¹ The traditional date of the foundation of Cumae, in so far as it can be treated as serious historical evidence at all, can only refer to some westward movement of the Greeks in the period of migrations. Eusebius' (Jerome's) words, ‘Mycena in Italia condita vel Cumae,’ perhaps indicate a confusion of the tradition that Cumae was the oldest Greek colony in Italy and Sicily with that of the migration of Halesus and the Argives to Falerii and Alsion. An archaeological terminus post quem for the foundation of the Greek colony is provided by the presence of two Greek geometric cups (Monumenti Antichi xxii, pl. xviii, nos. 7 and 9) in the graves of the native settlement which gave place to the Greek colony. These cups can hardly be earlier than c. 800 B.C, and consequently the foundation of the Greek colony must be placed after that date. (Gabrici's archaeological chronology has been completely discredited by the work of Johansen, Les vases sicyoniens). A terminus ante quem is provided by the contents of the earliest colonial graves at Syracuse. Protocorinthian pottery of the globular-Aryballus period is comparatively common in the Greek graves at Cumae, while at Syracuse it is confined to the earliest tombs. (Nos. 223, 312, 466, N. d. Scavi, 1895.) A date for Cumae in the period c. 775 to c. 750 B.C. seems to be indicated, and this agrees well enough with Strabo's statement that Cumae was the oldest Greek colony in Sicily and Italy (Strabo 243). The first Greek settlement in Pithecusae (Strabo, 247; Livy, viii, 22) was probably earlier, as Livy indicates, but there is as yet no archaeological evidence in support of this available from Ischia.

² Annual of the British School at Athens, 1933/4.

³ Pl. XX, nos. A1, A2. Montelius, La civilisation primitive en Italie, pl. 260, nos. 5 and 6. Payne, Necrocorinthia, p. 4, note 2, recognised the first as Boeotian or Cycladic.

⁴ N. d. Scavi, 1916, p. 217, fig. 26. Recognised as a Cretan import by Payne, Necrocorinthia, p. 4, note 2.

⁵ Pl. XXI, no. A3. N. d. Scavi, 1914, p. 333, fig. 24. Recognised as a Cretan import by Payne, Necrocorinthia, p. 4, note 2.

⁶ (Tomb CVII.) No. 15265 in the Villa Giulia. Cf. Dugas, , Delos, xvGoogle Scholar, pl. xxvii, nos. 24 and 31.

⁷ Tombs 779 and 785 in the Vila Giulia.

⁸ Villa Giulia, No. 5642.

⁹ Montelius, op. cit., pl. 290. One vase is illustrated in pl. XXI, no. A5 (‘The Warrior's Tomb).

¹⁰ These vases are in the Cerveteri room in the Louvre under the general heading of ‘Vases de style géometrique trouvés en Italic’ No. 18, a Cycladic Geometric amphora, pl. XX, no. A4 (Pottier, Vases antiques du Louvre, pl. xxix, D.18), is certainly from Cerveteri, and so also is one of the Cretan examples.

¹¹ So far as our evidence goes, the importation of Greek Geometric pottery to Etruria antedates its importation to Sicily by a period of at least fifty years. In Sicily there seems to have been a long gap between the last Mycenaean import and the arrival of the first Greek Geometric vases at about end of the ninth century. The Greek imports to Sicily of a date before the foundation of Syracuse are also far fewer than those to Etruria in the same period. See BSA, 1933–4, p. 170 ff.

¹² The more successful vases of this class are, of course, more easily recognisable as local than as Barbarian. For example, the drawing of no. C8 (pl. XX) is in no way inferior to that of many Greek Geometric vases. It may be that I have included in Class C certain vases which are the work of Greek hands and should have been placed in Class B, but the existence of this last class seems to me to be of such great historical importance that I do not want to run the risk of bringing it into disrepute by including in it any members of dubious origin.

¹³ Other examples of Class C:—N. d. Scavi, 1928, pl. ix, Capodimonte. (The drawing of some of these examples might well be pure Greek work, but the local hand is betrayed by the barbarising shapes.) Op. cit., 1914, p. 320, fig. 13 and p. 312, fig. 6, Vetralla. (For the Greek model of the latter, cf. Dugas, , Delos XVGoogle Scholar, pl. xxvii, no. 31. Sedulous but heavy-handed imitation.) Corpus Vasorum Antiquorum, Musée Scheurleer i, ‘Style Italo-Géometrique,’ IVb–IVc, pl. i, 1. ‘Rome.’ Pl. XX, no. C8. (Either a Cretan import or local work closely modelled on a Cretan original.) Nos. 4432 and 4433 in the Villa Giulia, Falerii. (Very careless work.) N. d. Scavi, 1907, p. 231, fig. 33, Tarquinia.

¹⁴ There is, so far as I know, no example of a Lydian ‘Lydion’ dating from as early as the eighth century B.C., but the resemblance of this early Etruscan example to the Lydian vases of the seventh (?) and sixth centuries is very striking indeed, and makes very tempting the hypothesis that both the eighth-century Etruscan, and the seventh—sixth century Lydian examples, derive from a common Lydian archetype.

If this hypothesis is correct there is thus some slight additional evidence of Lydian (not merely Oriental) commercial contact with Etruria at least as early as the eighth century B.C. [I am, of course, not concerned with the theory of the Lydian origin of the Etruscans. The literary tradition can only be made to agree with the archaeological evidence by doing violence to its implied chronology, and the bulk of the archaeological evidence in itself only supports Oriental (and Greek), as opposed to the more specific Lydian, importation (and perhaps immigration) in the ninth and eighth centuries B.C]

¹⁵ This does not, of course, include imported ‘Lydia’ of the sixth century such as the Laconian III ‘Lydion’ from Orvieto at Philadelphia and the late sixth-century Lydian ‘Lydion’ from Cerveteri in the Villa Giulia.

¹⁶ No. 13 in the Cerveteri Room in the Louvre.

¹⁷ E.g. Montelius, op. cit., pl. 214, nos. 5 and 6, from Chiusi, pl. 206, no. 25, pl. 207, nos. 1 and 9, pl. 210, no. 4, from Pitigliano, pl. 291, no. 5 from Tarquinii and N. d. Scavi, 1914, p. 323, fig. 16 from Vetralla.

¹⁸ The possibility of this class having been made by Greeks at Cumae is excluded for the following reasons:—

(a) Class B is of earlier date than the foundation of the Greek colony at Cumae: that is to say, vases of the style of Class B have not been found in even the earliest graves of Greek Cumae, while late examples of Class A (which is contemporary with Class B) have been found in the Barbarian settlement which was supplanted by the Greek colony.

(b) Cumaean clay, as known to us from the late eighth- and seventh-century pottery of that colony, is different from that of any of the examples of Class B.

There remains the possibility that these vases were manufactured in the Greek settlement at Ischia, which may have antedated the mainland colony at Cumae, but this possibility is practically excluded, for at least some members of the class, by the fact that their clay is characteristic of the part of Etruria in which they were found.

¹⁹ The drawing of this vase in Randall-Maclver, Villanovans and Early Etruscans, pl. xi, no. 12, gives but a poor idea of its shape and practically no idea of the essentially Greek character of the drawing.

²⁰ The rival claims of the pre-Cumaean Greek immigrants to Etruria and of the Greek settlers at Cumae to be the transmitters of the alphabet to Etruria are discussed below, p. 138 f.

²¹ I have purposely avoided attempting a full chronological classification of these pre-Cumaean Greek influences in Etruria because we lack as yet a detailed authoritative chronological classification of Greek Geometric art earlier than c. 800 B.C. from the hands of any Greek archaeologist of the first rank. I am, however, forced to attempt a rough sketch of their chronology because of Schachermeyr's statement that the first appearance of Greek merchants in the Tyrrhenian sea and of Greek Geometric pottery in Etruria can be dated from the period c. 820 to c. 800 B.C.

Schachermeyr's great work, Etruskische Frühgeschichte, is not only far the most convincing exposition of the theory of Oriental origins but also makes far fuller use of the Greek archaeological material than any of its predecessors. It is with full appreciation of the value of that work that I venture on the following criticism of this conclusion, a criticism which is in part based on archaeological evidence not available to Schachermeyr.

The publication of the material from Arkades Annuario, x–xii) and the series of burials in the Geometric tombs at Cnossos, excavated by Payne, Brock and myself in 1933 and 1935, have provided us with a continuous stylistic series which makes the stylistic dating of Cretan Geometric by fifty-year periods reasonably secure. This is also true of Corinthian and, to a lesser degree, of Argive Geometric, thanks to the temple deposit of Hera Akraia at Perachora and the work of Payne. Unfortunately we are not nearly so sure of our ground with Cycladic Geometric; but the publication of Délos xv and the context of Cycladic exports at Perachora, Cnossos, Locri, Lentini and Finocchito (for the last three places see BSA, 1933/4, pp. 176–180, 184–191) make the dating of these fabrics much easier than it was six years ago. It is on the basis of this new knowledge that the following chronological attributions of the earliest Greek influences of the pre-Cumaean period have been made.

Four vases of Class A from Cerveteri, the Cycladic amphora (Pottier, pl. xxix, D. 18) and three Cretan Oinochoai (nos. 16 and 17 and one unnumbered in the Cerveteri room in the Louvre) are certainly not later than c. 850 B.C. and may very well be much earlier.

One vase of Class B (no. 2) from Tarquinia (Poggio di Selciatello-Sopra. Tomb 160–56. Excavations of 1905), whether judged by its place in the stylistic development of Greek Geometric art, or by the early ‘Second Benacci’ character of the cemetery, can hardly be later than the first half of the ninth century.

One vase of Class C, pl. xx, C8, from ‘Rome,’ which betrays strong Cretan influence (CVA, Musée Scheurleer, i, IVb–IVc, pl. i, 1), and four vases from Capodimonte (N.d.Scavi, 1928, pl. ix) imply the importation of Greek originals of mid-ninth century date.

One vase of Class B, pl. XXI, B3, from Falerii is in the Protocorinthian style of the period c. 850 to c. 800 B.C.

It will be noticed that these earliest Greek influences are confined to sites whose Etruscan foundation Schachermeyr dates to his first Etruscan immigration in the year c. 1000 to c. 950 B.C., or to sites (with the possible significant exception of Falerii) penetrated by the Etruscans before their second migration, which he dates c. 810 to c. 800 B.C. In fact, his error as to the date of the first appearance of Greek influences in Etruria does not affect his theory of a double wave of Etruscan immigration. But the error is nevertheless a serious one for an estimate of the cultural origins of Etruscan civilization.

Greek influence on Etruscan civilization and individual Greek immigration into Etruria begin, then, not later than the first half of the ninth century B.C.; Greek importation perhaps even earlier. How far this throws any light on the traditions of Greek Heroic Migrations to the West, such as Evander and the Arcadians at Pallanteum (Livy, i, 6, 7. Ovid, Fasti i, 471; v, 99. Dionysius, i, 31–33. Vergil, Aeneid viii, 355), the Thessalians at Caere (Dionysius, i, 16. Pliny, NH iii, 18. Strabo, 220, 226. Servius, ad Aen. viii, 479; x, 183) and Tarquinii (Justin, xx, 1), and Halesus and the Argives at Falerii (Dionysius, i, 17. Ovid, Fasti ix, 73. Cato ap. Pliny, NH iv, 8. Servius, ad Aen. vii, 695. Steph. Byz. s.v. (Justin, xx, I, Chalcidians)) and Alsion (Silius Italicus, viii, 476), I must leave to those who are more competent than I to weigh the historical value of semi-mythical tradition. Certainly local patriotism and antiquarian philhellenism can no longer be regarded as adequate explanations of every Roman tradition of a Greek origin for an Italian city. Compare Strabo 214, on the Greek origin of Spina, with the discoveries (N.d.Scavi, 1927) in the swamp district of Comacchio.

²² For the chronological value of the foundation dates of Syracuse and Gela and the contents of the earliest colonial graves from those sites, see Johansen, Les vases sicyoniens, pp. 179–185.

²³ For the chronology of the end of the Protocorinthian style see Payne, Necrocorinthia, pp. 21–27.

²⁴ For the chronology of the end of the Early Corinthian style see Payne, Necrocorinthia, pp. 55–57. The foundation date of Marseilles, 600–598 B.C. (Pseudo-Scymnus, 209 ff.; Eusebius), and the archaeological evidence from that site (Jacobsthal, ‘Gallia Graeca,’ Préhistoire, ii, fasc. 1) give strong general confirmation of Payne's chronology at this point. There is very little pottery earlier than Middle Corinthian, and that little can be accounted for by Aristotle's indication (fr. 503 R = Athenaeus, 576a) that the Phocaeans had traded with the neighbourhood before founding their colony.

²⁵ For the date see p. 129, note 1.

²⁶ It is a fact of which the historical significance is but seldom appreciated that there is less evidence of Greek trade with Rome in the fifth than in the sixth and even in the seventh centuries B.C. While Rome was under Etruscan rule she was worth the attention of Greek merchants; the decline which followed the winning of her independence temporarily destroyed her value as a market for Greek goods. (It is possible that the immediate cause of the falling off in Greek trade with Rome at the end of the sixth century B.C. was Rome's first treaty with Carthage (of late sixth century date ?), but this explanation is made improbable by the fact that the Carthaginian-Etruscan alliance of the time of the battle of Alalia seems to have had little effect on Greek commerce with Etruria.)

²⁷ The doubtful example is no. C 8, said to be from Rome. For the distribution of pre-Cumaean Greek imports and influences see Map (fig. 17).

²⁸ There is, perhaps, less archaeological evidence of Greek craftsmen working in Etruria after c. 775–750 B.C. than would be expected from the evidence of their activity in the previous period. See, however, below, pp. 144–147. There is plentiful evidence of their work at Cumae.

²⁹ There is nothing new in this conclusion, but as the evidence for it has often been confused by the admission of irrelevant and partially relevant material and by faulty dating of the inscriptions concerned, I have stated my case for it in full.

³⁰ Johansen, pl. xv, no. 5. Facsimile of the inscription in Roehl, Inscriptions Graecae Antiquissimae, no. 524, see pl. iii, no. 2.

³¹ MA, xxii, p. 361.

³² ‘Chalcidian’ vases can be dated to the period c. 575 to c. 525 B.C. See Rumpf, Chalkidische Vasen. In a recent paper (California Publications in Classical Archaeology i, 3) Smith argues for the manufacture of these vases by a settlement of Greek potters at Caere. I am not convinced by his arguments for Caere, but the evidence he collects makes a very good case for their manufacture at some Chalcidian colony in the West. If further evidence becomes available in support of this view, the inscriptions on these vases may form a valuable series of documents for comparison with sixth-century Etruscan and Latin inscriptions, but even so they can throw but little light on the problem of the origin of their alphabets.

³³ Inscriptions Graecae Antiquissimae, nos. 372–376.

³⁴ Op. cit., no. 525.

³⁵ Montelius, La civilisation primitive, pl. 186, 10 a–d.

³⁶ Montelius, op. cit., pl. 189, no. 11.

³⁷ Pl. xxii, no. 1. Memoirs of the American Academy in Rome, iii, pl. 3, nos. 3–5. Röm. Mitt., 1887, p. 37. Montelius, op. cit., pl. 370, no. 3.

³⁸ E.g. Randall-Maclver, ‘A little after 800 B.C.’ He fails to recognise the Protocorinthian cup (which he describes as Geometric) which appears in pl. 21 of Villanovans and Early Etruscans, and also the Protocorinthian influence on the Bucchero cups on the same plate.

³⁹ Johansen, p. 89. To this must probably be added the Oenochoe, from which the paint has disappeared.

⁴⁰ The tomb in which the stele was found had been completely plundered.

⁴¹ Probably native Etruscan. Compare the axe from Poggio Pepe, N. d. Scavi, 1895, pp. 304–6.

⁴² Röm. Mill., 1887.

⁴³ The chronology of Greek and Italian fibulae is still most uncertain and cannot be compared with that of Greek vase painting and sculpture.

⁴⁴ Montelius, La civilisation primitive, pl. 188, 3.

⁴⁵ Montelius, op. cit., pl. 370, 4 a and b.

⁴⁶ Montelius, op. cit., pl. 224, 5 a and b.

⁴⁷ Fragments of a Protocorinthian cup of the Sub-geometric class are assigned to the Bernardini tomb by Curtis (Memoirs of the American Academy at Rome, 1919). (See also Johansen, p. 183.) Randall-MacIver (Villanovans and Early Etruscans, p. 219) suggests that the evidence for this is doubtful. The gold cup (Montelius, pl. 370, 5) clearly derives from Protocorinthian work of the early seventh century, and the bronze relief (Montelius, 367. 6) from Greek Daedalic sculpture of the same period. The Bernardini Tomb thus cannot be earlier than c. 700 to 675 B.C.

⁴⁸ The Italo-Protocorinthian vases of this tomb derive from Protocorinthian of Payne's Late Protocorinthian and Transitional styles (Necrocorinthia, pp. 16–34.) The Greek imports are Transitional, i.e. of c. 640 to c. 625 B.C. date.

⁴⁹ Perhaps earlier.

⁵⁰ See pl. XXII, la and 2a; 2a is the Manios inscription written in the letters of the Tataie Aryballus.

⁵¹ Not later. For, if it were of sixth-century date. it would presumably show the influence of the sixth-century ‘Chalcidian’ alphabet. Further, there is, so far as I know, no example of a fibula of this type in a sixth-century context.

⁵² There is, of course, the third possibility that the Etruscan alphabet, though deriving in the main from that of Cumae, was also influenced by other Greek alphabets not before c. 775–50 B.C. (i.e. after the foundation of Cumae). I find it difficult to believe that such a peculiarity as the letter San, and such an irregularity as the form of the Etruscan Koppa could find then way into the Etruscan alphabet at the same time as, and in spit of, the strong Cumaean influence. San and the irregular Koppa are intelligible as survivals. They are unintelligible as by-blows.

⁵³ For the mixed origin of the settlers at Cumae see Pseudo-Scymnus, 238; Strabo, 243 and 247; Dionysius, vii, 3; Livy. viii, 22.

⁵⁴ That the Greek alphabet was in existence in the ninth and eighth centuries B.C. is (pace Rhys Carpenter) proved by (a) the Hymettus Geometric inscriptions (Blegen, AJA, 1934, 10); (b) the δς νûν ὀρχηστῶν inscription on the Dipylon jug; (c) possibly (the evidence is, I think, doubtful), the newly discovered inscriptions at Corinth (Stillwell, AJA, 1933, 605). (a) and (b) show not merely that Greek writing was in existence in the ninth and eighth centuries but also that it was already used for frivolous purposes.

⁵⁵ Bulletino di Paletnologia Italiana, xx, pl. iii, 8. BSA, 1933/4, p. 191. From Finocchito. The graffito is found on a vase which belongs to a class unrepresented in the Greek colonial graves in Sicily; it may even be older than the foundation of Cumae, where the class is likewise unrepresented.

⁵⁶ For imports see p. 130, and notes 4, 5, 10, and for vases of Class C, p. 131, note 13.

⁵⁷ MA, vol. xxii, pl. xxxvi, 2, xl, 2, xl, 7, xli, 6, xlii, 4, xliv, 5, xlix, 2, p. 471, fig. 172; vol. xiii, pp. 272, 273 and 274; Cretan imports. For an example of Creto-Cycladic, see MA, xxii, pl. xxxv, 2.

⁵⁸ See, however, Kunze, Kretische Bronzereliefs, p. 270. ‘Kesselattaschen’ from Praeneste and Vetulonia.

⁵⁹ Kunze, op. cit., pp. 236–237.

⁶⁰ The problem of the origin of much of the Orientalising metal-work of Etruria of the lateeighth and early-seventh centuries is still unsolved; but since Poulsen's Der Orient und die frühgriechische Kunst the wholesale attribution of all examples and of all the varieties of style to the Phoenicians and immediate Phoenician influence has become impossible. The silver bowls of the Amathus bowl type from the Bernardini and Regolini-Galassi tombs (Montelius, pl. 367, 3, 8a and 8b; pl. 368, 5; pl. 369, 7a and b; pl. 338,1–5), and the example of a more primitive type from the Tomba del Duce (Montelius, pl. 187, 10a and b) are probably Phoenician imports, but the influences on such work as Montelius, pl. 367, 1a–c, 6; pl. 364, 7, 8, 12; pl. 365, 13a; pl. 369, 1, 8; pl. 370, 6, 7 (from Praeneste); pl. 332, 9, 10a–b; pl. 336, 10a–b; pl. 339, 10–17; Pl. 340. 5a–d, 7; P1. 341. 1–15b; pl. 335, 1, 2, 5, 8 (from Cerveteri); pl. 188, 1a–c; pl. 179, 2; pl. 193, 1–5; pl. 194, 11a–b; pl. 201, 1–5; pl. 202, 6a–b (from Vetulonia); pl. 266, 8b–c; pl. 267, 4; pl. 261, 1a–b (from Vulci); pl. 217, la–b (from Chiusi); pl. 294, 1, 10a–b, 11; pl. 295, 3 (from Tarquinia); pl. 327, 13 (from Falerii); pl. 173,4 (from Cortona) are obviously very complex.

Kunze's discovery is exceedingly important as showing how the strongly ‘Assyrian’ type of the Barberini ‘Kesseluntersatz’ came to Etruria not direct from the Orient but through the medium of Cretan art.

⁶¹ To those artistic influences may perhaps be added the introduction to Rome, and possibly also to Etruria, of the Cretan god Fελχανός See Rose, , JRS, XXIII, 1933, 48–50Google Scholar and 62.

⁶² During the period c. 735–690 to c. 600–575 B.C. I know of no site in the western Mediterranean, Greek or Barbarian, whose Greek pottery imports are not mainly Corinthian. In the Annual of the British School at Athens, 1933/4 I have discussed the possible implications of this fact for the history of the first wave of Greek colonisation in Sicily and Southern Italy, and for the relations of Corinth, Chalcis, Eretria and Megara at the end of the eighth, and for Corinth and Miletus at the end of the seventh centuries B.C. The distribution of Corinthian exports to western sites in the period c. 734 to c. 650 B.C. is shown in the map on p. 139 (fig. 18). For the period c. 650 to c. 550 B.C. see the list in Payne, Necrocorinthia, pp. 188–9. Of the distribution of Italian imitations of Protocorinthian and Corinthian, I can only say that I do not know of any seventh-century site in Etruria where this fabric is not well represented. For the rarity of other Greek fabrics in the Western Mediterranean during the years c. 735–690 to c. 600 B.C., see BSA, p. 204, note 1. The little East Greek pottery exported to Etruria in this period is only enough to mark the contrast with the large quantities of Corinthian. Indeed, as a generalisation, it can be said that in this period Corinthian imports are far more numerous on all Etruscan sites than those of all other Greek states put together. Statistics as to the number of Corinthian vases found on any given site are seldom available, and when available of dubious value. For Protocorinthian see Johansen, pp. 18–19, 88–90, but to his lists must be added many vases discovered since the publication of Les vases sicyoniens and a few corrections made by Payne. The greatest quantities of Protocorinthian and Corinthian in Etruria come from Tarquinia, Cerveteri and Vulci. The last two places ‘have probably produced more Corinthian vases than any other Italian sites’ (Payne, p. 189).

⁶³ Both Etruscan metal-work and Etruscan relief bucchero of the seventh century often show signs of the immediate influence of Argive-Corinthian bronze-work.

⁶⁴ Payne, Protokorinthische Vasenmalerei, p. 24.

⁶⁵ Johansen includes as Cumaean two Cretan imports to Cumae. MA, xxii, pl. xliv, 5 and xlix, 2.

⁶⁶ By far the best short sketch of this class during the period c. 640 to c. 550 B.C. is the incidental discussion of Payne. (Necrocorinthia, pp. 206–209.) It is indicative of what might be accomplished by a comprehensive and detailed study that Payne in his incidental treatment has isolated the work of two painters in the ‘dot-rosette’ group. (Necrocorinthia, p.206.)

⁶⁷ This class also contains a certain number of hybrids with drawing derived from Corinthian vasepainting and shapes from East Greek imports such as Montelius, pi. 344, no. 3, but most of the examples of this type belong the sixth century.

⁶⁸ The Class iii examples from Falerii are Villa Giulia, nos. 4942, 4971, 4742. No. 4942: clay, dirty white; paint, dark brick-red. (The clay and paint of this vase are very similar to those of Villa Giulia, nos. 5012, 5013, 5014, which are undoubtedly of Faliscan workmanship. Nos. 5013 and 5014 are illustrated in Montelius, pl. 326, nos. 9 and 11.)

No. 4971. Clay, chalky white; paint, dark brick-red. (The clay and paint of this vase are exactly the same as those of Villa Giulia, nos. 4432 and 4433, which are the crudest possible examples of Faliscan work.)

No. 4742. Clay chalky white; paint dirty brownish-black. This vase might be a Cumaean import as far as its paint is concerned, but the clay is indistinguishable from that of no. 4742.

No. 4791, apart from clay and paint, compares very favourably with the Protocorinthian imports to Falerii. (Villa Giulia, nos. 5006 bis, Montelius, Pl. 326, 8, and the two examples numbered 5008 in Johansen, pl. xix, 1.) The drawing and shape are more competent than those of the average Cumaean product, and the vase is very possibly the work of a true Corinthian craftsman.

The large amphorae and Oenochoai from Falerii in the Villa Giulia are mainly of Faliscan work manship with a few Cumaean imports and perhaps one example which is both local and Greek.

Of the possibility of the existence of vases of Class iii from Tarquinia and Cerveteri, I can only say that not all the Cumaean vases from Tarquinia and its neighbourhood are of the same clay, and that at least two Oenochoai from Cerveteri in the Louvre are in shape and decoration characteristically Greek though made from the same clay as undoubtedly Barbarian vases. For Greeks at Caere in the seventh and sixth centuries see Strabo, 220, 226.

⁶⁹ Polybius, vi, 2, 10.

⁷⁰ Dionysius, iii, 46. Cicero, De Re publica ii, 19–20. Livy, i, 34, iv, 3. Strabo, 219 and 378. Pliny, NH xxxv, 16 and xxxv, 152.

⁷¹ It is even said that the story is chronologically impossible. So, indeed, it is, if Beloch's reconstruction of the house of Cypselus is accepted; but not otherwise. Of Pais' interpretation of the Tarquins, all that need be said has been well said by Ure (Origin of Tyranny, pp. 236–240). The theory of the Tarquin story as ‘ Herodotus translated into Latin ’ does not really affect the Demaratus tradition.

⁷² cf. Busolt, GG 1², 640.