmann, Jr.2



1 Department of Anthropology,University of South Florida, 4202 East Fowler Ave., SOC107, Tampa, FL 33620-8100 USA;

2 Museum of Fine Arts, 465 Huntington Ave., Boston, MA 02115 USA;







Normally colored marble is identified by eye, but it may eventually be possible to sort out some ambiguities with archaeometric support. The case of dark gray marbles may be especially in need of this kind of clarification.  The borderlines between Bigio morato, the various Bigio anticos, Marmo bigio, Nero antico, Nero antico of Chios, and the gray marble of Aphrodisias seem particularly ambiguous. The gigantomachy ascribed to sculptors and marble from Aphrodisias from Valdetorres de Jamara in Spain has recently been termed both Nero antico (Castellano in I marmi colorati della Roma imperiale, 2002, cat. 7) and Bigio antico (Gregarek, Kölner Jahrbuch 1999, cat. E10). With this problem in mind, three dark gray marble sculptures in the Montemartini Museum were sampled. Their stable carbon and oxygen isotope ratios were determined by mass spectrometry and the results compared with published databases.

A Dancing Maenad belongs to a widely distributed group of replicas and variants with drapery in dark gray or black stone and flesh parts (usually missing) carved separately in white marble.  Heike Gregarek has collected examples in the Palermo Museum, Naples Museum, and Benevento Museum, Italy; Antalya Museum, Turkey; the Bardo Museum, Tunisia; and the Egyptian Museum, Cairo (Gregarek 1999: 231-234).  She has identified them variously as Bigio morato, Bigio antico, Nero antico, and basalt.  She terms the example in the Montemartini Museum Bigio antico (a stone coming from various sites in Asia Minor, according to Borghini et al., Marmi antichi, 1998: 158-159). The isotopic ratios of the dancer in the Montemartini Museum are compatible with a provenance from the quarries of gray marble at Aphrodisias (Lazzarini et al., ASMOSIA 5), and it seems legitimate to ascribe this statue to the Aphrodisians, who were known to have worked at Rome. It also seems legitimate to hypothesize that Aphrodisians may have produced some of the other dancing maenads found around the Roman Empire and that in some cases at least, they used their native stone. Gregarek calls the replica in Antalya Nero antico, a stone that comes from Gebel Aziz, Tunisia (Borghini et al. 1998: 254-255).  It seems more probable that this statue too was carved in stone from Aphrodisias, but proof will have to wait for future testing.

Two other gray stone sculptures in the Montemartini Museum, the “Bocchus Relief” of the Republican period, and a fragment of an Antonine statue with blowing drapery, have isotopic ratios that are not compatible with the known data for Aphrodisias.  Given the paucity of data on quarries of dark gray marble it is impossible at present to identify scientifically the sources for the material of these sculptures.









R. H. Tykot1, G. E. Borromeo2, K. Severson3



1 Department of Anthropology, University of South Florida, Tampa, FL 33620,

2 Museum of Art, Rhode Island School of Design, Providence, Rhode Island 02903 USA

3 74 The Fenway, #43, Boston, Massachusetts 02115 USA,







Eight marble sculptures from the Museum of Art, Rhode Island School of Design were analyzed in order to determine the source of the raw material used. Stable carbon and oxygen isotope mass spectrometry was used in conjunction with grain-size measurements and visual information to attribute each sample to possible and likely sources. The sculptures include a Headless Female Figure, two Male Torsos, a Torso of a Fighting Giant, the Pamphylian Sarcophagus, a Portrait of a Julio-Claudian, and a Male Figure in the Guise of Hermes. The Hermes actually consists of at least six parts and each was tested to determine whether marble from one or more quarries was used for this one sculpture. Similarly, both the lid and body of the Sarcophagus were also tested. Isotopic results for all samples were compared with published databases (Gorgoni et al. 2002; Herz 1987) for Mediterranean quarries, and the matches further narrowed b ased on grain-size measurements and visual analysis. Our results suggest that while the Hermes was made of marble from at least two if not three sources, both the body and lide of the Sarcophagus may be attributed to Afyon (Dokimeion). The other sculptures may be attributed to Afyon, Aphrodisias, Carrara, Paros, and Penteli.








R. H. Tykot1, M. Archambeault1, L. Becker2



1 Department of Anthropology, University of South Florida, Tampa, FL 33620,

2 Worcester Museum of Art, 55 Salisbury Street, Worcester, MA 01609 USA







Fifteen marble sculptures and fifty-five marble mosaic tesserae, mostly from the Roman site of Antioch in southeastern Turkey, were analyzed in order to determine the source of the raw materials used. Stable carbon and oxygen isotope mass spectrometry and x-ray diffraction analysis were used in conjunction with visual and other information to attribute each sample to possible and likely sources. Among the sculptural samples analyzed were two grave reliefs, two capitals, two sarcophagi, and figures of Asklepios, Aphrodite, Dionysus, harpocrates, a Private Citizen, and a Woman Holding a Jewel Box. Both the Asklepios and Aphrodite sculptures consist of multiple pieces and each was tested to determine whether they came from the same quarry. Isotopic results for all sculptural samples were compared with published databases (Gorgoni et al. 2002; Herz 1987) for Mediterranean quarries, and the matches further narrowed based on x-ray diffraction and visual analysis. Both pieces tested for the Asklepios and Aphrodite sculptures likely came from the same quarry. The mosaics were made of both calcitic and dolomitic marble coming from sources not known to have been used for sculptures. For several mosaics, it appears that all tesserae, both white and pink, came from the same source, while for at least one mosaic at least two sources were used. Further study will be necessary to identify the specific sources used for the tesserae.









M. De Angelis d'Ossat



Director Museo Nazionale Romano - Palazzo Altemps, Via S. Apollinare, 8 - 00186 Roma, ITALY

tel. 0039 6 6872414 - fax 0039 6 6897091,







The cameo found in new recent urban excavation in Palazzo Altemps was subject of research by Geological Departement of University of Roma "La Sapienza" in collaboration with Soprintendenza Archeologica di Roma.

Result is reported of stone provenance, comparated with similar other rare archaeological cameos.









E. Cantisani1, F. Fratini1, P. Pallecchi2, E. Pecchioni3, S. Rescic1.



1 Istituto per la Conservazione e Valorizzazione dei Beni Culturali -CNR sede di Firenze, Via degli Alfani, 74, 50121-Firenze, Italy

2 Soprintendenza Archeologica per la Toscana, Via della Pergola, 65, 50100-Firenze, Italy

3 Dipartimento di Scienze della Terra, Università di Firenze,Via la Pira ,4, 50121- Firenze, Italy







Among the many finds that came to light during the excavation of the ancient harbour of Pisa San Rossore there are numerous stones used for ballast and others for stowing the material transported by the ships. The ballast found in the excavation site is represented by large blocks, mainly rounded, of various sizes but with diameter not less than 10 cm. One sample was taken for each type of rock identified by preliminary macroscopic observation.

Classification and mineralogical-petrographical characterisation of the ballast stones give important information in the search for the possible areas of origin of the rocks themselves. In addition, information on the origin of the stones could shed light on the routes taken by the ships, although it cannot be ruled out that the materials may have been re-utilised. In order to classify the different lithotype we have performed geochemical, mineralogical and petrographical analyses.

We have analysed 21 samples of rocks coming from the areas of the archeological site defined as zone 3, zone 5 and “Southern Extension”. A point still to is determined whether the samples taken from rocks found during excavations of the “Southern Extension” also include stones pertaining to the harbour structures as well as to the ship’s ballast.

This research permits to recognise the different kinds of rocks, coming from geological areas with different characteristics. The mineralogical-petrographical analyses performed on the rock samples revealed the existence of markedly different lithotypes, certainly deriving from a very broad area. Rocks of sedimentary, magmatic and metamorphic origin were identified.

The stones that were sampled in zone 3 (ship B), are made predominantly of metamorphic rocks (marble, quartz, mica schist, schists and quartzite) while those belongings to zone 5 (ship D) are made of sedimentary rocks (sandstone, micritic and biosparitic limestones) and rocks of magmatic origin (basalts, and microgranitic porphyries). These preliminary observations suggest a probable different trade ship taken by the ship B respect to ship D.

Particularly interesting is the way of stowing used by Romans: the amphorae where settled on an incoherent groundmass incorporating blocks of stone, lumps of brick and plant rnatter, laid on the underbeck. The groundmass show a carbonatic composition and the framework is composed by magmatic rocks with pyroxenes badly preserved, leucite and carbonatic fragments.

This mineralogical association can be referred to a particular geological setting located in the Latium –Campania area (Italy).

For the stones used as ballast a caparison with the bibliographic data and suitable sampling of probably provenience area can be useful for a following search step.








N. Lombardini1, G. Tucci2, M. Iwade3



1Architect, Milano, Italy

2DINSE Politechnicodi Torino, Torino, Italy

 3 Institute of Art and Design, University of Tsukuba, 1-1, Tennodai, Tsukuba, Ibaraki, 305-8574, Japan,







The church of St. Vitale at Ravenna (Italy) is one of the most important buildings of the Empire of Justinian the great. (6th century). The building system, the mosaics of the presbytery and the apse, and the octagonal plan represent the characteristics of the grandeur of the epoch. The historic, archaeological and artistic value of the church is also given by the marble, still remaining inside the church. Greek and Turkish marbles were used for the 36 original columns distributed on the women's gallery and the ground floor, the veneering of the eight pillars sustaining the dome and the partial veneering of the six pillars sustaining the boundary wall.

The aim of this presentation is to report the first phase of the geometrical and material survey of the veneering marble of the eight internal pillars in order to produce the graphic support useful for the study and the characterisation of the different types of marbles and the causes of their decay.

Each pier is covered with more than 180 marble slabs, the surface of which amounts to approximately 62 square meters. The purpose of this research is obtaining a graphical documentation of these marbles as fundamental information for their conservation and maintenance. The following procedures were applied for this purpose.

1) Measurement of marble slabs of the 8 pillars with traditional instrument.

2) Photo shooting of all the marble slabs with the digital camera.

3) Drawing up the elevation of each pier applying the rectified digital photographs on the basis of the geometrical survey.

The result of the research is a set of accurate elevation drawings of the eight piers with coloured marble patterns and description of decay. This graphical documentation will greatly contribute to the future conservation and maintenance of this important heritage.









K. Knowles, F. Lewis, D. Peacock



University of Southampton, Department of Archaeology, Highfield, Southampton, SO17 1BF, UK,







The ‘Stone in Archaeology’ project is in the final year of its three-year development period. Funded by the Arts and Humanities Research Board, its purpose is to create a digital catalogue of all archaeological stone known to have been exploited in Antiquity throughout the British Isles.

The aim is to create a database centred on the extensive collection at the Department of Archaeology, University of Southampton. The digital resource would allow archaeologists to explore avenues of research that were often non-existent, or previously slow, difficult and expensive to implement.  In so doing, it is hoped that it will help realise the immense potential of the study of stone in archaeology.

The digital resource will incorporate a searchable relational database, which will be regularly updated, dynamic and available on the Internet.  It will hold information regarding rock type, quarry location, stone usage and bibliographic references. There will be facilities available to identify rocks, to view microscopic and macroscopic images of stone samples, and to examine the changing use of a particular stone over time. Furthermore, the resource will answer specific questions about trade and exchange, movement of materials and distribution of stone by incorporating a GIS element into the database.

By creating this resource, it is hoped that it will improve the breadth and depth of archaeological knowledge in the field of lithics and beyond. 








P. Storemyr 1, T. Heldal 2, E. Bloxam 3, J. A. Harrell 4



1 Expert-Center fuer Denkmalpflege, CH-8005 Zurich,

2 Geological Survey of Norway, N-7491 Trondheim,

3 Institute of Archaeology, University College London, London WC1H 0PY, UK,

4 The University of Toledo, Toledo, Ohio 43606-3390, USA, JHarrel@UTNet.UToledo.Edu







Basalt was an important material to the ancient Egyptians, and was used primarily for vessels in the Predynastic period and for architectural elements like floors and retaining walls at Old Kingdom pyramid mortuary temples. Moreover, a number of small sculptures are known from the Middle Kingdom and Graeco-Roman periods. Until recently the only known ancient Egyptian basalt quarry was located in the so-called Haddadin flow of Oligocene age at Widan el-Faras in the Northern Faiyum desert. From earlier investigations, it seems certain that this quarry was the main source of basalt for pyramid temple floors and, moreover, that the Haddadin basalt flow was the main source of Predynastic vessels.

The Widan el-Faras quarry was reopened by the Romans, as evidenced by pottery scatters and technical features related to the stone extraction. Whether the basalt used for Roman statuettes came from this location or from reuse of pyramid temple floor blocks is not yet known. It is also possible that it came from a recently discovered small Roman quarry in the El-Minya basalt flow at Tilal Sawda near El-Bahnasa (and the Graeco-Roman city of Oxyrhynchus) in Middle Egypt. This quarry was found during a survey of Egyptian basalt outcrops in 2002. The El-Minya basalt flow is also of Oligocene age, but has a much more fine-grained texture than the Haddadin basalt.

The surviving quarry workings at Tilal Sawda are modest. A main extraction site measuring some 10x15 m is located in the top flow of the basalt, on the side of a small hill in this undulating desert landscape rich in sand dunes. Quarrying was obviously undertaken with simple means, using wedges to split open existing columnar joints. Scatters of Late Roman to Early Islamic pottery can be found in the quarry area, as well as along the scree of waste from the quarrying operation. Other quarries have not been found in the vicinity, but extensive basalt outcrops to the east and closer to the Nile Valley, where ancient quarrying may have started, would have been destroyed by the modern quarrying operations currently taking place here. In the area with the surviving ancient workings, it seems that the top basalt flow, where there is reasonably good-quality stone, has been systematically tested. This is evidenced by tiny extraction sites on the nearby hills. Because it is easy to texturally distinguish El-Minya from Haddadin basalt, it should be theoretically possible to find out whether some Roman statuettes might originate from Tilal Sawda. This work is yet to be undertaken.








Jillian B Phillips



Dept. of Archaeology, University of Southampton, Avenue Campus, Southampton SO17 1BJ, United Kingdom or







Mons Porphyrites quarries are known as the source of the Imperial Porphyry much valued in the Roman and Byzantine worlds.  The priceless purple porphyry came from a series of quarries situated at the tops of mountains in the harsh remote environment of the Gebel Dokhan in the Red Sea Mountains of Egypt.  This unique series of quarries are part of an industrial landscape supported by a network of quarry villages, ancillary buildings, slipways, interconnecting footpaths and forts.   The remote location provides the perfect opportunity to study an area that had been visited only a handful of times since the Romans left in the 5th century. This paper shows the archaeological evidence of five known areas that were exploited and looks in detail at the quarry complexes at Lepsius, Lycabettus, North-West and Foot quarries to show the extent to which the extraction of porphyry was undertaken.  The logistics of the support and transportation are also considered.  The complexes were vastly different in size and were active mainly in the 1st century AD but continued production until the 4th possibly 5th century.  This paper presents an overview of the remarkable journey purple porphyry travelled from its humble beginnings in the Egyptian mountains to the Imperial Courts in Rome.








Ergün Laflı



Deparment of Archaeology of the Dokuz Eylül University, T.C. Dokuz Eylül Üniversitesi, Fen-Edebiyat Fakültesi, Arkeoloji Bölümü, Oda No: A 461/1, Tınaztepe Kampüsü, Kaynaklar Yerleşkesi, Buca, TR-35160, İzmir, Turkey.,







Kahramanmaraş Valley is located in southeastern Turkey, lies between mountain ranges and is very suitable for agriculture. Between 1993 and 1995, a survey was done in the valley by a team led by Dr. Elizabeth Carter of University of California at Los Angeles. About 250 sites were located in this valley ranging from the Neolithic to the Medieval/Islamic Period. The main object of these surveys was to understand the dynamics of settlements and settlement patterns (especially one possible determinant for urbanism in the valley) in earlier and classical ages. The first results of these surveys provided the evidence that beside the world metropolis Antioch the Kahramanmaraş valley had rather an isolated character in the classical antiquity.

At these surveys we were able to determine some sites, which were obviously used as stone quarry during the classical antiquity. The most important characteristics of them were that they were not depending to any greater settlement in their neighbourhood, and were used for the stone consumption of nearby greater cities, such as Germanica. Since our survey is aimed to be a regional project, we have observed various aspects of these specific sites with special purposes individually. As for the first step we tried to understand the dynamics and characteristics of these stone quarries. What we attempted was to understand how stone transportation was being implemented and in what sort of technical structure these quarries were working. For these purposes we have observed all stone samples of later ages in the region and tried to comprehend whether they were gathered from our local quarry sites. Our researches brought interesting patterns of stone transportation in rural environment of classical ages. What we also were planning was to look after the dynamics of stone consumption during the classical antiquity. Thus our work could be a sample survey for understanding the dynamics of stone quarries in southeastern Anatolia.







Georgina Bradbury



Allandalle Avenue Gien, Osmond 5064, (University of Adelaide), Australia







Most scholarly visitors and tourists who visit the Greco-Roman sites of Pamphylia and Pisidia overlook one important aspect; the stone used to build these cities. Many such cities on the plain relied upon local conglomerates to lay their foundations and plaster or marble revetment for decoration.  Termessos, lies on the border of Pamphylia, and Pisidia to the north of modern Antalya, and unlike cities of the plain could use stone of a higher quality in all construction. There are two types of building stone available at Termessos; a limestone, and a fine-grained, low metamorphic grade marble.  These two building materials provided the city with all construction needs, and the local quarries probably played an important role in the city’s history.

Lanckorownski (1892), who surveyed the city in the 1890’s, noted two small quarries, one on a path leading past the rock cut tombs before reaching the city walls, and the other behind the large necropolis beyond the necropoleis and the city itself. Recent observations, (Bradbury 2001), have revealed more extensive quarrying than previously recorded.  This paper describes newly discovered quarries near the necropoleis of the city.  The extent and methods of quarrying are also discussed. These are evidenced by considerable quantities of quarry debris and also by quarrying marks remaining on the quarry faces.  This paper describes the potential importance of the quarries at Termessos and their potential supply to the cities on the Plain of Pamphylia.







Garagunis K., Vgenopoulos A., Katsinis D., Oikonomidis L., Papadakis M. Pikopoulou-Tsolaki D.



National Technical University of Athens, 9, Heroon Politechniou Str., 15780 Zografou, Athens, Greece








 The “Via Egnatia” was one of the most famous roads in the years of Roman Empire. It was constructed for commercial and military reasons and thousands of pedestrians, soldiers, horses and various types of carriages used it every year. “Via Egnatia” had two starts one to Epidamnos (Today the biggest port of Albania) and the ancient city of Appolonia at the coast of Adriatica sea. The road passed through Macedonia and ended at Kypsela near the river Evros. Its construction was done between the years 146 BC and 120 BC. In 109 BC the Roman Emperor Trajanus managed to make “Via Egnatia” a most safe road for the transportation of military troops, various goods and travelers.

“Via Egnatia” was constructed from the ancient Greeks and later from the Romans according to 3 basic criteria: the morphology, the geology and the climatology. They used to construct roads in areas with smooth morphology in order to avoid great pitches that would cause problems to their transports. We were impressed when we visited “Via Egnatia”, the section between Salonica and Kavala, and we noticed that the road had been constructed in smooth morphology areas close to lakes and thermal water springs. So, they managed to develop the microclimates of the areas so as to control the temperature and to construct resting areas. Also, we noticed that in the parts of the road that were constructed on hard rocks, like granite, there were any or few ruinations. Contrary to that, in the parts of the road that were constructed on loosely rocks, like clay, there were corrosions that ruined the road. So, they filled the road with materials from the nearby areas. At the end, in the moderate strength rocks, like limestone and marbles, the corrosion was not as great as in the one in the loosely rocks.           

The present work is based in a graduation thesis completed in the National Technical University of Athens, School of Mining Engineering, Section of Geological Sciences.

The conclusions of the research done can be summarized in three parts. a) Characterization of the geological structures where the road is situated, b) Seismic activity of the area and c) Geological and petrographic recognition of the construction materials in the parts of the road that are still preserved.








Kyriaki Polikreti and Yannis Maniatis



Laboratory of Archaeometry, Institute of Materials Science, N.C.S.R. “Demokritos”, Aghia Paraskevi Attikis 153 10, Greece,,







The present work concentrates in the atomic scale of marble surface processes and contributes to understanding the behavior of carbonates in general, under weathering conditions.

The redistribution of ions (Mn2+ and Fe3+ substituting for Ca2+ in the CaCO3 lattice) observed on marble surface layers, was studied by Electron Paramagnetic Resonance spectroscopy on nineteen marble samples with different types of weathering and depositional layers. Samples of various grain sizes and manganese concentration, from quarry fronts and ancient monuments were used.

Mn2+ does not show a systematic behavior but Fe3+ (in the calcite lattice) decreases at the surface, up to 70 or 90%, compared to the marble bulk. The depletion starts at a depth at least 1 cm from the surface. Various treatments by heat, exposure to sunlight or boiling in clay-water solutions failed to reproduce the decrease in ferric ions. A possible explanation for the Fe3+ depletion is solid-state diffusion, probably aided by other lattice defects. The rate of the diffusion must be very slow and the ions may finally form iron-oxide crystallites on the surface, which may cause color alterations. The same phenomenon was not observed for Mn2+, maybe because it fits into the calcite lattice more easily than iron due to its smaller charge and ionic radius closer to that of Ca2+.

Another alteration observed in the EPR spectra of surface samples is the decrease of the peaks at g=2.0037 and 2.0044 attributed to lattice defects. The peaks were shown to diminish after one-month exposure to the sunlight. The phenomenon is probably a result of charge transfer towards or from the surface to the interior and can be connected to phenomena of thermoluminescence “bleaching”.

Except for the information given by the previous results on weathering mechanisms and effects in the marble lattice, possible implications to provenance investigation are discussed. 





Sourcing the Stone Tools and Vessels from the North Sinai Survey Collection



Joan S. Schneider1, Eliezer Oren2 and Miko Gabay2,


1University of California, Riverside, USA

2Ben Gurion University of the Negev, Israel







Over one thousand archaeological sites were recorded in the course of the North Sinai Survey project (Oren 1993).  Processing tools and vessels of a variety of stone materials, as well as fragments of items, numbering well over 1,476,were collected from 31 of these sites -- sites attributed to the Chalcolithic, Bronze Age, Iron Age, Hellenistic, Roman, and Persian periods.  A sample type collection was studied in-depth.  One aspect of the study was determining the probable source of stone for these objects, especially important because stone is not available within the survey area -- the northernmost strip of the Sinai Peninsula, bordering on the Mediterranean. In addition to the examination of hand samples, thin-section petrography and potassium argon dating methods aided in the identification of materials and determining possible sources.  The stones ranged from those of common varieties available in neighboring areas to rare and exotic materials available only in specific outcrops.  The results of the study indicate that stone was transported to the Northern Sinai, either in raw or finished form, from southern Europe, the northern Galilee, the Aswan area of Egypt, and other widely dispersed regions. Moreover, hypotheses regarding methods of transportation are considered.





Ancient marbles from monumental structures in the Circus Flaminius in Rome AND plastered TRAVERTINE ARCHITECTONICAL elements from the Temples of Bellona and apollo sosianus



M. De Nuccio



Sovraintendenza BB.CC. del Comune di Roma – Via del Portico d’Ottavia n. 29 – 00186  ROMA







One of the most important archaeological areas in Rome is the one of the Theatre of Marcellus, which includes the ruins of the Porticus of Octavia, the Theatre of Marcellus, and the Temples of Apollo Sosianus and Bellona, surrounded by a large porticus. During the excavations carried out in the 1930s, in fascist period, the archaeological structures were cleared of the houses built throughout the centuries.  At the time the medieval quarter was demolished to expose all Roman remains, especially those of the Augustean period

A joint research project of the Università “la Sapienza” and the Sovraintendenza examines the development of the ancient, medieval and modern history of the area. The study is focused on the temples of Apollo Sosianus and Bellona and their porticus, based on a preliminary analysis of all pertinent architectural and sculptural finds. A catalogue of about 40-50.000 of these fragments has been set up as an indispensable tool for understanding the sequence of construction phases from the republican to the medieval period.

These materials have been stored in the arcades of the Theatre of Marcellus: considering their high number and precarious state of preservation, efforts are being made to reach optimal preservation conditions.

The temple of Bellona is a peripteral, hexastyle temple with 11 columns on the sides, set  on a high podium with a staircase on the front.   A series of marble and plastered travertine architectural   elements, with similar decorative pattern, have been attributed to the building.  Like the nearby Temple of Apollo Sosianus, the Temple of Bellona was also built using two different materials: white marble on the front and travertine for the sides.

Here we present the restoration and the analysis of the architectural fragments of the cornice and fluted travertine drums of the external colonnades.  All fragments feature remains of a thick layer of painted plaster in yellow and blue.








Athanassios Nakassis



62, Neapoleos st., Maroussi 151 23, e-mail:

Phone-Fax: 2106847578. Cell phone: 6932448467







A problem for buildings in all periods is the effect of earthquakes. Attempts at the systematic examination of earthquakes, free from approaches relied on the supernatural, had already begun in historical times among the Greeks of Ionia and southern Italy.

The selection of a location with less seismic activity for settlement existed simply as a result of experience and precaution.

The basic construction materials used up until the last third of the 7th century BC were wood and clay. Wood was more appropriate than clay and even more so than stone to absorb damage which could come from earthquakes. Wood allows the building to sway, because of its properties, such as its flexibility, elasticity, and its ability to withstand back-and-forth movement. Of course, all scholars agree to the benefit of wood in the case of an earthquake. Wood was used for reasons of general stability. In the North Europe, and elsewhere, there was a wide use of wood in the absence of violent earthquakes. The same conclusion, of course, holds for the historical period. There is no proof that the known wood-joined constructions were used for anti-seismic reasons.

The width, reinforcement and assembling of corners, piers, doors, windows and other delicate areas were done for general reasons, and they show static knowledge.

In all periods, high-quality architecture and construction is considered that which the ratio of the useful to the built space is maximized so far as possible. Consequently, we might define consciously anti-seismic preparation as constructions that remain standing over time without excessive strengthening of their constituent parts. This definition excludes buildings that are strengthened after the fact and after observed damage. Such improvements are an integral part of the ongoing conservation of a building in use, and do not indicate any special knowledge either by the original architect or the subsequent builders.

Seismic activity is not a predicable phenomenon, thus it did not coincide the first priority in builders’ mind. Ancient builders cared more in static problems and in “everyday” weather phenomenon. The Roman architect Vitruvius, although he refers at length to construction techniques does not mention anything about seismic precaution.









S. Chlouveraki1,  S. Lugli2



1 INSTAP Study Center For East Crete, Pachia Ammos, 72 200, Greece

2 Department of Earth Sciences, University of Modena and Reggio Emilia, Largo S. Eufremia 19, I-41100 Modena, Italy







The aesthetic properties of gypsum, its availability, and the fact that it can be easily quarried, cut and shaped, were the main reasons for its extensive use in Minoan “Palatial” or “Elite” architecture, where large quantities of cut stone were required. A piece-by-piece record of four thousand gypsum architectural members, from the major Minoan sites, reveals a striking range in the variety and function of the gypsum rocks.

The main function of gypsum in Minoan architecture is a decorative one. It was principally used as an ornamental stone, in the form of thin paving slabs and revetments, which were applied as facing to rubble structures such as walls and benches. Other functions include, floor slabs, staircase treads, balustrades, door frames, pillars, wall piers.

Gypsum rocks show a great diversity in crystal morphology even within the same outcrop. One of the most common varieties is ‘selenite’, which consists of large transparent crystals that form a very attractive reflective surface. Selenite is used in a great extend at Knossos, Myrtos Pyrgos and the Megaron Nirou. Fine-grained varieties, which show very different macroscopic characteristics and aesthetic value, occur at Knossos, Megaron Nirou, Phaistos, and Agia Triada. These can be found in light hues but are often darkened by iron oxides/hydroxides, argillaceous material, or other minerals, as for example the purple, orange or brown gypsum rocks of Phaistos and Agia Triada.

The most popular but also controversial form of gypsum is the fine-grained nodular variety, also termed ‘alabaster’, particularly in the snow white or light coloured varieties. Although alabaster is usually thought as the most widely used variety of gypsum at the Palace of Knossos, petrographic analysis has shown that it has been used in very few rooms and certainly not at the West Façade or any other outdoors elements, which are usually made of coarse grained selenite. The white crystalline gypsum of the orthostates of the West Façade is a result of alteration of primary selenitic crystals due to dehydration caused by fire.

The petrographic analysis of a series of samples from all sites has allowed the detailed classification of the varieties used, and has shown that the current terminology used for the gypsum remains of the Minoan palaces, differs considerably from the geological one. It has also indicated that, because of fire damages and wheathering, the original appearance of some varieties was far different from what we see today.









M. Mariottini1, E. Curti2 and E. Moscetti3




1 Museo Litomineralogico - APAT, Via Curtatone 3, 00185 Roma, Italy

2 Museo di Scienze Naturali, Gruppo Speleologico di Guidonia - Montecelio,

Via Douet 2, 00012 Italy

3 Inspector of the Soprintendenza Archeologica per il territorio del Comune di

Guidonia-Montecelio, 00012 Italy







There are witnesses of the great amount of marble brought during the centuries to Rome, which has been already testified by the latin authors, not only in the ancient Rome but also on the outskirts of the town, i.e. in some patrician villas which still keep the remains of their ancient magnificence.

         Often, in the surroundings of Rome, agricultural activities and roadworks brought to light even the most insignificant traces of those remains whose preservation is still an unsolved problem.

         The purpose of the authors of this paper is to report, even through a simple cataloging activity based upon macroscopic, mesoscopic and textural observation, the splendor of these coloured marbles which sometimes assumed a particular political and ideological importance owing to their finding place or to their lithotypes.

         This paper intends to give a useful contribution to the knowing and to the comparison of the coloured stones used in the Roman Age in order to decorate the villas built in that part of the Rome suburb between the Nomentana and the Tiburtina roads.






Yannis Maniatis
ASMOSIA VII Conference
Laboratory of Archaeometry
Institute of Materials Science
N.C.S.R. Demokritos
Aghia Paraskevi
15310 Attiki
tel. + 30 1 6503389
fax + 30 1 6519430