^{page 1 note 1} So Rumpf, , J.d.I. 1933, 62 n. 11.Google Scholar

^{page 2 note 1} By ‘Rhodian’ I mean generically the various local styles and phases, which have been called Rhodian, Milesian, Rhodian-Milesian, Camiran and Wild Goat style. These terms include genuine Rhodian, Samian, Aeolian, and the little-known products of Ephesos and Marseilles. The local styles have rarely been distinguished (e.g. E. R. Price East Greek Pottery II). I have kept Aeolian separate; and where I have referred to particular vases, have usually mentioned their provenances. On the whole, since the mass of this ‘Rhodian’ that is published comes actually from Rhodes, the inverted commas could often be removed: where I have used Rhodian without the inverted commas, I feel sure that the style is the local style of Rhodes.

The development of ‘Rhodian’ is clearest in Rhodes, thanks to Maiuri and Jacopi and, in a lesser degree, to Biliotti and Salzmann; and I have presumed, without much direct evidence, a parallel development for the other local styles that go to make up ‘Rhodian.’ Miss E. R. Price has divided this medley into two classes, which she calls A and B. Class A includes all the vases without incision on them: class B is of vases on which incision occurs (East Greek Pottery 14-15). This is now misleading. Roughly about 600 B.C., witness Clara Rhodos, the dominant Rhodian style changes to a coarser and clumsier style, which sometimes uses incision—notably on the shoulder of oinochoai, dinoi and kraters, while it only very occasionally intrudes on the plates, which are the commonest shape, and not at all on the dishes and amphorae. The change goes much deeper than the simple introduction of incision : it extends to shape, technique and, to some extent, to the motives of decoration (note, for instance, the dividing bands in Fig. 10, and the greater popularity of the dog). The plate, now the commonest shape, is at least rare earlier. The dish becomes smaller and clumsier. The oinochoe gets a long cylindrical neck and a narrow oval body (cf. Kinch figs. 79, 114). The amphora (e.g. J.d.I. 1886, 140, fig.) is closely related to the new oinochoe, and contrasts strongly with the stylistically middle Rhodian amphora, Louvre N 2342 (see below pp. 55–56). The bowl and the krater are new. The dinos is rare earlier and perhaps does not change. (The aryballoi, of which there are several at Mykonos and Delos, are perhaps all Cycladic imitations.) The slip often deteriorates to a thin wash; and often there is no slip. The technique of drawing becomes clumsier altogether. The term B is used in this article to cover all the products of the incised period, which continues through the first quarter of the sixth century. Then we lose the thread: for the second quarter there is a lacuna in the evidence of grave-groups and stratification. By the term A is meant all that goes before the period of incision. Possibly much of what I have classed as A overlaps with the B style: for such pieces I have sometimes used the term Late ‘Rhodian’ A. Middle ‘Rhodian’ is the phase of the last third of the seventh century, the floruit of the familiar wild goat oinochoai. Late ‘Rhodian’is equivalent to ‘Rhodian’ B. In Naucratite incision seems to have swept away almost completely the old style: Naucratite A, therefore, means the unincised, B the incised ware.

^{page 2 note 2} J.d.I. 1933, 62 n. 11, 69–83.

^{page 2 note 3} Prinz 39–42.

^{page 2 note 4} Pfuhl i 155–8.

^{page 2 note 5} East Greek Pottery 19–21.

^{page 3 note 1} See the list of illustrations in Pfuhl i 156. The drawings in Tanis ii are rarely accurate.

^{page 3 note 2} Kunze, , Ionische Kleinmeister, A.M. 1934, 81–122Google Scholar, discusses these cups fully. Dr. Kunze very kindly lent me the proofs of the article, so that I have been able to make references to it. For the relationship of these cups to Fikellura see in particular Kunze 85–86, 92–3, 117–118.

^{page 3 note 3} Louvre F 68, from Italy. C.V.A. Louvre viii, pl. 510, 3, 5 and p. 61, where bibliography is given. PLATE 11 b.

^{page 3 note 4} Kunze, A.M. 1934, 101–104Google Scholar, Beil. 8, 1–2; Price, in C.V.A. Oxford 2, p. 89Google Scholar; Beazley, in J.H.S. 1932, 169 n. 13Google Scholar,—‘pure East Greek—Fikellura style.’

^{page 3 note 5} Gf. the faces. The man on the Vineyard cup has a shorter torso than most Fikellura men.

^{page 3 note 6} Kunze says not, A.M. 1934, 117.

^{page 4 note 1} Cl. Rh. vi–vii, Nisyros grave xx, pl. 1 (a coloured drawing, but accurate enough for this comparison).

^{page 5 note 1} E.g. in the middle ‘Rhodian’ dinos B.M. A 741 (Kinch fig. 73; Price, E. R.J.H.S. 1924, 195 fig. 22)Google Scholar, and in a large ‘Ionian’ bowl in Mykonos, (Délos xvii pl. 52Google Scholar, Vases ‘Naucratites’ no. 3); in late ‘Rhodian,’ Kinch pl. 15, 1 (krater).

^{page 5 note 2} The neck of the dog has now been inserted.

^{page 6 note 1} Otherwise the object in the right-hand corner does not seem explicable. The Πὁτνια θηρῶν is rare in East Greek vase-painting. For an earlier Artemis holding two lions compare Technau A.M. 1929 Beil. 14, 4. A later Artemis carries one lion on the stele from Dorylaion (Körte, A.M. 1895, pl. 1Google Scholar). On both these also the lion is carried with tail downwards.

^{page 7 note 1} Cf. pp. 86–7 below.

^{page 7 note 2} Technau, , A.M. 1929 Beil. 19, 1Google Scholar; Kunze, , A.M. 1934, 83 no. 1; 85–6, pl. 6, 1.Google Scholar

^{page 7 note 3} Technau, , A.M. 1929 Beil. 19, 3Google Scholar; Kunze, , A.M. 1934, 83 no. 2, pl. 6, 2.Google Scholar

^{page 8 note 1} PLATE 1 a, c.

^{page 8 note 2} Dugas, , Délos xvii p. 71Google Scholar, cites three other instances of such sports—Caeretan hydria, Louvre E 696, Pottier pl. 52; Amphiaraos krater, F.-R. ii pl. 122; Italian lid, Schaal Frankfurter Vasen 73. Cf. also the fragment of a Laconian column-krater, p. 150 below.

^{page 8 note 3} For the meaning of this provenance, see p. 87, n. 1.

^{page 15 note 1} See Rumpf, Gnomon i 326Google Scholar. Dr. Ackermann tells me that the provenance is Aegina.

^{page 16 note 1} I have been unable to find this vase, and rely on the drawing in Tanis ii.

^{page 17 note 1} The same holds true for the Berezan fragments, no. 9, which also provide another example of a comast with head turned back. No. 10, if the drawing is to be believed, is in its subsidiary decoration conveniently close to the Altenburg vase and also connects with this group K 4.

^{page 18 note 1} Note especially the bearded pigmy. See Beazley, B.S.R. 1929, 2Google Scholar, and cf. the Northampton amphora and the François vase.

^{page 18 note 2} Cf. possibly the dinos stand on a late Geometric vase from Athens, (A.M. 1892, 209 pl. 10Google Scholar). I am indebted to Professor Webster for this suggestion.

^{page 18 note 3} Cf. perhaps the cubes up which the Naucratite partridge is stepping (J.H.S. 1924, 213 fig 51).

^{page 21 note 1} Birmingham University has a number of pieces of which it is not recorded whether they came from Cyprus or from Biliotti's sale. It seems, however, a reasonable assumption that the Cypriot is from Cyprus and the East Greek from Rhodes, and I have given Rhodes as the probable provenance of the four vases mentioned here (L 5a, M 3a, K 5a and the oinochoe cited on p. 57, n. 2).

^{page 21 note 2} J.d.I. 1886, 141—‘entbehrt des Ueberzugs, ist jedoch geglattet’—is, I think, mistaken.

^{page 24 note 1} The linea alba, the penis (set much too high) and the rudimentary beard line would then be misunderstandings or else simplifications of Attic. The Halle amphora is published by Langlotz Zeitbestimmung pl. 1, 1. This is the best explanation of the figure. It has also been suggested that the figure might be a pigmy and the vase a companion piece to some such vase as no. 6. But there is nothing by which to identify the figure as a pigmy: the thickness of the thigh and other apparent distortions are characteristics of this painter's work.

^{page 24 note 2} Cf. two Attic B.F. cups of about 560 in Boston (the first, A.J.A. 1913, 2 fig. 1 and 1923, 427 fig. 2; Buschor Greek Vase-Painting 100 fig. 92: the second, A.J.A. 1923, 426 fig. 1). Etruscan B.F. amphora, Louvre E. 723, with lion-headed man (Mus. Map. III pl. 59; Perrot and Chipiez ix fig. 217; Cook, A.B.J.H.S. 1894, 117 fig. 12Google Scholar; Zervos fig. 46). On gems in the British Museum are a panther-headed man (Furtwängler pl. 8, 13), a fox-headed and a donkey-headed man. I have found no other examples of hare-heads.

^{page 25 note 1} Dr. Laurenzi kindly had the vase dismantled. The illustration is, of course, mainly wrong. Actually the shape is rounder than usual. The drawing of the animals is fair. The S-shaped crescents resulted from a wrong join, which also distorted the shape. The animals look earlier than I had supposed, and perhaps belong rather to Group B II.

^{page 26 note 1} Grave no. 2 in the same cemetery (Papatislures) also seems to have been Used twice at much the same periods—end of seventh century and middle of second half of sixth century. It also is a chamber tomb. Cf. also graves x and xii at Makri Langoni, Cl.Rh. iv.

^{page 30 note 1} See Jacobsthal, G.G.A. 1933, 1–2, 8Google Scholar, and Rumpf, J.d.I. 1933, 61.Google Scholar

^{page 31 note 1} Assuming that the amphora described by Boehlau is that illustrated by Zervos. The other Fikellura piece in Basle (Y 6) mentioned by Boehlau was recatalogued also, as 1906.257.

^{page 33 note 1} Flower as FIG, 9, 29. Cf. N 1.

^{page 33 note 2} See p. 26 n. 1.

^{page 36 note 1} Not from an amphoriskos, as stated in C.V.A. Oxford ii 85.

^{page 36 note 2} E.g. the fragments of a deep band cup in the British Museum and Boston 350.3 (Fairbanks pl. 38).

^{page 37 note 1} Another fragment of this or a similar oinochoe is in Vathy Museum, from Tigani.

^{page 38 note 1} Compare e.g. the dish B.M. A 710 and the cup Kinch pl. 18, 3 and fig. 129 for the single form of no. 2. For the double form of no. 1, the cup Kinch pl. 18, 2 a. On these cups the spacing and alternation of the ornaments is the same. The roundel filling of the angles is typically ‘Rhodian.’

^{page 38 note 2} See p. 58.

^{page 38 note 3} See p. 78.

^{page 38 note 4} See p. 78. As on many ‘Rhodian’ dishes.

^{page 39 note 1} Cycladic imitations of ‘Rhodian’ are not uncommon. Several are illustrated in Délos x.

^{page 40 note 1} Formerly in the possession of Captain Bunbury, who on two or three occasions received small fragments of pottery from the Egypt Exploration Society.

^{page 40 note 2} Prof. Jacobsthal has sent me a photograph of another fragment in the B.M. (A 1728), perhaps from the same vase. Bars, alternately varnished and reserved. From Naucratis also.

^{page 40 note 3} The Clazomenian riders with similar distortion are also women (e.g. B.M. B 116¹: Tanis ii pl. 29, 4).

^{page 41 note 1} The head may date as late as 530: compare the fragments of K 7 and K 8 (PLATE 10 d and e), but this looks clumsier and earlier—I think about 540. The treatment of ear and ear-ring is similar to a Clazomenian type. The hair seems to be done in a krobylos. The ornament in the centre looks like a lotus-flower springing from volutes, much as in Gaeretan (e.g. Berlin Inv. 3345, Neugebauer pl. 18). The photograph came to my notice too late for the piece to be properly incorporated.

^{page 41 note 2} B.M. 1934.12–1.217. From Naucratis or Tell Defenneh.

^{page 43 note 1} From the shape of the vase, this graffito must have been made after breaking.

^{page 43 note 2} For the dividing bands cf. the fragments from Ephesos in the B.M. (Hogarth Excavations at Ephesus pl. 49 and fig. 45).

^{page 43 note 3} E.g. Payne Necrocorinthia no. 887.

^{page 43 note 4} See E. R. Price East Greek Pottery 27.

^{page 45 note 1} Another similar fragment is in Kassel also from Samos. Rays of star at irregular intervals. Part of the ring foot is preserved.

^{page 46 note 1} For East Greek Gorgons see Payne Necrocorinthia 88. Fragments of an Ionian B.F. cup in the B.M. from Naucratis show snakes, probably from a Gorgoneion (see Kunze, A.M. 1934, 93–94 no. 8. Beil. 7, 3Google Scholar).

^{page 46 note 2} Cf. pp. 58–9.

^{page 46 note 3} See p. 40, n. 7.

^{page 48 note 1} Among the earliest vases with shield aprons are a B.F. lekythos in New York, 26.60.76, of about 500; and the Berlin kyathos of the Panaitios painter (Gerhard Auserlesene Vasenbilder pl. 51, 4) of 500–490. I am indebted to Professor Beazley for these references.

^{page 48 note 2} Close, Y 16 from Rhodes; Y 19 from Aegina. Less close, Y 18, unknown provenance; Y 24 from Rhodes.

^{page 52 note 1} This pattern, on a small scale, is not uncommon in Late ‘Rhodian’: e.g. the Munich sherd, PLATE 17 e. Cf. also G 6.

^{page 52 note 2} Cf. the amphoriskoi Y 25 and 34.

^{page 53 note 1} PLATE 10 a, b.

^{page 53 note 2} Descriptions (e.g. those in Cl.Rh.) often omit slip. Berlin Inv. 2975 (L 7) is said to have no slip (J.d.I. 1886, 141), but is actually thinly slipped. Probably the same is true of the Hague amphora, E 2.

^{page 54 note 1} The coloured plate Cl.Rh. iv pl. 5 is adequate on this point.

^{page 55 note 1} For other olpai see Payne Necrocorinthia 299. But the Fikellura olpe has no obvious connection. The ‘Rhodian’ olpe is of very different shape: I know only three examples —Berlin Inv. 2931 (J.d.I. 1886, 139, fig. in text), Mykonos Rh. 15 (Délos xi pl. 42, Vases Rhodo-ioniens no. 15) and Turin, (J.H.S. 1926 pl. 9, 2)Google Scholar. An earlier East Greek olpe in close imitation of a Middle or Late Protocorinthian olpe is Rhodes 14709 (Cl.Rh. vi–vii 355 fig. 102).

^{page 55 note 2} The shape is sufficiently distinctive to determine as Fikellura the vase Rhodes 13206 (Cl.Rh. iv grave lxxxiv fig. 198; Camiros), the surface of which has almost completely perished. Jacopi and Jacobsthal (G.G.A. 1933, 12) say that this is a Fikellura amphora unslipped and unpainted. But there are definite traces of varnish, e.g. inside the lip.

^{page 55 note 3} N 2342. From Carniros. C.V.A. Louvre i pl. 18, 12.

^{page 55 note 4} Rhodes 1388. Annuario vi–vii grave iv p. 264 fig. 164. On the neck perhaps traces of meander and square. The handles are two-reeded.

^{page 55 note 5} The zigzag band is rare in Fikellura; the running loop below unknown. I therefore consider this fragment late ‘Rhodian.’

^{page 55 note 6} E.g. Louvre B 561. From Myrina. B.C.H. 1884, 509, fig. in text.

^{page 56 note 1} I. Colmar. Bulle A. A. 1904, 49 fig. I. From Camircs (?). 2. Rhodes 12892. Cl.Rh. iv grave lxxiii fig. 177. From Camiros.

There is also an amphora in Berlin, Inv. 2987, of Fikellura shape; the reeds of the handles are, however, not properly distinct. Height about ·30 m. The vase was painted over with dark varnish in imitation of Bucchero in the same way as the oinochoe described on p. 57, n. 21: here no traces of earlier decoration are to be seen. From Rhodes. There is no evidence to determine the date of this piece.

^{page 56 note 2} C.V.A. Louvre i pls. 22, 49.

^{page 56 note 3} Rhodes 12976 is related to Fikellura: amphora of oval body and with cylindrical handles, set vertically. Ht. ·33 m. Clay, slip and varnish as normal in Fikellura. Lip, handles and foot varnished. Neck, scraggly palmette, terminating on each side in a lotus flower (neither ornament of ordinary Fikellura type). Shoulder, tongues. Belly, alternate thick and thin bands. From Makri Langoni, Camiros. Cl.Rh. iv grave lxxviii figs. 190, 191. C.V.A. Rhodes i pl. 421, 4. Found with the banded amphoriskos Y 43 and with an Attic Floral Band Cup: so probably end of sixth century. Shape and decoration distinguish this vase from Fikellura; but the technique is similar and there is no other group with which to connect it.

^{page 56 note 4} Occasionally with ivy leaf—D 1, E 1 (?), Q,1: doubled—Z 10. Once with lotus flowers and buds, unjoined—Z 12. Inside the lip a band or two of varnish is painted.

^{page 56 note 5} Once, L 16, the outer reeds are varnished, the inner reserved: as once, Y 13, the outer reeds reserved, the inner varnished. On E 7 the handles are ‘solid black.’

^{page 56 note 6} Simple meander—L 15.

^{page 56 note 7} Lotus flowers and buds—E 7, M 2, PP 1 and another fragment from Berezan. Eyes—L 6, L 16. Vertical zigzags—P 7. For the eyes, cf. numerous ‘Rhodian’ stemmed dishes and eye bowls.

^{page 57 note 1} The late ‘Rhodian’ oinochoe does not affect the Fikellura shape. Fikellura, here as often elsewhere, draws its origins from middle ‘Rhodian.’

^{page 57 note 2} Kinch fig. 130; PLATE 18. Ht.·24 m. From Rhodes. There is an oinochoe of similar shape and size in the B. M., A 634 (64.10–7.1576), from Rhodes. On the body are traces of a band of meander decoration, as on C 1, an animal frieze and, below that, lotus flowers and buds approaching the Fikellura type. Of the animals a goat and a dog can be made out. The vase was later coated with a thick varnish and fired hard to imitate Bucchero; and since further the surface is damaged, it is impossible to make out details of the earlier decoration. Still, the general shape of the animals looks ‘Rhodian’ rather than Fikellura. Cf. the amphora described on p. 56, n. 1, at end.

A smaller oinochoe of similar shape is in the collection of the University of Birmingham. Bucchero with red and white decoration. Height to lip ·18 m. Neck, meander and square (?). Shoulder, lotus flowers and buds, unjoined; the flowers white, of normal Fikellura type, and the buds red, outlined in white and with white chevrons on them (see FIG. 13, 11). Belly, fine meander and square: buds, unjoined: longish red rays. Probably from Rhodes. This piece must be close to early Fikellura.

^{page 58 note 1} E.g. the eye of ‘natural’ shape—S 1: the circular eye—R 1 (PL. 16 a.).

^{page 58 note 2} Miss Price describes the shape as ‘tiny phiale or bowl’ and cites Boehlau pl. 8, 1 as an example of the shape (C.V.A. Oxford 2 p. 85 and J.H.S. 1924, 184). There is no evidence at all for phialai.

^{page 58 note 3} E.g. Kinch pl. 18, 3, 7; 34, gr. 1, 7 (with Early and Middle Corinthian); Munich, Sieveking-Hackl 480, pl. 18 (see Kinch 58). These cups are much the same size as the Fikellura group.

^{page 58 note 4} I have preferred the name amphoriskos to lekythos-amphora: amphoriskos suggests the derivation of the shape better.

^{page 59 note 1} But see Price, J.H.S. 1924, 183.Google Scholar

^{page 59 note 2} See PLATE 15 for a comparison of amphora and amphoriskos, shape (a).

^{page 59 note 3} Illustrated in Cl.Rh. passim. For a specimen from outside Rhodes, see Izv. (Arch. Komm.) 35, 26 fig. 12: this vase has only one handle: from Taman or Kertsch. On the whole, vases of this group were not much exported.

^{page 59 note 4} Cf. the small group of amphoriskoi, Y group 4.

^{page 60 note 1} Excepting the isolated Vathy fragment (A 1).

^{page 60 note 2} Berlin F 3917. Diameter ·275 m. From Camiros. Salzmann pl. 55. Brunn Kunstgesch. in Bildern i, fig. 111; Zervos fig. 333. See p. 67. PLATE 19.

^{page 60 note 3} Samos, German Excav. Outer zone, goats; second zone, lions; third zone, boar hunt; fourth zone, palmettes. Germ. Inst. in Athens, photograph no. 1180.

^{page 61 note 1} E.g. Cl.Rh. vi–vii 541 fig. 81; from Nisyros and probably local. lb. 63 fig. 63; from Papatislures, Camiros.

^{page 61 note 2} Miss Price, East Greek Pottery 21, records a centaur, which unfortunately I have not-been able to trace.

^{page 61 note 3} A 1 (PLATE 14 b).

^{page 61 note 4} See pp. 4–5.

^{page 61 note 5} B 1 (PLATE 2 c); B 2 (PLATE 3 b). For another lion see pp. 93–5.

^{page 61 note 6} See p. 7.

^{page 61 note 7} PLATE I f (B 1) and PLATE 3 b (B 2). Another panther may be assumed for B 4.

^{page 61 note 8} On the Vlastos oinochoe. So also with the pantheress on the oinochoe in Leningrad (Rayet and Collignon fig. 30).

^{page 61 note 9} Taris II, pl. 31, 18 and 19.

^{page 61 note 10} B.M. 88.2–8.26; from Tell Defenneh.

^{page 61 note 11} Ducati Pontische Vasen, passim.

^{page 62 note 1} Kassel T 434. Lion, panther, siren. Shortly after 550.

^{page 62 note 2} B.M. B 286. B.M. Cat. Sculpture figs. 176, 177.

^{page 62 note 3} B.M. B 90. Op. cit. fig. 41.

^{page 62 note 4} Head Cat. Ionia pl. 1, 1; 3, 22; 34, 4, 5 and 6.

^{page 62 note 5} PLATE 3 b. For a second specimen see p. 93.

^{page 62 note 6} PLATE if.

^{page 62 note 7} Furtwängler, Kl.Sehr. I 488Google Scholar observes that a break in the centre of the spinal crest is characteristic of the Ionian type of boar. This is true for ‘Rhodian,’ Naucratite and Clazomenian.

^{page 62 note 8} Stamboul 6574: from Larissa.

^{page 62 note 9} PLATES 2 b, 3 a.

^{page 62 note 10} Cf. the Naucratite fragment in the British Museum, illustrated in J.H.S. 1924 pl. 10, 8. The Fikellura collar is certainly a genuine collar: the style does not go in for fancy decoration on its figures. Attic B.F. dogs of the middle and third quarter of the sixth century often wear collars, sometimes on a lead (e.g. Jacobsthal Orn. pl. 19b). On Laconian vases of the early sixth century dogs with collars are found chasing foxes without collars (e.g. the cups Louvre E 66$=C.V.A. Louvre 1 pl. 25, 4, and B.M. B 7). The same distinction holds generally for Protocorinthian and Corinthian, though the Chigi vase (Payne Protokorinthische Vasenmalerei pl. 27) shows hares with what on dogs might be considered collars.

^{page 63 note 1} PLATE I a, c.

^{page 63 note 2} E.g. Naucratite, fragment in B.M., (J.H.S. 1924, 211 fig. 43)Google Scholar and a fragment in Aegina. From Rhodes, the Perseus plate (PLATE 19). From Samos, the Boar hunt dish (p. 60, n. 3). For a contemporary East Greek dog, see p. 7, and n. 2.

^{page 63 note 3} On Y 12 a looking back.

^{page 64 note 1} Munich 2044. F.R. I, pl. 42: Pfuhl fig. 231: Buschor Greek Vase-Fainting fig. 93.

^{page 64 note 2} Stebbins Dolphin in Literature of Greece and Rome; list of dolphins on vases, pp. 97–115 (and p. 13, n. 20). Add to her East Greek list, besides this recently discovered piece, a small plate from Rhodes, Berlin, Arch. Sem. d. Univ. D 82. A decapod and, below, a dolphin. Full silhouette. The Fikellura dolphin is of the ‘Simple Curvilinear type.’

^{page 64 note 3} So the relief from Miletus, Berlin 1614 (Stephanos für Wiegand 8–9: Jacobsthal Orn. pl. 136 b).

^{page 64 note 4} As are the wings of the Boreads, B7 and X 1. The wing of the dog-headed monster, K 9, is in contrast scalloped.

^{page 64 note 5} For a ‘Rhodian’ example of the distinction between beak and forehead, cf. Louvre A 1007 (C.V.A. Louvre 1 pl. 21, 10). On the whole, in ‘Rhodian’ the angle is less pronounced. For another set of sphinxes and griffins see pp. 93–5.

^{page 64 note 6} PLATES 7 b (J 12), 8 e (K 1), 14 c (T 4).

^{page 64 note 7} N 94–6.89, G 24. J.H.S. 1924, 213 fig. 51: C.V.A. Cambridge 2 pl. 18, 45. See Payne Necrocorinthia 174 n. 1. To his list may also be added a partridge on the R.F. headpiece of a Clazomenian sarcophagus in Athens, and another on a late archaic relief from Thasos in Stamboul (no. 576); a banqueting scene—the partridge is placed below the stool on which a woman is seated. There is a Laconian plastic vase in the shape of a partridge, but without the barring on the breast, in Sparta; cf. infra p. 156 and PLATE 37 c.

^{page 65 note 1} B.M. B 106¹⁰; Tanis II pl. 26, 9. On the dating of the situlae see Rumpf, J.d.I. 1933. 66.Google Scholar

^{page 65 note 2} E.g. C 2. PLATE 4 c (of C 1) shows the body markings of this type, except that a long reservation has been added below.

^{page 65 note 3} E.g. B.M. 88.6–1.411; J.H.S. 1924, 183 fig. 3; Kunze, , A.M. 1934, 92–3 no. 7, Beil. 7, 1.Google Scholar

^{page 65 note 4} E.g. K 2, PLATE 9.

^{page 65 note 5} E.g. G 1, PLATE 15 a.

^{page 65 note 6} FIG. 2. See also pp. 93–4.

^{page 65 note 7} Once, D 1, two griffins and a deer. So too B1a, pp. 93–5.

^{page 66 note 1} Both forms on the same vase, J 6.

^{page 67 note 1} See p. 60, n. 3.

^{page 67 note 2} See p. 60, n. 2 and PLATE 19.

^{page 67 note 3} The fragmentary plate from Ephesos with man, spear and bush (Achilles ?) is near to the Perseus plate; it is in silhouette with reserved details (Stamboul 4156–8 and B.M. 1907.12–1.770; the B.M. fragments illustrated in Hogarth, Excavations at Ephesus fig. 53). In B.F. technique: Oxford 1925, 608, d. 1; fragment of a late ‘Rhodian’ dish; head of flautist; from Naucratis; C.V.A. Oxford, pl. 395, 18.

Outline heads: from Rhodes, B.M. A 749, the Euphorbos plate (Schaal Bilderhefte III pl. 5, 8; etc.); Berlin Inv. 3724, Rider plate (Prinz pl. 3 a; Neugebauer pl. 17). ‘Rhodian,’ B.M. A 721, fragment of dinos, from Naucratis (J.H.S. 1924, 193 fig. 21). Aeolian, e.g. Louvre B 561 (B.C.H. 1884 pl. 7; Cl.Rh. vi/vii 209 fig. 251). From Marseilles, Vasseur pl. 5, 5.

^{page 67 note 4} See p. 3, and n. 3–5.

^{page 67 note 5} PLATE 10 d.

^{page 67 note 6} PLATE 8 a.

^{page 67 note 7} B 7 (PLATE 10 c), J 12, K 8 (PLATE IO e). It is difficult to make out the exact form of S 12 (Fig. 6): it seems to be a krobylos.

^{page 67 note 8} So the Berezanxs satyr (K 1; PLATE 8 e), who has side locks also; and Dionysos (Y 12). On sphinxes the hair hangs loose: on F 2 (PLATE 14 a) the ends of the locks are shown.

^{page 67 note 9} E.g. François vase—Moirai and various female figures (F.R. I, pl. 1). Exekias, Amasis painter, Antimenes painter, Berlin painter, etc. Laconian, cup in Taranto of 530–520 (F.R. III p. 212, fig. in text). Caeretan, cf. Webster, J.H.S. 1928 nos. 11, 16 and 18.Google Scholar

^{page 67 note 10} As on the krater from Cyme in Aeolis, B.M. 1904, 6–1, 1 (R.M. 1888 pl. 6); and on several B.F. sarcophagi.

^{page 68 note 1} K 1 (PLATE 8 e), K 2 (PLATE 9), K 3, L 2 (PLATE 13), Y 12.

^{page 68 note 2} B 7 (PLATE IO c), K 8 (PLATE 10 e).

^{page 68 note 3} There are no signs of sexual excitation about any extant Fikellura figures—of the satyrs, of course, only the heads are preserved. This contrasts strongly with Clazomenian. Naucratite vase-painting shows a similar decorum, though here are parts of two phallus-vases in the B.M. (88.6–1.496 and bis: from Naucratis) and of one in Athens (Acropolis Museum).

^{page 68 note 4} Cf. H 1, J 1 and K3.

^{page 68 note 5} Castanets, k 5 (PLATE 8 c): kantharos, J 4: amphora, J 1 (PLATE 5 a).

^{page 68 note 6} E.g. on same vase, H 1.

^{page 68 note 7} Y 12, L 1 and L 3. See remarks in Catalogue.

^{page 68 note 8} PLATES 10 C and 8 d and FIG. 6.

^{page 68 note 9} B 7 (PLATE 10 c).

^{page 68 note 10} K 1 (PLATE 8 e).

^{page 68 note 11} PLATE 9.

^{page 69 note 1} See pp. 38, 58.

^{page 69 note 2} PLATE 14 b.

^{page 69 note 3} From Smyrna, Ö.Jh. 1932 Beiblatt fig. 92, 15, 19 and 7 (here together with 20a); J.H.S. 1924 pl. 8, 16. Apparently a late form. A fragment in the B.M., 86.4–1.1053, gives the form 20a with dots.

^{page 70 note 1} On Clazomenian, e.g. the krater from Cyme in Aeolis (p. 67, n. 10). Sculpture, a relief from Miletus, Berlin 1614 (Stephanos für Wiegand 8–9: Jacobsthal Orn. pl. 136 b).

^{page 71 note 1} Compare the fragment of an Ionian B.F. cup in Samos (p. 7, n. 2).

^{page 71 note 2} PLATE 18.

^{page 71 note 3} Hogarth Excavations at Ephesus pl. 49 and fig. 45. But these are unrepresentative of the pottery from Ephesos.

^{page 73 note 1} On P 7 the members are unjoined.

^{page 73 note 2} On neck of amphora, L 15, and a small amphoriskos, Y 10: on body, B 17, E 7, J 11.

^{page 73 note 3} E.g. Kinch figs. 131, 132. In late ‘Rhodian’ the form is still larger, e.g. ib. fig.

^{page 73 note 4} Attic and Corinthian crescents, independent or in whirligigs, are polychrome. It is therefore more probable that Laconian derives its crescents from Fikellura than from them.

^{page 73 note 5} On the other hand, the crescent occurs sporadically in the seventh and early sixth centuries elsewhere than in a whirligig. For instance, there is a rudimentary form on the outer zone of an early plastic shield from Samos (Technau, A.M. 1929 Beil. 15, 1Google Scholar; and ibid. 26, where the comparison with Fikellura is made). There are a few examples of the use of polychrome crescents in Attic; e.g. Graef I, 454–456, pl. 25, and on a fragment of a large kotyle in Eleusis. The precise derivation of the Fikellura crescent is, anyhow, of little importance, especially when one considers the simple nature of the ornament.

^{page 73 note 6} E.g. fragment of late ‘Rhodian’ plate from Naucratis, B.M. A 997: all the lower angles are filled with palmettes. A simpler type is found on some Vroulian cups, e.g. B.M. A 1004 (Kinch fig. 62). A related ornament on the shield of Menelaos on the Euphorbos plate (Schaal Bilderhefte III pl. 5, 8) has V-shaped copulas instead of palmettes.

^{page 73 note 7} So on C 1 (PLATE 4c), C 5, K 2 (PLATE 9) and K 3. On the belly below a lost main zone, C 3 (PLATE 17 ƒ), C 4, C 4a. On the shoulder, the zone below lost, Z 4 and 5. As a frame to the central panel of a Louvre Group oinochoe, S 9 (PLATE 17 9). But between two minor zones on E 7.

^{page 74 note 1} G 11, P 17. And on the lid of the new amphora from Kremasto, F 5.

^{page 74 note 2} So also Kunze, A.M. 1934, 100Google Scholar, where further instances are given of Laconian influence on East Greek art.

^{page 75 note 1} See p. 55, n. 3.

^{page 75 note 2} And so, occasionally, in Attic; e.g. on the Amasis amphora, Jacobsthal Orn. pl. 35 a. Here D 1, E 1, Q 1 and, repeated, K 9. Cf. the early ‘Rhodian’ squat oinochoai, e.g. Berlin F 295 (Neugebauer pl. 17 shows the general type of this group).

^{page 75 note 3} Zervos fig. 89. From Camiros.

^{page 75 note 4} See p. 43.

^{page 75 note 5} Kinch pl. 18, 3; 25, 5.

^{page 75 note 6} E.g. Louvre A 309: C.V.A. louvre 1 pl. 19, 6.

^{page 77 note 1} N 94–6, N 28.67. To be published in C.V.A. Cambridge ii.

^{page 77 note 2} E.g. Kinch fig. 89 etc., and an unpublished oinochoe of the Banded group in Gotha. On the Gotha vase the shoulder ornament is a chain of lotus flowers of normal Fikellura type and buds with arrow-shaped reservation; in the centre is a large lotus flower with enclosed outer petals, the space between which is filled in with diagonal chequer pattern. Cf. also the polychrome Bucchero oinochoe at Birmingham University (p. 57; FIG. 13, 11).

^{page 78 note 1} C 3 (PLATE 17ƒ) K 7 (PLATE 10 d).

^{page 78 note 2} See p. 55, and PLATE I 7 e.

^{page 78 note 3} E.g. Leningrad 15847: Izv. (G.A.I.M.K.) 5, pl. 12, 1.

^{page 78 note 4} PLATE 4 c

^{page 78 note 5} In the descriptions in the Catalogue, unless the contrary is stated, it is to be assumed that the flowers and buds are joined.

^{page 78 note 6} PLATES 17 b (U 6) and 1 b (W 6).

^{page 78 note 7} B.M. 1924.12–1.217. Fragment of a lid, from Naucratis or Tell Defenneh.

^{page 79 note 1} The Tell Defenneh amphora (B.M. 88.2–8.57: Tanis ii pl. 32, 4), of local manufacture (?), owes as much to Clazomenian as to Fikellura. The cable pattern on the shoulder is taken from Fikellura, the form of the scales and the plastic ring below the lip are rather Clazomenian.

^{page 79 note 2} J.d.I. 1886, 142.

^{page 79 note 3} See p. 36 n. 2.

^{page 79 note 4} So C 6, neat and small: Z 6, larger and clumsy. This motive, small and with T-shaped filling, is fairly common in late ‘Rhodian’ A; e.g. the amphora fragment in Munich (p. 55, PLATE 17 e).

^{page 79 note 5} Y 25, 34 and Z 7. On Y 36 instead of crescents, as the chevrons on Y 41.

^{page 79 note 6} FIG. 3.

^{page 81 note 1} Jacobsthal Orn. 43 is too elaborate.

^{page 81 note 2} Ib.

^{page 81 note 3} D 438. From Rhodes. Auktionskatalog Helbing, 28–30.10.1913, pl. 5 no. 90. Jacobsthal Orn. 179 (the shoulder ornament is drawn out). Berlin Inv. 2932, which Jacobsthal cites (p. 44, fig. on p. 45; from J.H.S. 1885, 181 fig. I—an unfair drawing) is not before the middle of the sixth century. Conceivably the vase with which the satyrs are busy is meant for Fikellura.

^{page 81 note 4} Schaal, Bilderhefte III pl. 5, 8. FIG. 15, 2.Google Scholar

^{page 81 note 5} Cf. Jacobsthal Orn., plates passim.

^{page 82 note 1} On L 5 a, filled with chevrons. The side ornaments may be omitted (L 5, L 18).

^{page 84 note 1} PLATE 10 e.

^{page 84 note 2} In ‘Rhodian’ the spine is attached to the spiral below. Cf. the right half of the handle-ornament of L 2 (FIG. 17, 2).

^{page 84 note 3} Jacobsthal Orn. 38 ff. discusses this ornament in detail. The ‘Rhodian’ origin is clear: an organic development can be traced for the motive from the pair of simple spirals of the type of B.M. A 460 (Kinch fig. 108) to the more substantial forms of the normal oinochoe (ib. fig. 87 c-e): d and e reproduced here, FIG. 18. In origin both the great volutes and the handle-ornament of Fikellura come from the simple spiral; the volutes by way of the palmette tree and the Euphorbos plate ornament (FIGS. 15, 1,2). The similarity of the curves of the spirals on the stelai is a natural expression of the current East Greek feeling for curves; there is no conscious imitation by vase-painters.

^{page 85 note 1} B 16, Y 30.

^{page 85 note 2} Graef I 446–449 (Boehlau no. 60).

^{page 85 note 3} Y group 1, at end.

^{page 85 note 4} Mostly casual finds from the island of St. Kiriak. Y 22 a was dredged up with miscellaneous pottery from the very end of the seventh century to the Roman period in the channel between the island and the promontory where the modern town stands. This material is now in Sofia and Burgas.

^{page 85 note 5} Mme. Lambrino La Céramique d’ Histria—Série rhodo-ionienne, Dacia III–IV 366, mentions ‘assez nombreux fragments.’ Prof, and Mme. Lambrino kindly let me view the unpublished finds.

^{page 85 note 6} K 2, L 6, N 8 are presumably from the Delta. At the head of the Delta one sherd from Memphis (P 23 c) and one from Abusir (T 8).

^{page 85 note 7} Of the relevant sites on the West coast of the Pontus only Istria has been excavated. Apollonia Pontica, to judge from casual finds, is rich in East Greek pottery from the end of the seventh century.

^{page 86 note 1} Pharmakowsky, A.A. 1914, 232.Google Scholar

^{page 86 note 2} Unfortunately no stratification is possible. The finds extend from about 620 to the Roman period. Gjerstad, E. (Studies in Archaic Greek Chronology, I: Naucratis, L.A.A.A. 1934, 67–84)Google Scholar attempts to use the stratification of the site of the temple of Apollo, as recorded by Petrie in Naukratis i. Unfortunately Petrie's classification of wares (op. cit. 17–20) is vague and obscure and he did not think it necessary to publish the context of individual objects. The data are, therefore, unsatisfactory, and the conclusions that Gjerstad reaches on the chronology of East Greek Pottery must anyhow be suspect. The Fikellura he dates after 550: this may well be true. But it is incredible that early Attic B.F. should be found in a late sixth-century stratum, or, when one considers the scientific excavations of Jacopi in Rhodes, that Naucratite A, ‘Rhodian’ and Ionian Bowls should persist to the end of the sixth century. The relative chronology is also impossible. If Gjerstad is right in his interpretation of Petrie's classes—and he would seem to be right—the strata of the Apollo site must have been seriously disturbed.

^{page 87 note 1} G 6, P 17 and Q,2 are illustrated in Tanis ii. B.M.: 1924.12–1.1060 goes with fragments (D 2) of known provenance.

There are certain extra numbers, not ordinary B.M. numbers, on some of these sherds. They occur on sherds known to have come from Tell Defenneh, but not on sherds known to have come from Naucratis. They may be dig numbers used at Tell Defenneh. I have listed sherds bearing such numbers ‘from Tell Defenneh (or Naucratis).’

^{page 87 note 2} There was no Corinthian: Payne Necrocorinthia 187. For general remarks on the dating of the Situlae, see Rumpf, J.d.I. 1933, 61Google Scholar: but they cannot yet be used to fix the chronology of Tell Defenneh.

^{page 87 note 3} Cf. Rumpf, Gnomon i 330Google Scholar and J.d.I. 1933, 60, for the dating and identification of Tell Defenneh. Herodotus is not definite. The position he gives for Stratopeda should not be far from Tell Defenneh (ii 154). The similarity of name favours Daphnae, and anyhow its site should also be somewhere near Tell Defenneh. Daphnae must lie on the strategical route from Asia, since the frontier fort is there (ii 30) and Sesostris is casually mentioned as returning by Daphnae (ii 107). A settlement of veterans might very well have been placed close at hand in support of the important frontier post of Daphnae. But if Daphnae and Stratopeda were situated closely together, it is indeed curious that Herodotus does not mention Daphnae in his description of the situation of Stratopeda. It is unlikely, to judge from the statements of Herodotus, that the two places should be equated. Lastly, it does not follow that on the concentration of the Greeks in Naucratis the Greek mercenaries were recalled from frontier duty. In fact the reason given for the depopulation of Stratopeda is that Amasis wanted them as a bodyguard in Memphis (ii 154), from which the inference is, that the Naucratis regulations did not apply, as one would anyhow expect, to the mercenaries. If then Tell Defenneh is Daphnae or indeed any frontier post, the finds of Greek pottery there may have no bearing on the date of the concentration in Naucratis. Tanis ii pl. 24, 6 must be a stray.

^{page 87 note 4} There are two or three earlier pieces.

^{page 88 note 1} Blinkenberg Lindos 284 nos. 992–994.

^{page 88 note 2} All the Fikellura pieces in Kassel, except for the amphora P 21, are from Boehlau's excavations. A few dull sherds went to Bonn.

^{page 89 note 1} A brief report by Technau, , A.M. 1929, 26.Google Scholar Lately considerably more and finer Fikellura sherds have been found.

^{page 89 note 2} It was ‘relatively frequent,’ Prinz 41.

^{page 89 note 3} In the Archaeological Collection of Vienna University. Fragment of neck of amphora with double cable pattern; from Keil's excavations.

^{page 89 note 4} By the Miltners. See Ö. Jh. 1932 Beiblatt 127–188. A selection of some hundreds of sherds from this campaign are now in the Archaeological Collection of Vienna University, where Prof. Schober kindly let me study them. There was one Fikellura sherd, which I would date shortly after the middle of the sixth century. It must, I think, be a stray. There are also fragments of two Attic band-cups, one of the third, the other of the last quarter of the century. The Fikellura piece is anyhow an import. Much of the finds from Old Smyrna, presumably local, seems Aeolian rather than ‘Rhodian.’This is not clear from the publication.

^{page 90 note 1} O 1.

^{page 90 note 2} See pp. 91–2.

^{page 90 note 3} See p. 56.

^{page 91 note 1} See p. 2, n. 1.

^{page 91 note 2} See PP. 3, 7.

^{page 91 note 3} All dedicatory, except perhaps that of Y 42a, the ΝΙΚΟ of Y 42 and the Δ painted on the base of U 8, which might be a potter's mark. On N 3 also the dedication was painted on the base, after firing. The rest are graffiti: the graffito on U 1 was not scratched till the oinochoe had been broken. The pieces with inscriptions are: B 3, N 3, R 10, U 1, U 8, W 12, Y 42, Y 42 a.

^{page 91 note 4} Boehlau 61.

^{page 92 note 1} But there is enough fin de siècle Attic B.F. in Rhodes, and even a Late Corinthian Warrior Aryballos with a large dent in the side.

^{page 92 note 2} So Kunze more definitely, A.M. 1934, 81 n. 4, 118.

^{page 93 note 1} PLATES 2 b, i ƒ and FIG. 8, 1.

^{page 94 note 1} Cf. fragments of an amphora neck, B.M. 1924. 12–1. 32 and 1087, from Tell Defenneh (or Naucratis): this may also be Fikellura. The piece from Samos, German Excavations (Technau, A.M. 1929 Beil. 12, 4Google Scholar) should now be Fikellura.

^{page 96 note 1} From the campaigns of Biliotti and Salzmann: G I, II; J 12; L 1, 2(?), 3, 5 a ^*, 7, 8, 9^*, 13(?), 14, 15^*, 19(?); M 1, 2^*, 3, 3 a ^*, 5(?); N7; O3,6^*, 7,8; P 3, 4, 6, 7,8, 10^*; Q,7(?),9^*, 13(?), 14; R 1, 5a ^*; S 1 (?); Y 6, 10, 14, 24, 26^*, 27, 28, 31, 42.

^{page 97 note 1} See p. 87, n. 2.