² Robertson, R. H. S., Fuller's Earth: A History of Calcium Montmorillonite (London, 1986Google Scholar). At present it is Melos rather than Kimolos which is one of the world's major producers of this clay, known as environmental clay because of its application as a liner to landfill sites, in filters for water-supply, and as a drilling mud in hydrocarbon exploration and production. For a full review of Melos' currently exploitable resources see M. G. Stamatakis, U. Lutat, M. Regueiro, and J. P. Calvo, ‘Milos, the mineral island’. Industrial Minerals (February 1996), 57–61.

³ Caley, E. R. and Richards, J. F. C., Theophrastus on Stones: Commentary (Ohio State University, Graduate School Monograph 1: Columbus, 1956), 177–8Google Scholar; Robertson (n. 2), 37–40.

⁴ Shipley, G., A History of Samos 800–188 BC (Oxford, 1987Google Scholar) appendix 2; Theophrastus, De lapidibus 62–4; Dioscorides, , De materia medica v. 158Google Scholar. Lemnian and Samian earths are currently the subject of a separate archaeological/geological investigation by two of the present authors in association with the Institute of Geological and Mining Exploration (IGME), Athens. It should be noted that archaeology does not attest the existence of ‘Samian ware’ or ‘Samian Sigillata’, although Samian Earth may have been stamped for sale as a remedy for ailments during the Ottoman period.

⁵ Pliny, , Natural History xxxv. 191Google Scholar.

⁶ Ibid. 192; Dioscurides, , De materia medica v. 170Google Scholar; Eretrian earth has previously been identified as magnesite given that the island of Euboea is well known for its major deposits of this mineral; see Bailey, K. C., The Elder Pliny's Chapters on Chemical Subjects (London, 1960Google Scholar).

⁷ Theophrastus, De lapidibus 52–3; E. Photos-Jones, A. Cottier, Hall, A. J., and Mendoni, L. G., ‘Kean miltos: the well-known iron oxides ot Antiquity’, BSA 92 (1997), 359–71Google Scholar. The fineness (particle size less than 10 μm) of Kean iron oxides from the mines of Orkos and the site of Trypospilics have been shown to match those of finely processed modern iron oxides, thus corroborating Theophrastus' statement that Kean miltos was the best.

⁸ Cordelas, A., Ή ῾Ελλὰς ἐξεταϚομένη γεωλογικῶϚ καὶὀρυτολογικῶϚ (Athens. 1878Google Scholar).

⁹ Singer, C., The Earliest Chemical Industry: An Essay in the Historical Relations of Economics and Technology Illustrated from the Alum Trade (London, 1948Google Scholar).

¹⁰ Pliny, , Natural History xxxv. 199Google Scholar.

¹¹ Stephanides, G., The Mineralogy of Theophrastus (Athens, 1896Google Scholar).

¹² For an attempt to clarify some of the confusion see Pittinger, J.. ‘The mineral products of Melos in antiquity and their identification’. BSA 70 (1975), 141–8Google Scholar: Sparks, B., ‘Production and exchange in the Glassieal and Roman periods’, in Renfrew, G. and Wagstaff, M. (eds), An Island Polity: The Archaeology of Exploitation in Melos (Cambridge, 1982), 228–35Google Scholar.

¹³ E. Photos-Jones, ‘The Aghia Kyriaki Melos Survey: industrial minerals in the Aegean in the Roman period’, paper presented at the opening of the Melos Mining Museum. Voudia, Melos, May 28th 1998 (a copy of the paper is at the Melos Mining Museum): Photos-Jones, E., Hall, A. J., Atkinson, J. A., Tompsett, G., Cottier, A., and McRobb, A.. ‘The elusive earths: prospecting for Industrial Minerals in the Aegean in Roman Aghia Kyriaki’, unpublished report prepared for the Melos Mining Museum, September 1998Google Scholar.

¹⁴ Ross, L.. Reisen auf den griechischen Inseln des ägäischen Meeres, iii (Stuttgart and Tübingen, 1845)Google Scholar: I. Chatzidakis. Ήίστορία τῆςνήσου Μήλου, ed. Z. A. Vaos Athens, 1972: first pub. 1927.

¹⁵ Mackenzie, D., ‘Ancient sites in Melos’, BSA 3 (1897), 71–88Google Scholar; he noted (p. 80): ‘Haghia Kyriaki and Palaeochori present exactly the same features as Komia, and belong to the same late Roman era. They are emporia for the Roman exploitation of the mineral wealth of East and South Melos, and we know that the mill-stone quarries of Melos lie in the hill region north-east of Palaeochori.’

¹⁶ J. F. Cherry, ‘Register of archaeological sites on Melos’, in Renfrew and Wagstaff (n. 12), 304, site 78.

¹⁷ I. Triandi (pers. comm.). These trenches concentrate in areas Z, X, F, I, AA: they have been incorporated within tile present topographical map.

¹⁸ I. Triandi (pers. comm.). Pottery from the site recovered in the course of rescue excavations between 1985 and 1989 is currently being studied, as part of a doctoral dissertation, by S. Raptopoulos.

¹⁹ The radiocarbon dates published by Traineau, H. and Dalabakis, P., ‘Mise en évidence d'une éruption phréatique historique sur l'île de Milos (Grèce)’, C.-R. Acad. Sci. Paris. 308 (1989), 24Google Scholar are given as 200BC–AD 200: no σ values are reported.

²⁰ B. A. Sparkes, in Renfrew and Wagstaff (n. 12).

²¹ Kakavoyiannis, E., ‘Μιά νέα άποψη για την λειτοργίατων πλυντηρίων μεταλλεύματος της Λαυρεωτικής κατάτους κλασσικούς χρόνους’. Proc. 1st Greek Archaeometry Society Symposium (Athens, 1992), 79–93Google Scholar.

²² Raptopoulos, S., ‘“… Duae et ejus species”. Ηστυπτηρία γη της Μρόνους’, paper presented at the First International Congress on Ancient Greek Technology, Thessaloniki, 4–7 September 1997Google Scholar. Melanterite, FeSO₄.₇H₂O, also called atramentum sutorium by Pliny and copperas by Agricola, may also contain Mg, Mn, Zn, or Cu in place of Fe. Apjohnite is manganese alum, MnAl₂(SO₄)⁴.₂₂H₂), is a member of the halotrichite group: it may contain Mg instead of Mn.

²³ AR 1989–90, 67: A. Delt. 36. Chr. 381.

²⁴ Numbers in parentheses are those given in Cherry (n. 16): Palaiochori (63) is situated on the adjacent bay to the NE: the valley is flatter than Aghia Kyriaki but is also dissected, albeit to a lesser extent, by erosion gullies. There are kaolin quarries at the head of the valley and offshore thermal springs. Amphorae and lckanai were reported, some with stamps, albeit worn out, Soleta (56–7) was discovered through the villa noted by Pittinger (n. 6) at Agios Pandcleimon; it may be the same as the Early Roman villa at the west end of one of the higher terraces. The site is extensive and stretehes along the terraces at the upper reaches oi the coastal plateau to the east of Palaiochori. A deep gully on the west side runs down to the sea to the east of Cape Spathi. Vitrified pottery wasters bloated and warped attest to pottery manufacture on site. There are Early Roman amphorae and basins with impressions COLO, G HN and also considerable quantities of Late Roman pottery: Asprokavo (96) is a small bay within the bay of Melos. It was formerly a coastal lagoon with a beach ridge cutting off the sea water lagoon. A 3 m high section in the dune has revealed pottery, walls, and floors of Early Roman structures including lekanai and amphorae. There is also a Late Roman site with domestic pottery and glass on the headland to the north, presently damaged by a bulldozer track. There are amphorae and lekane fragments as well as wasters. Emporio (95), a nonsite as described, is actually three sites on the edge of the plain with a sparse scatter of pottery with Early Roman chambered tombs in the plam itself On the east side there is Late Roman and Early Byzantine pottery. The site on the hills immediately to the south is Archaic and Frankish (13th-c.). At the west end at the ridge between Emporio and Asprokavo there is Early Roman pottery including sherds of amphorae and lekanai. The offshore remains are claimed to be medieval. Warped, bloated vitreous wasters point to pottery manufacture. At Kanava (24) there are pottery wasters and at Ambourdektakhi (97) and Achivadolimni there are amphorae and basins. At Plathiena (8) there are amphora and lekane fragments: at Tria Pigadhia (42). Kato Komia (43), and Sta Glastria 107 amphorae and basins. See HG. 1.

²⁵ Dakoronia, Ph., ‘Earthquakes of the Late Helladic III Period (12th century BC), at Kynos (Livanates, Central Greece)’, in Stiros, S. and Jones, R. E. (eds). Archaeoseismology (Fitch Laboratory Occasional Paper. 7: Athens, 1996), 41–4 at 43Google Scholar fig. 4.

²⁶ I. Trianti (pers. comm.).

²⁷ Fytikas, M., Geological Map of Greece: Melos Island (IGME: Athens, 1977Google Scholar): P. Shelford, ‘The geology of Miles’, in Renfrew and Wagstaff n. 121. 74–81.

²⁸ McKenzie, D. P., ‘Plate tectonics of the Mediterranean region’, Nature, 226 (1970), 239–43Google Scholar; Robertson, A. H. F. and Dixon, J. E.. ‘Introduction: aspects of the geological evolution of the Eastern Mediterranean’, in eid. (eds), The Geological Evolution of the Eastern Mediterranean (Oxford, 1984), 1–74Google Scholar: M. Fytikas, F. Innocenti, P. Manetti, R. Mazzuoli, A. Peccerillo, and L. Villari. ‘Tertiary to Quaternary evolution of volcanism in the Aegean region’, ibid. 687–99.

²⁹ A. J. Hall, E. Photos-Jones, A. Cottier, D. Turner, and A. McRobb. ‘The geological and geothermal setting of the Roman site at Aghia Kyriaki, Melos, Greece’, paper submitted for publication to the Journal of the Geological Society of London.

³⁰ Fytikas, M., ‘Updating of the geological and geothermal research on Milos Island’. Geothermics, 18 (1989), 485–96Google Scholar; M. Fytikas, J. D. Garnish, V. R. S. Hutton, E. Staroste, and J. Wohlenberg. ‘An integrated model for the geothermal field of Milos from geophysical experiments’, ibid. 611–21; Dando, P. R., Hughes, J. A., Leahy, Y., Niven, S. J., Taylor, L. J., and Smith, C.. ‘Gas venting rates from submarine hydrothermal areas around the island of Milos. Hellenic Volcanic Arc’. Continental Shelf Research. 15 (1995), 913–29Google Scholar.

³¹ Sulphur isotope work on natural and experimentally produced sulphur is now in the process of being investigated. In the last century sulphur was widely mined on Melos, principally m underground mines in the E and SE parts of the island (Theiorycheia. Tria Pigadhia, Paliorema, Firlingo, and Kalamos). Modern production briefly boomed between 1890 and 1905 with the discovery of its efficacy in treating vines alfected by phylloxera when some 300 labourers were employed (Wagstaff m Renfrew and Wagstaff (n. 12). 241) in extracting 125.000 tons of raw material with an average sulphur content of 14% (Stamatakis et al. (n. 2). 61). Formerly it was probably extracted in its pure crystalline form only where the mineral formed on the surface ot rock exposures and in fissures. It is possible that sulphur-rich matrices such as gypsum and clay were processed in kilns to extract the sulphur. Sulphur crystals melt at 113°C: in the process the liquid sulphur would trickle to the bottom of the kiln, from which it could be removed. Pale yellow liquid sulphur becomes increasingly viscous up to 180°C, at which point it acquires a red huc and thereafter decreases in viscosity. In air sulphur ignites and burns with a blue flame to produce sulphur dioxide at 363°C. The liquid could be poured into moulds for transportation or, in its viscous red state, turned into water to its (temporarily) plastic form. Sulphuric acid (oil of vitriol) produced by heating ferrous sulphate and alum exists naturally in water in volcanic regions (Singer, G., Holmyard, E. J., Hall, A. R., and Williams, F. (eds). A History of Technology). iii: From the Renaissance to the Industrial Revolution (Oxford, 1956Google Scholar)): as a medicine, the concentrated acid could be applied to venereal sores and warts and presumably its more dilute form in medicinal baths was moderately efficacious.

³² See Fytikas (n. 27).

³³ Fytikas, M., Innocenti, F., Kolios, N., Manetti, P., Mazzuoli, R., Poli, G., Rita, F., and Villari, L., ‘Volcanology and petrology of volcanic products from the island of Milos and neighbouring islets’. Journal of Volcanology and Geothermal Research, 28 (1986), 297–317CrossRefGoogle Scholar.

³⁴ See n. 29.

³⁵ Ibid. Soine hematite was probably also produced more conventionally by oxidation of iron-bearing minerals such as iron-rich illite/smectite and chlorite. The key reactants and products are represented in the (unbalanced) equations:

Fe³⁺ (sulphate) + H₂O→Fe₂O₃ + H₂SO₄

FeO(silicate + O₂→FE₂O₃ + Fe-free silicate.

³⁶ D. Davidson and C. Tasker, ‘Geomorphic evolution during the late Holocene’, in Renfrew and Wagstaff (n. 12). 82–94.

³⁷ Ibid.

³⁸ See n. 29.

³⁹ See n. 9: Singer et al. (n. 31), 242.

⁴⁰ Pliny, , Natural History xxxv. 37Google Scholar.

⁴¹ Theophrastns, De lapidibus 62: Vitruvius, De architectura vii. 7. 3: Plutarch. De defectu orraculorum 436 C.

⁴² Hippocrates, Steril. 225: Ulc. 11. 18.

⁴³ Levidis, A., Πλίνιος ο Πρεσβύτερος, Περὶ της Ελληνικής Ζωγραϕικής, 35. Βιβλίο της Φυσικής Ιστορίας (Athens. 1994)Google Scholar.

⁴⁴ Dioscorides, , Mat. med. v. 180Google Scholar.

⁴⁵ Levidis (n. 43), 210.

⁴⁶ Caley and Richards (n. 3). 108.

⁴⁷ Theophrastus de lapidibus. ed. Eichholz, D. E. (Oxford, 1963), 129Google Scholar.

⁴⁸ Pliny, , Natural History xxxv. 183–91Google Scholar: Bailey (n. 5), ii. 233, and the Encyclopaedia Britannica, 11th edn., both observe that Pliny's, description of alumen referred to several distinct minerals (‘plura et cuts genera’, xxxv. 183Google Scholar), principally iron sulphates including aluminium sulphate, potash alum, kalinite, and perhaps halotrichites. Iron sulphates. being strong astringents that can cause constipation, were used for treating diarrhoea and dysentry. When applied to broken skin ferric salts are powerful astringents coagulating blood, therefore important as local haemostatics and styptics.

⁴⁹ Pliny, , Natural History xxxv. 184. 188Google Scholar.

⁵⁰ Ibid. 185.

⁵¹ Ibid. xxxiii. 63.

⁵² Ibid. 88.

⁵³ Ibid. xxxv. 183.

⁵⁴ Ibid. 188.

⁵⁵ Claudius Quadrigarius fr. 81.

⁵⁶ Cottier, A., ‘Industrial Minerals in the Aegean: Melos in the Classical and Roman Periods’. Ph.D. thesis, University of Glasgow (in progress). Raftopoulos (n. 22Google Scholar) aptly showed that a closer look needs to be paid to the differenee between Pliny's species and genus, translated as ‘kind’ and ‘variety’ in the Loeb edition. Pliny labels one kind of alum as Melian earth, an observation which has inevitably led to much confusion.

⁵⁷ Pliny, , Natural History xxxv. 174Google Scholar.

⁵⁸ Ibid. 175–7.

⁵⁹ Ibid. 175.

⁶⁰ Agricola, G., De re metallica libri XII (Basle, 1556), book iii. p. 58Google Scholar.

⁶¹ Sparkes (n. 20), 233–4.

⁶² Keller, D. R. and Rupp, D. W. (eds.), Archaeological Survey in the Mediterranean Area (BAR S155: Oxford, 1983Google Scholar).

⁶³ Greene, K., The Archaeology of the Roman Economy (London, 1992Google Scholar).

⁶⁴ Photos-Jones, E. and Jones, J. Ellis. ‘The building materials and industrial remains at Agrileza. Laurion fourth century BC and their contribution to the workings at the site’. BSA 89 (1994), 307–64Google Scholar.

⁶⁵ Trianti (pers. comm.).

⁶⁶ See Traineau and Dalabakis (n. 19).

⁶⁷ A chronicler of 1650 in D. Vallianou. ‘New evidence of earthquake desetructions in Late Minoan Crete’, in Stiros and Jones (n. 25), 153-67, at 160.