¹ It is interesting to put on record that the clay of our terracotta corresponds closely to the ashen-grey earth of Eretria described by Dioscorides v. c. 170 (ed. Sprengel): Also Pliny, , Hist. Nat. xxxv. 54Google Scholar: ‘Namque et alba est et cinerea, quae praefertur in medicina. Probatur mollitie: et quod si aere perducatur, violaceum reddit colorerm.’ In view of this, it may be apposite to raise the question whether there was in Eretria in the fourth century a manufacture of terracottas dependent on those of Attica and Tanagra, yet distinct from them, such as has been already assumed for the fifth century lekythi (cf. Staes in Ἐφημ. Ἀρχ. 1894, p. 65). In the National Museum at Athens are some fifty terracottas from Eretria, all without exception from the excavations of the last seven years. With the exception of a vase in the form of a negro's head (published by Hartwig in Ἐφημ. Ἀρχ. 1894, pl. 6, p. 121), not one of them has as yet been made known, although some may rank as masterpieces. None are over one foot high, and compared with them our terracotta ranks very high. See Δєλτίον, 1889, pp. 74, 115, 171; and compare also a terracotta published by Miss Hutton in the current volume of this journal, pl. 4, fig. 2, p. 134.

² On Eros and his sepulchral signification see Furtwaengler in Roscher's Mythol. Lexicon, art. Eros; Collignon in Dictionn. des Antiq. art. Cupido; Pottier, and Reinach, S., Nécropole de Myrina, index s.v. Eros, and especially p. 329Google Scholar; Benndorf, , Bullet. municip. 1886, p. 70ffGoogle Scholar.

³ Published by Miss E. Sellers in Furtwaengler's Masterpieces, pl. 18.

⁴ This Eros in position and action of the arms corresponds minutely with the Dresden Artemis, which Furtwaengler, (Masterpieces, p. 324, fig. 139)Google Scholar regards as a product of the older period of Praxiteles. The head is gently inclined to balance the rigid left leg, and the bow is held in his left hand, while with the right he draws up his chlamys behind the shoulder just as the Diana of Gabii, which Studniczka, (Vermuth. zu gr. Kunstgesch. p. 18ff.)Google Scholar has identified with the Artemis Brauronia of Praxiteles. The type of countenance connects it with the Eros of Centocelle, as also do the long curls of hair, the relatively short wings, and the similarity of the whole conception (for instance, he makes no use of his weapons, but bends his head in reflection). Besides the three well-known examples we can point to a fourth Eros from the hand of Praxiteles, namely, that which Verres stole from the house of Mamertinus Heius at Messana, where it had been preserved for many generations. Cicero, in Verrem, iv. 24, calls it similar to the Eros of Thespiae: ‘idem prius artifex (= Praxiteles) eiusdem modi Cupidinem fecit illum qui est in Thespiis, propter quem Thespiae visuntur, nam alia visendi nulla causa est.’ If then the Centocelle statue represents the Thespian Eros, the fourth and similar Eros-type may be preserved to us in the above-mentioned Pompeian statue. In this case it would be an older work of Praxiteles, and as it were introductory to his later Thespian masterpiece.

⁵ For the history of Eretria, see Richardson, in American Journ. of Archaeol. 1891, p. 234ffGoogle Scholar.

⁶ I know it only from Reinach, S.'s Chroniques d' Orient, p. 336Google Scholar; in 1892 it was not exhibited, and it is not described in Kavvadias' vol. i.

⁷ On the influence of Lysippian art on the Smyrna terracottas, see Reinach, , Mélanges Graux, p. 156ff.Google Scholar; also Necropole de Myrina, p. 159.

⁸ The sculptured column from Ephesos, (J.H.S. xi. p. 280)Google Scholar and a terracotta, of unknown provenience (Lecuyer, 2^mecollection, pl. 13). The terracotta, if genuine, must be connected with the Ludovisi Ares.

⁹ I can recall only the following: (1) two gems in the Brit. Mus. (Cades, Abdrücke, Nos. 673, 681); the style of both is that of the fifth century; (2) a gem of severe style in Gerhard, Über den Gott Eros, pl. I., 8. In each case Eros is a perfectly mature youth, but how far the type approximates to Hermes, the small scale of the design forbids me to say.

¹⁰ I quote only examples from the British Museum: E 191, E 297, E 571, Eros with closed extended legs; E 13, E 187, E 293 (rev.), E 307, E 388, E 464, with bent knees; E 126, E 189, flying diagonally; E 293 (obv.), flying horizontally; E 129, F 37, upper part of body bent backwards. Flight correctly rendered, with or without aid of drapery: E 231 (?), E 241, and among Italian vases, F 138, 184, 306, 310, 314, 315, 399, 400, 458. Even on the Peleus-vase E 424, which in many respects shows a remarkable advance on the latest fifth century vases, Eros still flies without inclining his head, or bending the hips outwards, or having drapery underneath him.

The flight of Nike had as a general rule the same development. We see her flying diagonally on B 357 and E 584; with bent-in knees, without aid of drapery, E 179, 287, 453, 469, 513. But on the vases of the fifth century she already appears in natural and easy flight, as on E 312 and E 406; cf. E 432, E 498. This is explained by the fact that her drapery was at an early period applied to this purpose of making her flight as easy as possible. The Nike of Paeonios is another example; she cannot however be compared with those above-mentioned, as she has one foot resting on the ground. i.e. the base, and consequently does not fly clear.

¹¹ Cf. above all the Dictionary of Architecture issued by the Architectural Publication Society, s.v. Camera and Lacunar (with reference to ancient authorities); also Weale-Hunt, , Dict. of Terms used in Architecture, 4th edition, 1876Google Scholar; Durm, , Baukunst der Römer, p. 283Google Scholar.

¹² The former is proved from Helbig, , Untersuch. uber d. campan. Wandmalerci, p. 132aGoogle Scholar, 2. A classical example of the latter is a lacunariam in the Brit. Mus. from the Nereid monument, on which a head is painted in full face. Six, (J.H.S. xiii. p. 133Google Scholar) has adduced this in support of his view that this monument belongs to the first half of the fourth century, but his arguments are not convincing; and I am still persuaded that the monument dates from the fifth century. Moreover Six's interpretation of the B. M. astragalus vase is not to my mind satisfactory. Rather than Aurae and Aeolus we should regard the figures as nine personified astragali (ψῆψοι), by means of which the figure on the left (much restored) declares oracles. Cf. a similar relief from Sagalassos in Lanckoronski's Städte Pamphiliens u. Pisidiens, figs. 109, 110, and for the astragalus oracle, Petersen ibid.

¹³ I see no ground for rendering pucros ‘children’; I take it as used in contradistinction to barbati.

¹⁴ On the vases of the fifth century a flying Eros frequently occurs, with the lyre held downwards in his left hand, a sash held up in his right, e.g. E 191 in Brit. Mus. Another Eros playing the lyre and flying downwards is seen on the lekythos published by Benndorf (v. supr.), and on E 126 in Brit. Mus.