The Kamares cave on Mount Ida has been known now for more than twenty years as a prehistoric sanctuary, but its full excavation was only carried out in the summer of 1913 by a party from the British School, which thus resumed the work in Crete discontinued since Palaikastro gave way to Sparta in 1906. This early fame of the cave was due to the discovery in it in the early nineties of a number of vases and a few figurines, which Dr. Hazzidakis secured from the shepherd who had found them and placed in what was then the Museum of the Archaeological Syllogos at Candia. Some of the vases bear the marks of the rivets with which their peasant owner mended them.

The results of our excavation have so greatly supplemented this first instalment that the authorities of the Candia Museum have been able to do a great deal in the way of the reconstruction of vases, of which until now only small pieces were available, and these restorations have been put at our service for the photographs which accompany this report.

As soon as these vases were discovered they were at once recognised as belonging to a kind until then almost entirely unknown, and the name ‘Kamares,’ which was at once given to them, still remains current as a general description of that kind of Cretan pottery which has a black ground and white or polychrome ornament, although at the present stage of research rather as a convenient name for the technique than with any chronological meaning.

The village of Arkalokhóri in central Crete lies a short distance to the south-west of Lyttos. To the east of the village is an eminence to which the ruined church on the top has given the name of Prophetes Elias. Below the summit on the western slope is a cave, the mouth of which is shewn in Fig. 1. Many years ago the peasants found bronze objects and potsherds at this point, and began an excavation of the cave in the hope of finding buried treasure. On entering the cave they found more and more pieces of bronze, described as blades of knives and lances, as well as some beads, and in order to get more quickly to the bottom of the cave, where they supposed that the treasure would be, they broke up the greater part of it with dynamite. They say that they collected eighteen okes of bronze objects, for the most part rusted blades, and sold them in Candia as old metal. No report of this had ever reached me, but last year a peasant from Arkalokhóri brought some blades to the Candia Museum from this cave, and thus led me to go and make an examination of the place.

The mouth of the cave was then choked with large boulders which had fallen down from above; these I broke up and rolled down the slope. Beneath and amongst them I immediately began to find blades of swords and knives and double-axes, all of bronze with the exception of a small axe which was of silver. The area thus cleared after rolling away the boulders in front of the present cave, is about 4 metres in length, and in width from 1 to 2 metres, and seems to have been the floor of the original cave.

The objects described in this section were found on a small prehistoric site at Tchukurkend on the eastern side of the Beishehir lake between Eflatoun Bounar and Kirili Kassaba. The village of Tchukurkend lies some two hours from the Hittite monument at Eflatoun Bounar and fifty minutes to the S.E. of the small town of Kirili Kassaba. On the previous day, coming from Eflatoun Bounar, I had ridden for the greater part of the way across the lower ground by the shores of the lake and had thus missed the village, but a jeweller in Kirili to whom I showed a small celt, with an enquiry if any were to be obtained in the neighbourhood, informed me that they were often found by the peasants at Tchukurkend. There in the hands of the villagers I found the two human figurines (Fig. 1A, B), the two animals (Fig. 2c, d), and the small fragment (Fig. 2b). They were all found with certain others which had been destroyed, on a low hill immediately above the village. Here was clearly a small prehistoric site in antiquity, now much denuded, on which I found a large quantity of obsidian, principally of the Melian variety, and a few fragments of the red-faced pottery common on other early sites in the district.

Few terms in archaeology have become as ambiguous as the term Proto-Corinthian. Used at first not unreasonably of a class of small vases found over a wide area, which bear considerable resemblance to Corinthian pottery, the name has come to be applied to a number of vases which differ very widely from the fabrics originally so called. Furtwaengler first extended this term to two vases found near Thebes, and since then the appropriateness of the term has not been seriously questioned. Nowhere is this extension of the term more unsuitable than at Delphi, where a large quantity of Proto-Corinthian ware in the original sense of the term was found, as well as the Geometric pottery which Perdrizet describes as follows: ‘le géométrique delphien appartient à la catégorie appelée protocorinthienne par M. Furtwaengler: il est douteux qu'on puisse l'attribuer à une fabrique locale.’ The most cursory comparison of the Geometric pottery of Delphi, hitherto classed as Proto-Corinthian, with the Proto-Corinthian originally so called, makes it clear that whatever be the provenance of the Geometric, the same name cannot reasonably be applied to both fabrics. In the real Proto-Corinthian pottery a variety of shapes occurs, all of small size. The most characteristic are the aryballos, the lekythos, the pyxis, and the long-necked, flat-bottomed jug. The Delphic Geometric pottery on the other hand has little variety in its shapes, and, as will be seen below, these differ in size and form from the Proto-Corinthian. Again the distribution of Proto-Corinthian pottery extends over a very wide area; it occurs all over the mainland of Greece, in Italy and Sicily, and even in Asia Minor. The vases on the other hand which, as will appear later, may be brought in line with the Geometric pottery at Delphi, are few in number, and only found within a small area.

According to early investigators of the topography of the Megarid, the sites of Minoa and Nisaea were identified, the former with the small hill surmounted by a mediaeval castle standing in the middle of the coast-line in front of Megara, and the latter as the hill upon which is the church of St. George, above the hamlet of Pachi (Fig. 4). Spratt was the first to suggest this identification, which he based both on the existence of remains of ancient buildings around the small hill and on the assumption that it was originally severed from the land (Fig. 1).

His view was subsequently confirmed by Lolling.

It has, however, recently been suggested that the site identified as Nisaea is really the site of Minoa, while that given to Minoa is the real Nisaea. This reversal of the accepted identification is based on the fact that when Demosthenes in 424 made his attempt to capture the Long Walls of Megara, reinforcements were sent from Athens by land via Eleusis. These reinforcements, it is urged, would have taken up their position immediately outside the eastern Long Wall of Megara, as it would have been both difficult and dangerous to make the long circuit round the north of Megara and camp outside the western wall.

At the west end of the mediaeval castle on the Megarian Minoa the front surface of the west wall has fallen away. The foundations of the wall thus laid bare consist of large square blocks of poros and blue limestone from Hellenic buildings. The corner block is of blue limestone and is inscribed in three columns on its upper surface. On the face of the second and third columns are traces of mortar which show that there was originally a course above the stone. The block is 1·12 m. in breadth and ·55 m. in height. Its depth cannot be ascertained, as it is buried almost up to the surface. At the top is a margin 10 cm. in width, extending along the whole length of the stone. The inscription is below the margin, and the three columns are separated by vertical grooves. The first column is ·355 m. in breadth, the second ·32 m., and the third ·265 m.

The σύνοδος of the μύσται and τεχνῖται of Dionysos Breiseus at Smyrna has long been known to students from several series of inscriptions. These, although defective in themselves, can be supplemented by the inscriptions of analogous associations, so that the details of the society's organisation are tolerably familiar, and for the present require no further discussion. But certain questions have not so far been decided. Such concern the origin of the god, the meaning of the epithet, and the situation of his temple at Smyrna. The present paper is an attempt to answer these questions, some new evidence having recently come to hand.

The earliest certain home of Dionysos Breiseus seems to have been Lesbos, where the name of Brisa has been known for many centuries. Homer's heroine will be recalled: Androtion knew a Lesbian promontory called Βρῆσα: the Lesbos of to-day has a hill and a village called respectively Βρίσιον and Βρίσια.

The musical Notation of the Greek Church is decipherable as far back as the beginning of the Round System in the thirteenth century. Some examples of this were given in my article in last year's Annual. But before the invention of the Round System another kind of notation was in use, called for convenience the Linear System, the rules of which are very obscure. The reading of the Round System was made easy by the familiar little treatise called the Papadike, which occurs in many MSS., and explains the main principles of the notation. Nothing of the kind seems to have been known dealing with the Linear System. But the fragment which I am now to place before the reader may be expected to throw some light upon it. I photographed this fragment on my visit to the Monastery of Laura in 1912: and so far as I know, this is the first time that anything has been written about it.

The fragment is a single leaf of parchment bound up with the MS. Laura Γ 67, a small quarto codex containing some of the Stichera of the Triodium and Pentescostarium, i.e. the original hymns (excluding Canons and words set to tunes not specially composed for them) sung at the movable days in Lent, Holy Week and Eastertide. The manuscript is written in a bold well-formed hand, the text in black, the notes in red.

In April 1913, while at Melos, I took the opportunity of visiting the two churches of Kepos on the south side of the island at the South-eastern extremity of Mount Elias. The northern of these churches has in recent years been restored, and an account of it with a plan, has been published by Messrs. Fletcher and Kitson in the Annual of 1895–6. This plan, however, was made before the church was restored and before the collapse of the dome which made that restoration necessary. There are therefore certain points which need revision in view of the present condition of the church, and there are others which require an altogether more detailed treatment.

The restoration has been carried out comparatively well. The main structure has been retained without serious alteration, but the paintings which were inside the dome are no longer in position and at present lie on the floor of the church in fourteen fragments.

In the apse of the church is a cruciform font built up of slabs set on edge, of which separate plans are given in the article referred to above. The plans there given, however, do not correspond in all details with the font; these details can be seen from the plan and section given here (Figs. 1, 2), made from measurements which I took on the spot. Messrs. Fletcher and Kitson say that ‘on three sides are seats, on the fourth towards the door of the church, are two steps down, and in the middle a well 2 ft. square.’

The Melian font described in the preceding paper by Mr. Casson is paralleled by several other examples from various parts of the Aegean area; all have in common a more or less cruciform plan, one or more of the arms of the cross being provided with steps. Some of these fonts are now published for the first time: of the majority, descriptions have already appeared, but for the most part in publications not generally accessible. It therefore seems worth while to publish the series of measured drawings and notes given below, which have been made at various times in the last ten years. The series of fonts described is intentionally confined to the Aegean area and Constantinople; the references at the end of the paper will enable the comparisons to be extended to the whole group.

The area over which the Tsakonian dialect is spoken lies on the east coast of the Peloponnese between the Parnon range and the sea. Its northern boundary is roughly the torrent which, rising on Parnon above Kastánitsa, flows into the sea near Ayios Andréas, its southern the torrent which, also rising on Parnon, passes through Lenídhi to the sea. A mountain range stretches along the coast from end to end of the district, reaching its highest point (1114 metres) in Mt. Sevetíla above the village of Korakovúni. Between Tyrós and Pramateftí, the seaward slopes of this range are gentle and well covered with soil. Behind these coast hills there stretches a long highland plain, known as the Palaiókhora, which, in the north, is fairly well covered with soil, but gradually rises towards the south into a region of stony grazing land, and terminates abruptly in the heights above Lenídhi. The high hill of Oríonda rises out of the Palaiókhora to the west and forms a natural centre-point of the whole district. Behind it stretching up to the bare rock of Parnon, is rough hilly country, cut here and there by ravines and offering but rare patches of cultivable land.

A find of French and Neapolitan mediaeval silver coins made recently in one of the southern Sporades, probably Kasos, attracted my attention in a jeweller's shop at Smyrna in the spring of 1912. In 1913 I found the same hoard with some additions in changed hands. I then managed to secure a representative selection and an analysis of the whole collection.

Among the Mahommedan religious antiquities of Asia Minor the tomb-sanctuaries held to represent the resting-places of Arabs killed during the forays of the viii–ix centuries form a well-marked and extremely interesting group. Their authenticity is on general grounds more than doubtful. The campaigns of the Arabs led to no permanent occupation: the lands they had conquered for the moment were restored to Christendom or fell to alien races. Only in the borderlands, where in times of peace Christian and Moslem might meet on equal terms, can we expect a true tradition regarding Arab graves or a continuous veneration of them to have persisted.

Of these borderland Moslem cults supposed to date back to the Arab period we can point to two examples, the tomb of the ‘sister of Mahommed’ at Tarsus and the tomb of Umm Haram in Cyprus.

The former is mentioned by Willibrand von Oldenburg (1210) as still a place of Moslem pilgrimage under the Christian kings of Armenia. It was situated outside the church of S. (Beatus) Peter and S. Sophia in the middle of the town.

At the first appearance of the Ottomans, towards the close of the thirteenth century, Christian and Turk had already been living for two centuries side by side in the interior of Asia Minor under the rule of the Seljouk Sultans of Roum. The political history of this period is still emerging from obscurity: the social and religious history has hardly been touched. The Byzantine historians, concerned only incidentally with provinces already in partibus, give us no more than hints, and we have none of those personal and intimate records which are apt to tell us much more of social conditions than the most elaborate chronicle.

The golden age of the Sultanate of Roum is undoubtedly the reign of Ala-ed-din I. (1219–1236), whose capital, Konia, still in its decay bears witness by monument and inscription to the culture and artistic achievement of his time. Ala-ed-din was a highly-educated man and an enlightened ruler. He was familiar with Christianity, having spent eleven years in exile at Constantinople. One of his predecessors, Kaikhosru I. (1192–6, 1204–10) who likewise spent an exile in Christendom, nearly became a Christian and married a Christian wife.

The Karaosmanoglou dynasty, which during the eighteenth century and part of the nineteenth ruled the province of Saroukhan (Magnesia) in Asia Minor, stands almost alone in Turkish history as an example of a family which not only won and retained a wide local supremacy, but was conspicuous for family solidarity and wise administration throughout its tenure of power. Of the numerous pretenders to independence who disputed the Sultans' sway during the centuries in question few were able to make their claims hereditary and none could justly boast as could the Karaosmanoglou that their administration had raised their dominions from poverty and disorder to a degree of prosperity unknown probably since the Roman empire.

The history, real and mythical, of this great Turkish family affords an interesting illustration of the growth of folk-tradition and its relation to historical fact, since we have here the rare advantage of being able to compare and contrast fact and fiction, and even to trace the growth of the myth. Less than a hundred and fifty years from the rise of the family, which is not extinct at the present day, its real origin is completely obscured; its actual history is supplanted by a purely legendary set of incidents and associations by which the family gains in prestige no less than in antiquity.

In Turkish geographical nomenclature certain ‘round’ numbers are regularly employed in an arbitrary sense. Most important of these are ‘a thousand and one’ (bin bir), used to express the idea of ‘countless,’ and ‘forty’ (kirk), which is similarly used for ‘numerous.’ As examples of the first may be cited the well-known ‘thousand-and-one-column’ (Bin Bir Direk) cistern at Constantinople and the ‘Thousand and one Churches’ (Bin Bir Kilisse) in Lycaonia. For the second we may instance several rivers called Kirk Getchid (‘Forty Fords’, in Greek Sarandáporos), the town Kirk Agatch (‘Forty Trees’), springs called Kirk Gueuz (‘Forty Eyes’), districts called Kirk In, Kirk Er (‘Forty Caves’) and numerous others.

Side by side with names like the foregoing, which explain themselves if we read ‘numerous’ for ‘forty,’ we find certain localities denominated simply ‘the Forty’ (Tk. Kirklar, Gr. Saránda). They are especially common in Pontus but occur also elsewhere, as e.g. in Mysia, where there are at least two villages called Kirklar, and in Caria, where the name is applied to a site with ruins of a church near the ancient Loryma and to an ancient tomb east of Knidos. Similarly mysterious are names like Kirklar Dagh (‘Mountain of the Forty,’ not ‘Forty Mountains’) which like the foregoing, imply an association with forty persons. These ‘forties’ call for explanation.

The pictures on pp. 230, 231 (Figs. 1, 2) are taken from a red-figured amphora of Panathenaic shape in the Louvre (G 61). They are more curious than comely, but if devoid of merit they are not devoid of interest. Let us put them beside the pictures on p. 233 (Figs. 3, 4), which come from a vase in Count Stroganoff's collection. The lower part of the dress and the sandalled feet are almost line for line the same in both vases; the altars are the same except in one small detail: notice carefully the sleeve in Fig. 1, and compare it with the sleeve in Fig. 4. Other resemblances will strike the eye, but it will be enough to mention one: is there not something odd about the perpendicular fold-lines on the chiton in Fig. 4? in archaic drawing, perpendicular folds go right on until they meet a transverse line; but here they stop of their own accord at Athena's waist: now the grouped folds on the sleeve of Fig. 2 behave in the same obstinate way.

The Louvre vase G 61 has a pendant in the Louvre vase G 60, which is figured by Pottier in his Album, Pl. 95. The subjects are similar: on A of G 61, Athena bending at an altar, holding a spear and a flower; on A of G 62, Athena bending at an altar, holding a wreath and a spray of smilax; on B of G 61, a woman worshipper at an altar, holding a smilax spray; on B of G 60, a man worshipper at an altar, holding a phiale.

Attention was first drawn to this subject when Mr. Dawkins published in the Journal of Hellenic Studies in 1906, a full account of the masquerade performed by Greeks in the district of Viza in Eastern Thrace. Both he and Mr. Lawson had previously described the Skyros Carnival, and Polites had recorded several instances from Northern Greece. In 1908, Katsarov drew attention to the fact that such mumming was also known among the Bulgarians both near Adrianople and in Bulgaria proper. In 1910, I published in the sixteenth volume of this Annual an account of the Epiphany and Mayday masquerades that Mr. M. S. Thompson and I witnessed in Thessaly. Since then others have noted similar masquerades elsewhere and the evidence of these mumming plays has been used by writers on the origin of Greek tragedy and comedy. The most striking point brought out by recent study of the subject is that such plays are not confined to Greeks alone, they are performed by Bulgarians, Vlachs, Albanians and apparently, even by Gipsies. While travelling in Thessaly and Macedonia during the years 1910–1912, Mr. Thompson and I noted the existence of many similar masquerades.