^{page 46 note 1} Εφ. ᾿Αρχ 1894, p. 266 sqq.

^{page 46 note 2} Conze, Melische Thongefässe, Vase A.

^{page 47 note 1} Πρακτικὰ τῆς ᾿Αρχαιολογικῆς ῾Εταιρείας 1898, p. 100.

^{page 47 note 2} The modern Megále Delos.

^{page 47 note 3} No trace of the earlier purification of the island by Peisistratos (Her. I. 64) has been found, though the present find includes vase-fragments going back well beyond the middle of the sixth century. The whole question of the find in its relation to the literary authority, and the light it throws upon the system of purifications in Greece, remains to be discussed by Mr. Stavropoulos.

^{page 48 note 1} The aim of the present account is simply to call attention to the find, since the publication of Mr. Stavropoulos' final account of his work cannot be expected within a year or two. Many boxes of fragments still remain to be sorted, and after the final sorting there will still be much time required for putting the vases together, no light task for a man working with a single assistant. Whatever value the present description may possess is entirely due to Mr. Stavropoulos, who has exercised the most scrupulous care in sorting the fragments and gave the most generous aid in the examination of them. He is of course in no way responsible for the writer's mistakes or omissions.

^{page 48 note 2} The detailed description of these fabrics lies outside the scope of the present paper. The whole find is of extreme interest, not only for the presence of new fabrics but perhaps even more so for the unexpected absence of so many known fabrics. Only one Mykenaean vase occurs, and that of poor and late style. ‘Samian’ ware is represented by six amphorae, all of the same type; and ‘Rhodian’ by twenty amphorae (these also all of the same type) and a few plates.

^{page 49 note 1} The sketch here given represents no actually existing vase. It is a conjectural restoration based upon a number of separate fragments and is intended only to show the arrangement of the friezes and the general appearance of the vases.

In the case of this and of several other figures in the text it seemed best to give a mere transcript of the original sketch, since the rougher drawing is often more faithful to the original. The spiral designs and other cases of finer freehand work were kindly redrawn for me in Athens by Mr. Baker-Penoyre.

^{page 50 note 1} The description of the decoration here given must, in the present incomplete state of the sorting of the vase-fragments, be considered as tentative: nor has it been possible to enumerate all exceptional instances.

^{page 51 note 1} This upper shoulder-band seems to be copied from an actual necklace. The ornament forms used occur not only on the Heads painted on the vase-necks but also on certain early female statues from Delos.

^{page 52 note 1} This use of a single Eye under each handle has a parallel on the Theran hydriae already mentioned (p. 48).

^{page 52 note 2} Most regularly on a type of Attic black-figure amphorae (cf. Thiersch Tyrrhenisdic Amphoren, Pl. II). More isolated instances are the Myrina vase (B.C.H. 1884, Pl. vii.) and the Ionic amphora in Berlin No. 1674. An instance of a female head is on an amphora from Daphnae (Tell-Defennch, Pl. XXXI). Less directly analogous instances occur on Corinthian, Phalerou and Mykenaean vases and on Ionic Eye-kylikes.

^{page 54 note 1} These heads have an interesting analogy in that of the so-called Nike of Archermos from Delos. It is perhaps dangerous to lay much emphasis upon the resemblance in the shape of skull and profile, but the coiffure and dress at least afford safe ground for argument. The hair is arranged in the same way both over the temples and at the back of the head. The necklace is of the same form in both. The Nike has nothing but the upper rosette of the earrings still intact in the stone but the metal pendants are represented by dowel holes. She wears a similar short-sleeved chiton and the little incised circles that cover it may well be the guiding lines for a painted scale-pattern as on the chitons worn by the Head figures on the vases.

^{page 57 note 1} One form of rosette that is used in this way seems to be peculiar to the style. The whole centre of the rosette is reserved in the slip, and the black surface around it is again divided off into sections by narrow reserved lines.

^{page 58 note 1} I have picked up sherds of micaceous red clay as far apart as Aegina, Syra, Pergamon and Samos. The collection of modern Greek pottery in the Sèvres Museum well shows the wide extension of such clay at the present day. How far beds of similar clay may exist at different localities in the Aegean and how far there was and is one chief centre of export I do not know.

^{page 59 note 1} Cf. Riegl, , Stilfragen, pp. 166Google Scholarsqq.

^{page 59 note 2} The Theran hydriae at Mykonos constantly employ a band of trellis-work pattern, but it is of so simple a character that it can hardly be called distinctively Geometric.

^{page 60 note 1} Boehlau, , Jahrbuch 1888, p. 349Google Scholar.

^{page 60 note 2} A.J.A., 1900, p. 65.

^{page 61 note 1} It was this apparent uniformity of the Geometric style that gave plausibility to the theory that the Geometric ware came into Greece with the Dorian invaders as a fully developed style. But that theory is really inconsistent both with the character of the Dorian invasion and with the topographical distribution of Geometric ware in the Aegean. The same superficial uniformity of the style gave countenance also to the false use of the name ‘Dipylon’ to denote all Geometric ware in whatever locality it might be found. The error of that use of the name has already been pointed out by Dr. Wide and others in discussing the local variations of Geometric patterns: but the late Geometric hydria recently found at Kavousi, (A.J.A., 1901Google Scholar, Pls. III. and IV.) shows clearly that there is local variation not only in ornament forms but also in the technique of figure-drawing.

^{page 61 note 2} Besides the vases from Rhodes now in the Berlin Museum, there are other vases of the same style still remaining in private pcssession in Rhodes itself.

^{page 61 note 3} These graves contained nothing but the small, rudely decorated vases and a series of little sickle-shaped iron instruments. Mr. Stavropoulos suggests their connection with the Carian Islanders (Thuk. I. 8); but the discussion of the question must await the full publication of his researches in the Cyclades.

^{page 62 note 1} Athen. Mitthcil. XII. 227. Cf. J.H.S. XVI., p. 265.

^{page 62 note 2} B.C.H., 1884, p. 509.

^{page 62 note 3} Boehlau, , Aus Ion. und Ital. Nekr., p. 86Google Scholar.

^{page 63 note 1} Joubin, , B.C.H., 1895, p. 69Google Scholarsqq.

^{page 63 note 2} Athen. Mittheil, 1890, p. 318 sqq.: cf. the vases from the Marathon tumulus (A.M. 1893, taf. II. and III.) and from Menidi, (Jahrbuch, 1899, p. 107Google Scholarsqq.). Vases of this style occur at Delos, among the Akropolis fragments, at Eleusis and on other Attic sites. It is represented in many European museums.

^{page 63 note 3} E.g. Pottier, , Catalogue des Vasis du Louvre, II. p. 433Google Scholar.

^{page 63 note 4} This little detail is also of value in directly refuting the suggestion that the Vourva style is in any way an imitation of the Corinthian. The two styles are really quite distinct. The Vourva style came from Ionia to Attica not by way of Corinth but from the North-East. Its connexion with Eretria has already been pointed out (Boehlau, , Aus Ion. und Ital. Nekr. p. 116Google Scholar); and this connexion has recently been confirmed by further finds of Eretrian vases. Undoubtedly at times there is a certain measure of fusion between the Corinthian and the Vourva styles, but in essence the Vourva ware represents a wholly distinct influence, the influence upon Attica of the Ionian metal work, transmitted not directly up the Saronic Gulf but through the Euboean cities and the Eastern demes of Attica.

^{page 64 note 1} This class was first collected by Thierseh, (Tyrrhenische Amphoren, p. 146Google Scholar). He suggests that it may be Boeotian. The objections to that view are given in the text.

^{page 64 note 2} Besides the finds at Delos and Eleusis, complete vases were found at Vourva and a few fragments on the Kynosarges site by the Ilissus and in the Akropolis excavations.

^{page 64 note 3} No reference has been here made to the great finds of vases at Naukratis and Rhodes; for, although some of the minor fabrics there fouud were no doubt manufactured on the Ionian coast, there is no important fabric for which an Ionian origin is assured. The evidence drawn from Rhodes and Naukratis could never do more than confirm the existence of an early school of Ionic pottery, and until vases have been found in considerable numbers on some site in Ionia itself such evidence is inadmissible.

^{page 64 note 4} E.g. Pottier, , Catalogue des Vases du Louvre. I. p. 327Google Scholar, 377.

^{page 66 note 1} Annual of the B.S.A. Vol. IV. p. 37 sqq. No attempt has been made to assign any of the foreign fabrics found at Phylakopi to their proper localities. It is sufficient for the present argument if the unbroken series of native ware be taken as typical for the whole Cycladie area.

^{page 67 note 1} It does not fall within the scope of the present paper to trace out the influence of the Delian style upon contemporary and later fabrics. But what has been said thetein in, respect to ‘Ionic’ influence carries with it the position that much of the so-called Orientalizing influence upon the mainland fabrics—that influence which, e.g., developed the Phaleron and Early Attic styles out of the Dipylon—must be accredited to Delos.

^{page 68 note 1} Arch. Zeit. 1854. Pl. LXI.

^{page 68 note 2} Conze, A., Melische Thongefässe, Leipsic, 1862Google Scholar. These vases, lettered A, B, C, by Conze, are now in the Ethnic Museum at Athens, where they are numbered 911, 912, 913. They were unearthed in Athens at the time of their publication, when two of them were found in the Royal Palace, the third at the house of one of the ministers. All three however were said to have come from Melos. Vase A is the most important of these, and is referred to in the text as the ‘Apollo and Artemis’ vase, being that which has for its main figure-subject a representation of Apollo and three Muses in a chariot confronting Artemis who holds a stag by the horns. Those Who are familiar with the originals are well aware that the large reproductions in Conze's work are in some respects misleading. In the lithographed plates the colour of the ground is too brown and too dark. In the others the use of dotted surfaces for purple is confusing, and the original effect of certain kinds of field ornament cannot be given by tracing their outlines.

^{page 68 note 3} Jahrbuch, 1887, pp. 211—215, Pl. XII. No. 914 in the Ethnic Museum. This vase, which is considerably smaller than the others, is fitted with a cover. Strictly speaking it has no figure-subjects, the human, animal and mythologie forms being merely decorative. Its provenance is unknown.

^{page 68 note 4} Ephem. Archaiol. 1894, pp. 226–238, with a general sketch and two coloured plates by M. Gilliéron. This amphora, which is numbered 354 in the Museum, was said by the vendor, but without support of evidence, to have been found in Crete. It is referred to in the text as the ‘Herakles’ vase from its main figure-subject.

^{page 68 note 5} The whole vase is unfortunately in very bad condition. Large portions were missing altogether, and the surface and edges of those which remain were so much worn as to make their readjustment difficult. This task has been successfully accomplished by members of the British School and by the mender to the Ethnic Museum. In the drawings an attempt has been made to reproduce what a prolonged examination reveals to have been present once, rather than to suggest the features still visible on a casua inspection.

^{page 72 note 1} A second bowl, also acquired by Mr. Cecil Smith at Melos, is practically identical in shape but shows a decorative scheme of swans upon the shoulder. This vase bas a lid in the edge of which are holes corresponding to those pierced in the rim of the bowl and thus showing their purpose. Other than this the nearest analogies to our vase seem to be the Ionic deinoi published by Pettier, E., Bulletin de Correspondance hellénique, 1893, p. 424Google Scholar, all of which, however, are without the foot. The Nequada vases published by the Egypt Exploration Fund afford a curious though, I suppose, an accidental parallel. One of these is in the British Museum, a little bowl numbered A 1679,₂ in Case I, which contains vases mostly from primitive tombs in the Greek islands. It has no foot, and the holes for suspension are horizontal piercings in ears attached to the sides. It is completely covered with a decoration of thin and rather straggling spirals in red varnish paint on a buff ground. This bowl was brought by Mr. Greville Chester from Sameineh in Upper Egypt.

^{page 74 note 1} Unless the upper edge of the neck of the amphora published by Boehlan (cf. supra p. 68, footnote 3) presents the same feature.

^{page 74 note 2} Jahrbuch, 1897, p. 195.

^{page 74 note 3} Jahrbuch, 1887, pl. 3.

^{page 74 note 4} Id. pl. 4.

^{page 74 note 5} If, as is possible, the vincula of the spirals on the bowl were originally purple, we have a further resemblance.

^{page 75 note 1} In the Jahrbuch for 1897, pp. 195–199, Dr. Wide gives a list of these. They are Nos. 895, 896 in the National Museum at Athens, of which he gives photographs: two vases, one in Leyden and one in Paris reproduced by Conze, , Zur Geschichte der Anfänge der griechischen Kunst, Taf. XI. 1Google Scholar,2: and the Stockholm vase, which was the subject of Wide's article, and of which he gives a reproduction on Pl. 8, 14.

^{page 75 note 2} Archäologischer Anzeiger to the Jahrbuch for 1897, p. 78.

^{page 75 note 3} Cf. Böhlau, J., Jahrbuch, 1888, p. 325Google Scholar.

^{page 75 note 4} It also occurs in the large Eretrian amphorae Nos. 1005, 1006, in the Ethnic Museum at Athens. Cf. Couve's, M. article in the Bulletin de Correspondance Hellénique, 1898, p. 279Google Scholar.

^{page 75 note 5} Annali, 1877. Tav. d'Agg.CD. 5, 18.