Anyone who travelled in Albania before the war, as I did, is aware that the conditions were not favourable for excavation or archaeological research. The greatest credit is due to the Albanian archaeologists, Frano Prendi, Selim Islami, and Hasan Ceka, for their extensive and systematic excavations and for the reports which have been published in Albania. Where I differ from them in the interpretation and the dating of their discoveries, I do so in no critical or captious spirit but in gratitude to them for their services to European archaeology.

¹ I am grateful to Mr. S. Foltiny of the Institute for Advanced Study at Princeton who drew my attention to the report of an international conference in Yugoslavia at which mention was made of this periodical. I found some numbers of the periodical in the New York State Library and in the Library of Congress, and subsequently others in the Cambridge University Library, the London School of Slavonic Studies, and the Taylorian Institute in Oxford; these scattered numbers do not constitute a complete run of BUSS or of its successor Buletin i universitetit Shtëteror të Tiraneës, Seria Shkencat Shoqërore, hereafter referred to as BUST. Other abbreviations which I have used are as follows:

Alt-Ithaka = Dörpfeld, W., Alt-Ithaka (Munich, 1927).Google Scholar

Epirus = Hammond, N. G. L., Epirus (Oxford, 1967).Google Scholar

Mylonas, Minoica = ‘The Grave Circles of Mycenae’ by Mylonas, G. E. in Minoica, Festschrift Sundwall (1958) 276–86.Google Scholar

Mylonas, Studies = Grave Circle B of Mycenae in Studies in Mediterranean Archaeology VII (Lund, 1964).

PAE =

PrMac = Heurtley, W. A., Prehistoric Macedonia (Cambridge, 1939).Google Scholar

Tsountas = Tsountas, C. and Manatt, J. I., The Mycenaean Age (New York, 1897).Google Scholar

Wace and Stubbings = Wace, A. J. B. and Stubbings, F. H., A Companion to Homer (London, 1962).Google Scholar

² This sword is like sword B in Epirus 319 with fig. 19 B and pl. xxi a (1) and (3).

³ The M.H. Shaft Graves which are still being excavated at Kephalóvryso near Pylos have the same design, being cut in soft rock and roofed with stones resting on wooden beams. Two snouted knives of the kind found at Dodona and in the Mati tumuli (Epirus 328 f.) were found in one of the Graves, Shaft (PAE 1964, 82 f.).Google Scholar

⁴ The existence of the M.H. walls at this point only within Grave Circle B is in itself a strong argument for supposing that the curving wall here is part of an earlier Grave Circle. Indeed, I cannot see any other hypothesis which accounts for their survival.

⁵ If the broken base was thrown in after the funeral feast as part of the fill (as in PAE 1952, 442), the depth of soil was of course greater still. I differ in my deductions from Papadimitriou. He argued from the start of the excavations that there was no tumulus and that each grave had a low mound above it (e.g. PAE 1952, 434 and 442); when he came to the base above N, he said that when the roof of grave Ν collapsed ‘the sinking was not great’ (PAE 1953, 224); I do not follow his reasoning, since the base was heavy and the aperture was very large. See his diagram in PAE 1953, fig. 12. In PAE 1952, 434 he mentioned small pockets in the rock by two graves and also channels which he connected with the funeral feasts; but we must remember that the site was inhabited earlier in the M.H. period and such pits and channels need not have been contemporary with the graves.

^5a See also Mylonas, in AE 1962, 110Google Scholar who dates the original wall to M.H.

⁶ Arch. Reports 1955. A tumulus at Samikón in the north-west Peloponnese, 5·50 m. in diameter, its soil enclosed by a circular wall of rough stones, 0·60 m. high and 0·50 m. thick, and containing fourteen graves at varying levels, several of which were cut into the soft sandstone rock, was reported first in Arch. Reports 1954, 35, and the account of the excavation reached me when this article was in proof. Here we have similarities with the mode of burial in soft rock within a grave circle as at Mycenae and here demonstrably under a low tumulus of soil, and at the same time resemblances with the tumuli of Albania. In particular there are at Samikón two two-zoned cylindrical cups with ripple pattern closely resembling the one at Pazhok (here Plate 20, 1), and several hand-made dippers, a type of pot which occurred at Vodhinë (Epirus fig. 17, 2 and 6) and in Grave Λ² of the earliest Grave Circle at Mycenae, (PAE 1954, 263, fig. 12).Google Scholar These are shown in ADelt xx (1965), pl. 14, 2 and 3, pl. 18, 9 and pl. 19. The Samikón tumulus was in use for burials from M.H. down to L.H. IIIB; and the cylindrical cups and the dippers there are dated by the excavator, N. Yialouris, to L.H. I.

^6a The L.H. tholos tombs in Messenia were similarly covered with a low tumulus.

⁷ Dörpfeld argued that these tumuli were the latest on the a priori ground that the largest were the latest (Alt-Ithaka 222 f.). This is not the case in the Albanian tumuli.

⁸ The graves with a cist-tomb as the central burial and with stone pavements, are similar to those of Kakavi and Bodrishtë (see Fig. 9).

⁹ The skeleton at Sérvia was in a contracted position on its right side, to judge from PrMac fig. 50 on p. 54. The child in Λ² was in a contracted position, and the skeleton in a simple pit-grave at Pazhok lay in a contracted position on its right side (Plate 19, 4).

¹⁰ I follow Heurtley in PrMac 126 f. and not Milojčić in this matter; see Caskey, in CAH ² 1. xxvi (a) 7.Google Scholar

¹¹ See p. 82 above for coils at Pazhok. The Saratsé coils in PrMac 88 and 203, fig. 67 qq are of the same diameter, about 2·5 cm. For Cirrha, see Kirrha 100, 104, and pl. lx, no. 14, found in a context contemporary with Shaft-Grave VI at Mycenae.

¹² He pointed out in Alt-Ithaka 221 the parallel of partial burning in the burial of Patroclus, in Iliad xxiii. 250.Google Scholar See Epirus 388 for the many similarities between tumulus-burial in Albania and this burial in the Iliad.

¹³ An early tholos tomb near Pylos in the western Pyloponnese contained much amber. Marinatos dates the tomb to c. 1500 B.C. (PAE 1964, 94).

¹⁴ See Gimbutas, Marija, Bronze Age Cultures in Central and Eastern Europe (The Hague, 1965) 259 f.CrossRefGoogle Scholar, for a summary of the ‘Kurgan’ peoples' habits of burial.

¹⁵ Caskey, in CAH ² 1. xxvi (a) 6Google Scholar has entered a warning that ‘the zoological evidence needs verification’. I am sure Heurtley did not rely on his own opinion alone.

¹⁶ Op. cit.

¹⁷ Piggott, S., Ancient Europe (Edinburgh, 1965) 85.Google Scholar

¹⁸ The sentence runs thus: ‘Caractéristiques sont les vases à fond cylindrique haut ou tronconique, vides à l'intérieur.’ Plate 22, 1–3 have the truncated-conical base or lower part. I think that vases with a high cylindrical base are not illustrated, but a ‘fruitstand’ may be meant as in PrMac, nos. 115 and 145. Vases empty on the inside may be nos. 5, 7, and 13 on Plate 22, which are open in the sense that the top does not narrow.

¹⁹ ‘Brisées’; perhaps discontinuous rather than broken.

²⁰ There are interesting comparisons to be made (but outside the immediate scope of this paper) with sites such as Lerna and Eutresis, where the irruption of invaders has been traced. Of designs on painted pottery at Libonik which do not occur in Macedonia one may compare Plate 23, 8 with Hesperia xxiv (1955), pl. 21 i (probably E.H. III, at Lerna); Plate 23, 1 with Eutresis 152, fig. 209 and 154, fig. 214; Plate 23, 3–5 with Eutresis 117, figs. 155 and 156; and Plate 23, 12 and 13 with Eutresis 131, fig. 175; the three instances from Eutresis being E.H. III and M.H.

²¹ The absence of painted pottery in the later stages at Libonik seems strange if the settlement lasted through L.H., since painted pottery of Bouboústi type has been found, for instance, in the nearby Ochrid Basin (PrMac 229).

²² This gives me particular pleasure as I owe so much in friendship and in instruction to Heurtley who took me to Sérvia and Armenochóri and let me work with him on the Sérvia pottery in Salonica. His view is supported by this new evidence against that of Milojčić and recently of Furness in PPS 1956, 204 that ‘it would be reasonable to look eastwards to West Anatolia for the origins of the Macedonian neolithic’.

^22a A fuller report in Studine Histnrike i (1966) 41–70 reached me after this article was written.

²³ See PrMac 119 f.

²⁴ On the other hand, Epirus seems to have been peopled first in the south by Early Helladic peoples and then by immigrants from Macedonia, so that the tumulus-using chieftains at Vodhinë, Kakavi, and Bodrishtë ruled peoples of a different culture from the contemporary culture in central and northern Albania.

²⁵ See Crossland, R. A. in CAH ² 1. xxvii. 49 f.Google Scholar

²⁶ This old theory seems more likely than that advocated by Chadwick, J. in CAH ² 11. xxxix. 17.Google Scholar The dialects known as West Greek are best explained as imported into Greece by peoples from the northwestern area who had for centuries been speaking a retarded dialect of Greek in Albania and north Epirus.

²⁷ The place-names of southern Albania southwards from the Devoll (Apsus) and of southern Epirus indicate a pattern of Greek settlement (Epirus 394 f.) and not Illyrian settlement in the earliest times.

²⁸ See Epirus 401 f. and CAH ² 1. xxvii. 35, dating their move into Illyria ‘as late as the first quarter of the first millennium B.C.’.

²⁹ See BUST 1957, 2. 172 f. and Epirus 424 and 428.