The authors wish to express their very deep gratitude to Dr. N. Platon, Ephor of Antiquities for Crete, for his help and ready co-operation both during the excavation of the tombs and afterwards in the work on the material in the Museum at Herakleion, where his unrivalled knowledge of Cretan antiquities was of inestimable value. His ideas and suggestions have been freely incorporated in the text beyond a possibility of particular acknowledgement in each case. We are also grateful to Mr. S. Alexiou, Epimelete of Herakleion Museum, for much kind help in our work on the material in the Museum, to Professor V. G. Childe, Professor C. F. C. Hawkes, Professor J. M. C. Toynbee, Mr. R. D. Barnett, Miss S. Benton, Mr. P. E. Corbett, Miss D. H. F. Gray, Dr. H. Hencken, Mr. R. W. Hutchinson, Miss R. Levy, Miss H. L. Lorimer, Miss N. K. Sandars, Mr. L. C. D. Tait, and many others who have kindly discussed the material and offered valuable suggestions and references that have helped to elucidate it.

We are very much obliged to Mr. I. W. Cornwall, of the London University Institute of Archaeology, for examining the animal teeth and minerals, and to Mrs. F. L. Balfour-Browne and Dr. E. Trewawas of the Natural History Museum, South Kensington, for identifying the wood remains and a shark's tooth from Tomb I; to Dr. D. B. Harden, of the Ashmolean Museum, and Professor W. E. S. Turner, of the Society of Glass Technology, for reporting on the lump of glass from Tomb III; and to the Reverend V. E. G. Kenna, R.N., for a note on the seal from Tomb I, and advice and information with regard to the seals from Tomb III.

Other acknowledgements are reported in the appropriate place. The drawings, except for the plan, Fig. 1, are entirely the work of P. de Jong. The report on the tombs is by P. de Jong and M. S. F. Hood in collaboration. The account of the helmet and other objects from the tombs is by M. S. F. Hood. The photographs of the objects were taken by Mr. E. M. Androulakis of Herakleion using the School's plate camera.

The following abbreviations are used in this article, in addition to those ordinarily in use in the Annual.

AC Pendlebury, , The Archaeology of Crete, 1939.Google Scholar

Chamber Tombs Wace, , Chamber Tombs at Mycenae (Archaeologia 82), 1932.Google Scholar

Homer and the Monuments Lorimer, , Homer and the Monuments, 1950.Google Scholar

MP Furumark, , The Mycenaean Pottery, 1941.Google Scholar

New Tombs at Dendra Persson, , New Tombs at Dendra, 1942.Google Scholar

Prehistoric Tombs Evans, , The Prehistoric Tombs of Knossos (Archaeologia 59), 1906.Google Scholar

Prosymna Blegen, , Prosymna, 1937.Google Scholar

Royal Tombs at Dendra Persson, , The Royal Tombs at Dendra, 1931.Google Scholar

Schachtgräber Karo, , Die Schachtgräber von Mykenai, 1930–1933.Google Scholar

TDoAx Evans, , The Tomb of the Double Axes (Archaeologia 65), 1914.Google Scholar

¹ For ‘shaft-graves’ at Knossos, see Prehistoric Tombs 11.

² There were indications of a possible fifth tomb (see p. 251).

³ Not, however, the first Bronze Age metal armour recorded from Crete. See note, p. 260.

⁴ As Evans says (PM IV 785 ff.), ‘The last Palatial phase at Knossos presents a military and indeed militaristic aspect’, reflected for instance in the ‘Armament Tablets’ and the ‘Shield Fresco’ (here dated to the beginning of L.M. II: but see PM III 308, followed by AC 197–8, where it is called L.M. IA); and illustrated by the break in continuity visible in the remains of the smaller sites along the northern coast, and by the cessation of Phaistos and Ayia Triadha. For an opposite view of the Minoan Cretans as essentially pacific by contrast with the war-like Mainlanders, see AC 271, 286.

⁵ PM IV 849, fig. 832. This is clearly the tomb described in PM II 547, as containing L.M. IA relics, and marked on the plan (opposite p. 547) as ‘Rock Tomb L.M. I’ (at the bottom edge of the plan, just north-west of the ruined Chapel of ‘Hagios Kyrillos’). The objects from the tomb, which are now in the Ashmolean Museum at Oxford, consist of two short spear-heads or javelins like AJ (4) and I (11), together with one long ‘horned’ sword and fragments of another: the only vase recovered from the tomb was a squat alabastron, which emphasises the similarity with the tombs of the Hospital site (see p. 254, and n. 43). For other ‘warrior-graves’ at Knossos, cf. Tomb 36, the ‘Chieftains's Grave’, in the Zafer Papoura cemetery (Prehistoric Tombs 51 ff.), with long ‘horned’ sword, short ‘cruciform’ sword, and two spear-heads like II (4); and Tomb 44 in the same cemetery (ibid. 62), with long ‘horned’ sword and short sword.

⁶ For the ‘Chariot Tablets’, etc., see PM IV 786 ff. Compare the similar inventories of war-like stores kept in the Palace at Nuzi, together with records of the equipment bestowed upon individual warriors (Starr, Nuzi I 541).

⁷ As in the famous fresco of ‘The Captain of the Blacks’ (PM II 755, pl. xiii; IV 886).

⁸ The discovery was reported to Dr. Platon, who invited P. de Jong to complete the examination of the tomb.

⁹ These small spear-heads probably had quite short hafts, and were hunting spears or javelins (see p. 256, and the evidence from Tomb I, p. 265, under I (11)).

¹⁰ Compare Tomb V and the body on the ledge in Tomb I of the Hospital site.

¹¹ See the map, PMII, opposite 547. The stream is here shown as the ‘Gully’ which runs east from ‘Loukas Taverna’. The stream has now been filled in and a new bed dug for it about 100 metres further south to make a boundary for the Hospital grounds on that side.

¹² The floors of the tombs were very close to the surface of the levelled area. The actual removal of the soil was therefore a simple matter, and only two men were employed, Spiro Vasilakis and Grigori Kritsalakis, under the supervision of Manoli Markoyannakis, the School's foreman at Knossos.

¹³ The chambers, dug in the soft white ‘kouskouras’ rock of the region, had long ago collapsed.

¹⁴ All three types of tomb were found side by side in the Zafer Papoura cemetery (Prehistoric Tombs 1 ff.).

¹⁵ Tombs II–V were cleared by P. de Jong, Tomb I (which was the last to be discovered) by both of us in collaboration.

¹⁶ This may be copied from the door of a house with high stone socles supporting a wooden frame. Tomb 9, the most elaborate of the important group of fourteen ‘Tombe dei Nobili’ at Phaistos, had a similar façade, although the ‘pilasters’ were here rather lower (MA XIV 510, figs. 3 and 4). For other carved doorways, cf. the ‘Tomb of the Double Axes’ and Tomb 5 of the same group at Isopata (TDoAx, figs. 33a and 47), and Tomb 14 (the ‘Tomb of the Tripod Hearth’) at Zafer Papoura (Prehistoric Tombs, fig. 32). For rock-cut benches, cf. Tomb 9 at Phaistos (above) and Tomb 3 of the Isopata group (TDoAx, fig. 19).

¹⁷ There was no evidence that the bodies had been placed in the tomb at different times, and if anything the arrangement of the vases suggested a single occasion of burial. It is therefore worth bearing in mind the possibility of a wife or concubine killing herself or being killed on the death of her husband or master. Something of this kind was suspected by Persson in the tholos tomb at Dendra (Royal Tombs at Dendra 68, with a full discussion). For such practices among the Celts of Gaul, illustrated both by the literary and by the archaeological evidence, see Déchelette, , Manuel d'Archéologie IV (1927), 541 ff.Google Scholar

¹⁸ If coffins were employed they would probably resemble the clay larnakes with legs and gabled lids common in Late Minoan tombs. Such larnakes were clearly imitated from wooden models (Prehistoric Tombs 8 ff.). They are quite short, and the bodies were placed in them with the knees drawn up; so that a body in a wooden coffin of this type might well reach the flexed position seen in the case of the skeleton on the floor of Tomb I, cf. Tomb II (Fig. 5). For wooden coffins, see New Tombs at Dendra III ff., the only example established with certainty either in Crete or on the Mainland. This was about 1·90 long and 0·55–0·60 wide, and apparently had no legs. Clay larnakes are hardly found in tombs on the Mainland (Chamber Tombs 9), and wooden coffins may also therefore have been exceptional there: the occupant of this coffin at Dendra was accompanied by the unique bronze helmet! For (?) biers, see Tomb 9 at Phaistos, in the pit to the right of the entrance (MA XIV 522 and n. 1), and Prosymna, where traces were noted in three or four tombs (Prosymna 249).

¹⁹ Most of the vases in the south-east part of the tomb had been badly smashed and scattered. The amphora (7) was originally full of fine grey ash, which Mr. I. W. Cornwall, of the London University Institute of Archaeology, kindly examined and found to be highly calcareous, containing a considerable quantity of phosphate and scraps of charred wood; it could therefore be fire-ash: the wood scraps were identified by Mrs. F. L. Balfour-Browne, of the Natural History Museum, South Kensington, as olive and oak. A few tiny fragments of carbonised wood, identified by Mrs. Balfour-Browne as olive, were recovered from the alabastron (9); but these may have belonged to the contents of the amphora (7) which were much scattered. For the probable uses of alabastra, see p. 254.

²⁰ It is difficult to believe that this represents the position in which the vase had originally been placed in the tomb; and it had therefore presumably been thrown up in some way when the roof of the chamber collapsed. If the vase had been resting on top of the bier or coffin, this might be more easily understood.

²¹ Most of the fourteen ‘Tombe dei Nobili’ at Phaistos were of this shape (MA XIV 507, fig. 2): cf. Prehistoric Tombs 3 n. b, for other examples from Crete; but the shape is curiously enough not represented among the forty-nine chambertombs of the Zafer Papoura cemetery, although it is found in the Mavrospelio cemetery at Knossos, (BSA XXVIII 243 ff., figs. 1 and 27, Tombs iii and xiii).Google Scholar It occurs on the Mainland, e.g. at Mycenae (Chamber Tombs 135) and Prosymna (Prosymna 243).

²² There is no evidence from this tomb to indicate that the bodies may have been a man and a woman as in Tomb I. Beads and seal-stones were, of course, worn by men as well as by women.

²³ If we assume that the ivory adorned a sheath for the dagger, it follows that since the dagger was found lying on top of the decorated face of the ivory, it cannot have been inside the sheath, which is curious. It is possible therefore that the ivory belonged to a quiver, which was hanging, together with the dagger and a baldric fastened by the gold toggle, from a wooden peg in the wall of the chamber. There was some evidence in a Mycenaean tomb in the Agora at Athens that a garment had been hung on a peg set in the wall (Hesperia XVII (1945), 157). Note that in Tomb 10 (‘The Hunter's Grave’) in the Zafer Papoura cemetery the fifteen arrow-heads were found all together at a height of 0·40–0·60 above the floor, which suggests a quiver full of arrows standing upright in a corner of the tomb (Prehistoric Tombs 31, fig. 27).

²⁴ Cf. the shark's tooth from Tomb I (15).

²⁵ A number of sherds were recovered from the fillings of the dromoi of the chamber-tombs, and from the shaft-grave (Tomb II). These were mostly very fragmentary, and therefore difficult to attribute; but they appeared to range from M.M. I to L.M. I.

²⁶ Prehistoric Tombs 15 ff. The ‘cave’ was blocked off from the ‘pit’ by a stone wall. In many of the tombs of the Zafer Papoura cemetery nothing was found, and no trace of a skeleton was detected.

²⁷ Compare the turbans placed on the coffins of Turkish warriors. If in this tomb there was a wooden coffin with a gabled lid like the clay larnakes (see p. 248, n. 18), the comparison would be strikingly apt.

²⁸ The vases from the Palace itself which display the classic ‘Palace style’ of Late Minoan II are mostly large amphorae. Small clay vases, apart from stemmed goblets, were excessively rare; doubtless because, as Evans suggests (PM IV 353), the smaller vessels of the Palace during this period were made of metal—bronze, gold, and silver—and were therefore removed or plundered when the Palace was destroyed.

²⁹ Even in the tombs with two bodies the interval between the first burial and the second would not be long, since a close connection of some kind, such as that of husband and wife, may be supposed to have existed between the occupants of the tomb. For the possibility that there was only one occasion of burial even when there were two bodies, see p. 248, n. 17.

³⁰ Cf. PM IV 300 ff.

³¹ The combination of a greenish tinge in the clay with a uniform deep black colour in the decoration is probably to be explained by a particular circumstance of the firing rather than by a different centre of manufacture for these vases.

³² The vases had all been broken, and a few (I (4), III (6), and the lamps I (10) and III (11)) had been smashed beyond any possibility of restoration. The work of restoration was in any case difficult, since the fabric had become very soft and friable, and many fragments had actually warped out of shape. The School's technician, Stelio Katsarakis, showed great skill and ingenuity in the successful accomplishment of this task.

³³ Furumark (MP 167) states that ‘there seems to be no example of the argonaut in the known material’. Cf. ibid. 306, where he cautiously adds: ‘but it may have belonged to the Palace style repertory’.

³⁴ MP, fig. 16.

³⁵ See Pendlebury (AC 227) on the difficulty of distinguishing Mainland from Cretan products of this period by the fabric: ‘The old distinction between the “feel” of the sherds … can, as I can testify, no longer hold good’. It would be particularly interesting, as Pendlebury, loc. cit., points out, to settle the question of origins in the case of the Late Bronze Age pottery exported from the Aegean and found in Egypt and elsewhere abroad. Chemical analysis of the clay might help in this matter.

³⁶ Wace (Chamber Tombs 182 n. 2) reserves the term ‘kylix’ for the tall L.M./L.H. III goblets. The Knossian goblets of L.M. I–II, and the Mainland goblets of L.H. I–II, are then simply ‘stemmed goblets’.

³⁷ Some of the published examples are classified by Furumark as L.M. IIIA 1 (MP 171 n. 2, referring to PM IV fig. 302 c). This would still, of course, precede the destruction of the Palace (MP 169).

³⁸ PM IV 381 ff., fig. 323–4, cf. AC 200. The blue colour of the vase is presumably intended to represent silver. Furumark rightly points out that the lower part of the vessel is not preserved on the fresco, and argues that it is not a true goblet, but a deep cup of L.M. I type (MP 56). But these L.M. I deep cups seem normally at any rate to have only one handle.

³⁹ Pendlebury (AC 223, published in 1939) could cite only five examples. Furumark (MP 40 n. 4) adds two fragments from Zakro, which may belong to an alabastron (BMC Vases I, A 707: there classified as L.M. II, but Furumark would make them L.M. IB).

⁴⁰ These alabastra have mostly been found in tombs; but they occurred in the settlement at Korakou (Blegen, Korakou 50 ff.).

⁴¹ Furumark, however, expresses a strong faith that a number of alabastra found abroad (in Melos, Palestine, and Egypt), and decorated in a L.M. I style, are of Cretan manufacture (MP 40); while admitting that it is a ‘very remarkable fact that seven out of the fifteen or sixteen L.M. I vases found in East Mediterranean lands are of a type that is extremely rare in Crete’. See also MP 663, where he is clearly anxious to maintain his faith against Wace, and Blegen, (Klio XXXII 137 ff.)Google Scholar and Pendlebury (AC 223), who argue a Mainland origin for these early alabastra found abroad.

⁴² PM IV 938.

⁴³ E.g. the tomb on the ‘Acropolis’ hill (see p. 245, n. 5), and Tomb 3 (‘The Mace-bearer's Tomb’) at Isopata.

⁴⁴ See p. 254, n. 35.

⁴⁵ MP, Motive no. 32: ‘Rock Pattern I’.

⁴⁶ MP 322.

⁴⁷ TDoAx 24, fig. 35.

⁴⁸ Concentric circles on the bases of II (2), III (7) and (8), V (2), and probably V (1).

⁴⁹ Blegen (Prosymna 419–20) goes so far as to say that ‘In all examples (i.e. from Prosymna) dating from L.H. II or earlier the bottom bears as its characteristic decoration … a “wheel pattern” … In fact, I do not know of a single alabastron of contemporary date from any Mainland site that does not bear this distinctive mark in one form or another’. See, however, Furumark (MP, under Motive no. 68: ‘Wheel’, 404), who claims that four out of the nine L.H. I alabastra bases that can be distinguished have ‘simple circles, mostly concentric, once with a central disc’; but he admits that the wheel pattern is characteristic of L.H. II, though concentric circles appear again in IIB and in III supersede the wheel.

⁵⁰ PM IV 851. For a description of these swords see under A J (1), p. 261.

⁵¹ These swords are really short stabbing-swords rather than ‘rapiers’ in the strict sense.

⁵² The largest (V (6)), with a length of 0·49, was longer than the sword found in the same tomb. But even larger spearheads have turned up on the Mainland. One from the tholos tomb at Dendra, for example, attained the extraordinary length of 0·57 (Royal Tombs at Dendra 37, no. 18).

⁵³ PM IV 842. Cast spear-heads with a ring round the base of the socket appear at Ras Shamra in ‘Ugarit Moyen’ I (c. 2000 B.C.). But the ring is rare on spear-heads of the Late Bronze Age there according to Schaeffer, (Ugaritica II 55 ff.).Google Scholar

⁵⁴ For further examples see p. 267, n. 119.

⁵⁵ For the possibility that these early socketless blades were used as spear-heads as well as daggers, e.g. Childe, , Dawn of European Civilisation (4th ed. 1947), 32.Google Scholar

⁵⁶ But there was clearly no very rigid differentiation between weapons of war and weapons of the chase; and hunting might call for the organisation and armament of war. This is well illustrated by the famous ‘lion-hunt’ dagger from Mycenae, where men carry the figure-of-eight shield also used in warfare.

⁵⁷ Marinatos, , ‘Σιγύνη’ (BSA XXXVII (1936–1937), 187 ff.).Google Scholar From the Vaphio tholos came one other spear-head, which was larger and had a ring round the base.

⁵⁸ For Minoan helmets see PM IV 867 ff. For Aegean helmets in general, Kukahn, , Der griechische Helm (1936)Google Scholar; and more recently, Lorimer, , Homer and the Monuments (1950), 211 ff.Google Scholar

⁵⁹ The School's technician, Stelio Katsarakis, was employed for nearly three weeks on this exacting task.

⁶⁰ The metal is very thin at this end, which might indicate that it was the back part of the helmet.

⁶¹ Sprockhoff, , Zur Handelsgesch. der germ. Bronzezeit (Berlin, 1930) 44Google Scholar, apropos of the Oranienburg helmet, one of the group of three which most resemble our helmet (see p. 259).

⁶² These had escaped our notice at first; but they were observed by Mr. de Jong on making a careful examination of the knob with a view to drawing it.

⁶³ Possibly a long bunch of horse-hair. For representations of plumes flowing from the tops of helmets of similar shape, though not necessarily made of metal, cf. the ‘Battle of the Glen’ gold signet-ring from Shaft-grave IV (Schachtgräber, pl. xxiv, no. 241). What seems to be an actual plume of this type 0·22 long, composed of gold strips, was found in Shaft-grave V (ibid., pl. lvi, no. 639).

⁶⁴ PM IV 867. Homer and the Monuments 222–3. Cf. ibid. 210–11, for linen corslets in later times. The defensive qualities of such padded armour ought not to be underrated. The allied troops at the Battle of Lepanto found quilted corslets an effective protection against the formidable Turkish arrows (Wiel, A., The Navy of Venice (1910), 256 n. 1Google Scholar).

⁶⁵ TDoAx 27, fig. 37 b, reproduced in Homer and the Monuments 221, fig. 22.

⁶⁶ See Homer and the Monuments 212 ff.

⁶⁷ New Tombs at Dendra 43, 119 ff., pl. I.

⁶⁸ Cf. the views of Philopoemen on the importance of keeping metal armour bright: Μεγάλα γὰρ μὲν ἔφη τὴν λαμπρότητα συμβάλλεσθαι πρὸς ἔκπληξιν τῶν ὑπεναντίων. (Polybius XI 9. Quoted by Miss Lorimer, Homer and the Monuments 252 n. 4.)

⁶⁹ Cf. the helmet on the vase from Isopata T. 5 (n.65 above). It is just conceivable that the curious lead-filled disc V (8a), that was found with the remains of the helmet, might have acted as a weight in some way to such a neck-piece. The thin bronze coating on one side of the lead seems to make sense only if it was intended to be seen; and the disc might therefore have been sewn into the neck-piece with the bronze showing on the outside. This suggestion was put forward, although with every reserve, by Miss D. H. F. Gray of St. Hugh's College, Oxford, who has kindly permitted us to quote it. It is admittedly not very satisfactory; but no better explanation for the use of the lead disc has offered itself up till now.

⁷⁰ New Tombs at Dendra 43, 119 ff., pl. I. There is also a solitary bronze cheek-piece of a helmet from a Mycenaean tomb at Ialysos (Rhodes) in the British Museum (Walters, , BMC Bronzes (1899), no. 36: Homer and the Monuments 211, pl. xiii, IGoogle Scholar, where it is illustrated for the first time): this closely resembles our cheek-pieces in the thinness of the metal (about one millimetre), the slightly rounded section to agree with the curve of the helmet, and the little holes round the edges for sewing on to a padded lining; but it is rather larger (0·175 long by 0·125 wide at the top), and the profile of the front edge is concave instead of scalloped as on our cheek-pieces.

⁷¹ E.g. the helmets on fragments of a silver vase from Shaft-grave IV, and that on a gem from the Vaphio tholos tomb (reproduced in Homer and the Monuments, pl. xv, 1, and 217, fig. 20).

⁷² Cf. Homer and the Monuments 225: ‘The commonest form of Minoan-Mycenaean helmet is conical’.

⁷³ Woolley, , Ur Excavations II. The Royal Cemetery (1934), pl. 218.Google Scholar

⁷⁴ Davies, Ancient Egyptian Painting, pl. xxiv, from the Tomb of Menkheperra-senb, dating from the later years of Thutmose III (1475–48). The plumes are painted blue and red, the helmet yellow, which may be intended for bronze. This reference was kindly supplied by Mr. L. C. D. Tait.

⁷⁵ This unique monument dates from the period of the Hittite Empire. On some later ‘Hittite’ monuments from Carchemish and elsewhere, which all seem to belong to the Iron Age, helmets or caps surmounted by a large globe are worn by gods (Weber, , Die Kunst der Hethiter (Orbis Pictus, Vol. 9), pls. 2, 3, 14, 21 and 29).Google Scholar Humans on some Iron Age reliefs from Zencirli (ibid., pls. 24–6 and 33) are wearing what are apparently (Mr. R. D. Barnett informs me) leather caps and not helmets in any true sense of the word, but which carry a low knob or projection on top giving them a profile remarkably like that of the helmet on the Isopata vase. The standard Assyrian metal helmet of the Iron Age is high and pointed, but normally at any rate without cheek-pieces or plume or knob on top.

⁷⁶ But note the very high pointed cap, that might conceivably be regarded as a helmet, worn by the figure carrying shield and spear who appears to be a warrior-god, on a sealing from the Temple Repositories at Knossos of M.M. III date (PM I 505, fig. 363 b; III 465, fig. 324 A).

⁷⁷ PM I 308, figs. 228, 1 and 229 b. This might even be intended to represent a metal helmet.

⁷⁸ For the survival of a conical type of helmet in the Aegean and Cyprus into the Iron Age, see Homer and the Monuments 225 and n. 2. From Greece itself only two examples appear to be known, both from Olympia, (Ol. IV 172Google Scholar, pl. lxii, no. 1031. JdI LII (1937), Ol. Bericht. 52, pl. 6 right): neither is exactly comparable to our helmet, and neither has preserved any trace of a knob, if there ever was one. Furtwängler, in discussing the first of these, could only cite two parallels, both from Cyprus and both unpublished. The place of manufacture of these helmets would appear to be quite obscure. Models of detached cheek-pieces were found at Bassae, (AE 1910, 315, fig. 34)Google Scholar; Kukahn (Der griechische Helm, n. 45) thinks that these are comparatively early in date; but they are not like the cheek-pieces of our helmet.

⁷⁹ Merhart, Von, Zu den ersten Metallhelmen Europas (30 Bericht der Rom. Germ. Komm.) (1940), 4 ff.Google Scholar See also Sprockhoff, , Zur Handelsgesch. der germ. Bronzezeit (1930), 44 ff.Google Scholar, for the German examples. Von Merhart argues that the early metal helmets of Europe are mid-European creations, and certainly free from Aegean influence. He had, however, prophetically complained apropos of the Dendra helmet that in the light of the representations a metal helmet of the Aegean Bronze Age might have been expected to resemble a European ‘Bell-helmet’.

⁸⁰ But a large round knob appears in some Aegean representations of helmets, e.g. a fragment of fresco from Mycenae, (AM XXXVI (1911), 239, pl. xi, 2.Google Scholar Cf. Homer and the Monuments 217): this might be intended for a boar's tusk helmet; but admittedly no trace of divisions could be recognised. The colours in which the helmet was painted are not given.

⁸¹ Von Merhart (op. cit., 11) notes that these three helmets fall outside the main group of ‘Bell-helmets’ by reason of their smaller knobs. With them he places the two helmets from Corneto in Italy (see p. 259, n. 88).

⁸² B.M. no. 68.12–28. 248. Described by Sprockhoff (see p. 259, n. 79 above). In the files of the British and Mediaeval Department of the Museum is a very clear and detailed MS. summary of the literature on the Beitzsch helmet by Professor C. F. C. Hawkes. With his consent, and by courtesy of the Department and the Trustees of the Museum, the fullest use has been made of this account.

⁸³ These are also in the British Museum. The blade has elaborate incised decoration, and closely resembles the most ornate of the blades from the Neuenheiligen hoard (Childe, The Danube in Prehistory, fig. 143, top r. centre).

⁸⁴ For recent views on the dating of the European Bronze Age, see Childe, and Hawkes, in Proc. Prehistoric Society XIV (1948), 177 ff.CrossRefGoogle Scholar

⁸⁵ The Beitzsch helmet weighs only 353 grammes compared with the 695 grammes of our helmet without the cheek-pieces. The weight of the Beitzsch helmet was kindly supplied by Mr. J. W. Brailsford of the British and Mediaeval Department of the British Museum.

⁸⁶ It must, however, be noted that the standard of workmanship would tend to reach a very much higher level at a great capital city like Knossos, than at some provincial centre. This can easily be seen by comparing the spears and swords from our tombs at Knossos with contemporary weapons of the same types from provincial sites in Crete.

⁸⁷ Illustrated by Hampel, , Ältherthümer d. Bronzezeit in Ungarn (1890), pl. xxxiii, 2 c.Google Scholar In the case of the Beitzsch helmet at any rate the metal appears to have been pressed down while in a soft or molten state with some implement whose marks are clearly visible, cf. Sprockhoff, Zur Handelsgesch. der germ. Bronzezeit 44, apropos of the ‘Bell-helmet’ from Sehlsdorf.

⁸⁸ Two from Corneto (Etruria), two from San Canziano near Trieste, and one from somewhere in Italy, according to von Merhart's list. But there are also clay imitations of ‘Bell-helmets’ which were used, like the helmets themselves, to cover cremation-burial urns, e.g. Randall-MacIver, , Villanovans and Early Etruscans, pl. 11, 15, from Corneto.Google Scholar

⁸⁹ The two from Corneto. These are tentatively grouped by von Merhart with the three ‘Bell-helmets’ of p. 259 and n. 81, which most closely resemble our helmet.

⁹⁰ But with only four rivets in place of the seven on our helmet.

⁹¹ The ‘Bell-helmet’ is not the only early type of helmet in Italy for which Aegean Bronze Age parallels can be found. The distinctive ‘Villanovan’ helmet with its fore-and-aft ridge-crest strongly resembles a helmet depicted on a silver vase from Shaft-grave IV, as Kukahn and Miss Lorimer note (Der griechische Helm 7. Homer and the Monuments 216, pl. xv, 1. Cf. Schachtgräber, pl. cxxxi, g). The third type of early Italian helmet, which also occurs in Central Europe, is not very unlike some of the squatter forms of helmet seen on Aegean monuments (compare e.g. Randall-MacIver, , Villanovans and Early Etruscans, pl. 13, 12, 15 and 16Google Scholar, with helmets depicted on the Zakro, sealings, JHS XXII 79, nos. 24 and 25Google Scholar). In this context the similarity long ago noted between the ancilia of ancient Rome and the Minoan figure-of-eight shield may be kept in mind (cf. Evans, , Tree and Pillar Cult (1901), 31Google Scholar).

⁹² Von Merhart, , Zu den ersten Metallhelmen Europas 23, fig. 8, 3Google Scholar, from Pass Lueg, Reichsgau (Salzburg).

⁹³ Prähist. Zeitschrift XXXII–XXXIII (1941–42), 74, fig. 13. L. 0·155, W. 0·118. In shape not unlike the cheekpiece from Ialysos (Homer and the Monuments 211, pl. xiii, 1), although the metal is much thicker. This reference was kindly supplied by Dr. H. Hencken.

⁹⁴ E.g. BM Guide to Greek and Roman Life (3rd ed. 1929), 79, fig. 72, no. 230.

⁹⁵ For Roman helmets which continue the tradition of the Etruscan ‘Jockey-cap’ helmets, ibid., 81, and fig. 74; but the cheek-pieces are missing here. For the cheek-pieces, see BM Guide, Roman Britain (1951), pl. xxv, 6. Both these helmets were found in Britain!

⁹⁶ Compare the holes along the edges of the helmet and cheek-pieces from Tomb V (p. 257 and Plates 50 and 51).

⁹⁷ Homer and the Monuments 200, 246.

⁹⁸ PM IV 804. Homer and the Monuments 196 ff. Actual bronze scales were found in quantities at Nuzi (Starr, , Nuzi I 475, pl. 126Google Scholar). Part of a bronze ‘belt’ came from one of the tombs at Shamra, Ras (Syria XIX 240, fig. 32 w, reproduced in Ugaritica I, fig. 63 w).Google Scholar

⁹⁹ PM IV 803. For the possibility that the waist-belt of Minoan civilian dress may have been plated with metal on occasion, cf. the ‘Cup-bearer’ fresco, where the blue and orange of the belt might be taken to represent silver and copper as seems probable for the rhyton (PM II 705, col. pl. xii).

¹⁰⁰ It is not made clear from which of the fourteen tombs the various bronze plates came, or whether they were all from the same tomb.

¹⁰¹ PM III 184, fig. 128. Evans thought the rectangles were intended for boar's tusks, cf. Homer and the Monuments 220. But Persson (New Tombs at Dendra 129) believed that in Crete at any rate such representations implied metal plates, e.g. the Zakro sealing reproduced in PM IV, fig. 856. Metal discs from Shaft-grave IV may, as Karo suggests, come from a helmet (Schachtgräber, nos. 541–49, pl. lxx); Evans, (PM IV 869)Google Scholar cites, but does not illustrate, an ivory relief from Knossos showing a helmet which seems to be armed with similar metal discs.

¹⁰² For the slots, see above under A J (1).

¹⁰³ Crete: Khamaizi, (AE 1906, pl. 7, 5).Google Scholar Tip broken. L. as preserved 0·19. Presumably Mallia, M.M.I. (Études Crétoises VII 52, pl. lxi, 2–3, no. 2255)Google Scholar: from the interior of the necropolis building at Khrysolakkos. L. 0·28. Attributed by Demargne, but only on the grounds of its appearance, to the second period of the Palace. Cyclades: Amorgos, (AM XI (1886), 24, Beil. 1,8)Google Scholar: L. 0·23. Troy (Dörpfeld, , Troja und Ilion I 344Google Scholar, fig. 262 d): Troy II–V. N.B. On p. 345 Dörpfeld speaks of a mould found at Troy for a blade of this type. These blades were probably spear-heads (Childe, , Dawn of European Civilisation (4th ed. 1947), 54Google Scholar; for the system of mounting on a haft, ibid. fig. 26, reproduced from BM Bronze Age Guide (1920), fig. 173).

¹⁰⁴ There is one in the collection of Dr. Giamalakis at Herakleion (Κρητικὰ Χρονικά Δ´ (1950), 110 and 122, no. 358, pl. Γ′. Published by Miss Xenakis. Dr. Platon kindly brought this reference to our notice). L. 0·216. Miss Xenakis notes the resemblance of this blade to our blade A J (2). A second blade of this type was found in a hoard of L.M. bronzes at Tourloti, near Mouliana (Sitia), and is now in Herakleion Museum (No. 542): it is mentioned by Xanthoudides, (AE 1906, 135)Google Scholar apropos of the Khamaizi M.M. blade (see n. 103 above). Some daggers of the Late Bronze Age in the Aegean also have a leaf-shaped blade, e.g. Prehistoric Tombs 82, fig. 90 (from Knossos), and BM Bronze Age Guide (1920), fig. 173 (from Naxos).

¹⁰⁵ For differentiated boar-spears, see Marinatos, in BSA XXXVII (1936–1937), 191, esp. n. 2.Google ScholarCf. the fine spear-head from Ras Shamra with two boars springing from the ring at the base of the socket, and Schaeffer's, remarks (Syria XIV (1933), 118–19).Google Scholar

¹⁰⁶ There is a very similar spear-head in the collection of Dr. Giamalakis (Κρητικὰ Χρονικά Δ´ (1950), 112 and 124, no. 508, pl. Δ′); and another from Gournia (Boyd-Hawes, , Gournia, pl. iv, 48Google Scholar).

¹⁰⁷ For running spirals on jugs of this type, but of an earlier date, e.g. PM IV, fig. 195 (L.M. IA).

¹⁰⁸ Three handles are unusual. Cf. PM IV, fig. 298, from the S.W. part of the Palace at Knossos, dated by Evans to the beginning of L.M. II.

¹⁰⁹ Furumark regards these as iris (MP 190 n. 1. N.B. i should be j); but Evans seems to think that all the flowers on this sherd may be a version of ‘honey-suckle’ as seen on a fresco fragment from the House of the Frescoes, (PM II 469).Google Scholar

¹¹⁰ For the influence of the naturalistic designs of M.M. III wall-paintings on the decoration of L.M. pottery, see PM II 468 ff., cf. JHS XXIII 194. For a fresco with partridges, cf. PM II, fig. 51–4, from the Caravanserai, which according to Evans is L.M. IA (ibid., 116); one of these partridges has a raised wing (fig. 51), another a slightly parted beak (fig. 52), like the birds on our vase. Furumark (MP 195) suggests that representations like that of this partridge fresco may have served as models for some of the birds on L.M. III vases.

¹¹¹ See AC 243.

¹¹² See PM IV 314, fig. 250, for the evolution of trefoil rock-work from the triple group of rock and seaweed of the L.M. IB ‘Marine’ style.

¹¹³ See p. 249, n. 19.

¹¹⁴ For Minoan clay lamps in general, see BSA XXVIII 292, pl. xxii, where the lamp V 23 (at bottom r.) most resembles our small type I (10) and III (12). Cf. MP 77. N.B. The raised lip by the handle is to protect the hand from heat.

¹¹⁵ Dr. E. Trewawas of the Natural History Museum, South Kensington, who kindly examined the tooth, informs us that it matches well with the teeth of some of the species Carcharhinus, notably C. longimanus (Poey) and C. obscurus (Lesueur), both of which are known from the Mediterranean.

¹¹⁶ This is doubtless because, as Mr. Cornwall suggests, the bronze was in contact with the decaying flesh of the body, which, containing sodium chloride, would favour metallic corrosion. The contrast between the poor preservation of the sword and the unusually fine condition of the spear-head II (4) is remarkable.

¹¹⁷ Cf. a ‘horned’ sword from Tomb 9 of the ‘Tombe dei Nobili’ at Phaistos, (MA XIV 535–6Google Scholar, fig. 20 a).

¹¹⁸ The design of interlocking spirals shows signs of wear in places owing to frequent handling of the sword. This is some indication, if such were required, that richly decorated weapons of this type were not kept merely for occasional parade or ceremonial use.

¹¹⁹ For such loops on the gold hilts of ‘cruciform’ swords, see the sword from the tholos tomb at Dendra and that from the Zafer Papoura cemetery at Knossos described below. Cf. the ‘cruciform’ hilt-mountings of agate and faience from Mycenae, and one of crystal from the Palace at Knossos (Prehistoric Tombs 110; illustrated in PM IV 852–4, figs. 836–7).

¹²⁰ The peg may have been of wood or ivory and not bronze, as no trace of it was found.

¹²¹ Persson believed that these two swords actually came from the same workshop, which was situated on the Mainland (New Tombs at Dendra 61). The discovery of our sword suggests a Cretan origin at any rate for the two swords from Knossos.

¹²² Prehistoric Tombs 55, figs. 55–7. Cf. BM Bronze Agé Guide (1920), fig. 171, from Ialysos.

¹²³ Cf. a bronze double axe from Phaistos (Mosso, , Dawn of Mediterranean Civilisation (1910), 318Google Scholar, fig. 180). For butterflies in general as representing the ‘life’ of man, see PM III 148 ff., cf. II 787.

¹²⁴ Cf. III (14) (p. 271), where the wood was identified as olive.

¹²⁵ Cf. the rather similar motif on a jug from Korakou (Blegen, Korakou, fig. 69 and pl. V), classified by Furumark as Myc. IIB (MP no. 47, 1).

¹²⁶ For the rosettes, cf. a fragment of a L.M. II goblet from Knossos, (PM IV 360, fig. 302 a).Google Scholar

¹²⁷ BSA Suppl. I (1923), 74. Cf. MP 205, where Furumark notes the occurrence of stippling on Palace Style vases with designs both of the marine and of the floral class, and its survival into L.M. IIIA1.

¹²⁸ PM IV 306. Furumark agrees with Evans that this type of stippling was imitated from a fresco pattern, which probably represented sands (MP 423): but he doubts if such stippling was really meant to depict sea sands in marine scenes on Palace Style vases, and suggests that it may have served as a substitute for other patterns which were used to represent water in marine compositions of L.M. IB–II (ibid. 205–6).

¹²⁹ For the tentacles, cf. in a general way the types classified by Furumark as IIB–IIIA1 (MP, no. 22).

¹³⁰ See under III (4) above and nn. 127 and 128.

¹³¹ The figures on this vase are called by Wace ‘argonauts’; but Furumark classifies them as ‘sacral ivy’, and regards them as ‘a kind of pictorialisation under the influence of the argonaut’ (MP 269, no. 12, 11 = Myc. IIA).

¹³² Contrast the thick handles with their elaborate ‘metallic’ sections on the other amphorae from these tombs. The thin strap handles of this amphora may, of course, also be imitated from metal.

¹³³ Cf. BSA XXVIII 294, fig. 46 (ix. B. 12), from the Mavrospelio cemetery: also PM I 579, where the type is described and compared with Egyptian ‘candlesticks’ of Dyn. IV.

¹³⁴ Several tombs in the Mavrospelio cemetery produced fragments of lamps painted in unfixed colours, red, white, and blue (BSA XXVIII 294). Cf. TDoAx 26 ff., and Prehistoric Tombs 72.

¹³⁵ E.g. Crete: Gournia (Boyd-Hawes, , Gournia, pl. iv 60, 61Google Scholar): no. 61 has gold-capped rivets like our dagger. A. Triadha (Montelius, , La Grèce Préclassique, pl. 7, 20Google Scholar): from the second tholos tomb. Mochlos (Seager, , Mochlos 37Google Scholar, t. II 52): found close to the surface and probably intrusive. Seager thought that this ‘would appear to belong to L.M. I’ on typological grounds. Mainland: Dendra (New Tombs at Dendra 43, fig. 48, 1); Mycenae (Chamber Tombs 189, pl. vii 27): from its associations not later than the first half of L.H. II; Prosymna (Prosymna 330): three examples, all attributed to L.H. II: that on Pl. II (bottom) has gold-capped rivets like our dagger.

¹³⁶ cf TDoAx 15, fig. 21, which is almost exactly the same length (0·468) as our spear-head (the text mentions five holes in the base of the socket, with a wrong reference to fig. 22. Fig. 21 shows only the usual two ‘rivet’ holes). Also FLMV, Text Vol., pl. D, 4 and 5): these two spear-heads are from Ialysos (Rhodes) and are rather shorter than ours (0·41 and 0·42).

¹³⁷ The wood was kindly identified by Mrs. F. L. Balfour-Browne of the Natural History Museum, South Kensington. A knife from one of the Zafer Papoura tombs had an olive-wood handle (Prehistoric Tombs 80, no. 80 a). For wooden shaft tips well preserved in spear sockets, cf. Prosymna 199, fig. 511.

¹³⁸ The metal was identified as copper and not bronze by Mr. I. W. Cornwall, to whom a sample was submitted.

¹³⁹ PM III 299 ff., fig. 196 and col. plate xxiii; cf. IV 785.

¹⁴⁰ A rather similar toggle, but made of bone, was found in the lowest Copper Age Level I at Alişar in Central Anatolia (Schmidt, The Alishar Hüyük: Seasons 1928–9 (Researches in Anatolia IV) 71, fig. 85, b. 752). Another toggle from Level II of the same site is made of stone: it is fat and dumpy in appearance, but curiously enough has eight facets like our toggle. It is described by the excavator as ‘the most beautiful ornament from Stratum II’ (ibid. 171–2, fig. 221). Schmidt compares these toggles to some, apparently of faience, from pre-Dyn. IV contexts at Abydos, which Petrie thought to be toggles for fastening dress with a loop (Petrie, , Abydos II 26Google Scholar, frontispiece and pl. viii, nos. 141–3). Professor Childe has kindly drawn our attention to another toggle, said to be of ivory, from ‘Troy III’ (Schliemann, , Ilios (1880), 426, no. 536 = S. S. no. 7917Google Scholar); and to one of bone from as far afield as the ‘tholos’ tomb at Almizaraque in the south of Spain (Leisner, , Die Megalithgräber der iberischen Halbinsel (Röm.-Germ. Forschungen 17) (1943), 11, pl. 28, 22).Google Scholar

¹⁴¹ We are much indebted to Dr. Platon for identifying the stones from which the seals are made.

¹⁴² The Reverend V. E. G. Kenna, R.N., informs us that ‘this disparity which produces an elliptical shape seems intentional, and is not usually seen in lentoids of M.M. III–L.M. I, although it occurs in L.M./L.H. III’.

¹⁴³ The interpretation of the bow-like objects must to some extent depend upon whether they are regarded as being different from (so Evans and most authorities) or the same as one or other of the similar looking objects that appear on the following: (1) The gold signet-ring from Dendra (Royal Tombs 56, fig. 34 and pl. xvii). (2) The pin from Shaft-grave III, and the Aegina pendant (reproduced in BSA XLVI (1951), figs. 1 and 7). (3) Some Zakro, sealings (JHS XXII 82 ff.Google Scholar, nos. 57, 61–3, 66, 80–5, 88). For the view that all these are representations of one and the same thing, see most recently Marinatos, in BSA XLVI (1951), 102 ff.Google ScholarCf. Holland, , ‘Mycenaean Plumes’, AJA 33 (1930), 190 ff.Google Scholar

¹⁴⁴ For the knobbed shafts of short hunting-spears or javelins with heads like A J (4) and I (11), see Marinatos, in BSA XXXVII (1936–1937), 187 ff.Google Scholar

¹⁴⁵ Dr. Platon first pointed out to us that these creatures were lions. His opinion is endorsed by Mr. Kenna, who says that there can be ho doubt whatever that they are lions ‘on technical, stylistic, and associative grounds’.

¹⁴⁶ This reference was kindly brought to our notice by Mr. S. Alexiou.

¹⁴⁷ Cf. Murray, , Excavations in Cyprus (1900), 17, fig. 28Google Scholar, for the remains of an (?) ivory quiver with bronze arrow-heads.

¹⁴⁸ Cf. Frödin, and Persson, , Asine 256, fig. 181Google Scholar, for a bone mounting of trapezoidal shape, with running spiral decoration, from a late E.H. stratum (N.B. In the text ‘M.H.’ seems to be misprinted for ‘E.H.’): it is described as a mounting for a sheath.

¹⁴⁹ Mr. Cornwall, who kindly examined the lump, describes it as ‘a lump of mineral, largely composed of magnetite (magnetic iron ore)’.

¹⁵⁰ From tombs at Phaistos came ‘pezzi varii di una materia di colore turchino (κύανος ?) . . . Incerto e l'uso’ (MA XIV 522 and 557), which sound as if they might conceivably be formless lumps of glass or faience comparable to III (28).

¹⁵¹ professor Turner adds: ‘Colleagues or old students of mine (namely, Dr. Eric Preston, Mr. H. P. Rooksby, Mr. B. S. Cooper, Mr. A. C. Jeffkins, Mr. G. Warr) at the Research Laboratories, and laboratories attached to the Glass Works of the General Electric Co., Ltd., Wembley, were good enough to carry out the experimental work of which the following is a summary of the facts: I. The sample is an incompletely fused glass with all the characteristics, such as type of fracture, of a glass. II. The major constituents found by a spectrographic examination are silicon, sodium, potassium, calcium, and magnesium, and the elements found in smaller but appreciable amounts are copper, aluminium, iron, boron, and antimony. It is the copper which is the source of the blue colour. This was the common agent in use also in Egypt for producing the various shades of blue colour. III. A quantitative chemical analysis disclosed the following percentage composition: SiO₂, 64·;40%; A1₂O₃, 2·11%; MnO, 0·25%; Fe₂O₃, 0·36%; TiO₂, trace; BaO, trace; CuO, 1·62%; Sb²O₄, 0·5%; CaO, 9·88%; MgO, 2·56%; Na²O, 14·42%; K₂O, 2·40%; SO₃, 2·03%.

From the composition above no conclusion could be drawn whether the glass was made in Crete or was a somewhat crude piece derived from Egypt. All the samples of Egyptian Eighteenth Dynasty glass so far analysed vary distinctly in composition amongst themselves. There is a certain general pattern in the balance of the constituents, and the composition of this Cretan sample can be said to belong to the same pattern. IV. The whitish material associated with and embedded in crevices of the sample appears to have the same general composition as the glass and not to be merely a decomposition product arising from its weathering. This observation must not be assumed, however, to exclude the probability that the white material does include some constituents resulting from weathering.’

¹⁵² Bird's Nest stone bowls first seem to occur in M.M. I, and flourish into the L.M. period (Seager, Mochlos 38).

¹⁵³ The ‘button’ was found at the bottom level of the stomion (i.e. entrance) of Chamber-tomb 10. The tomb yielded no weapons; and the gold ornaments, etc., from it suggested that a woman had been buried here. On the other hand, all these ornaments, etc., came from pits in the floor of the tomb: and the tomb had evidently been opened, and the objects from the floor itself removed. The material recovered ranged from L.H. II to the beginning of L.H. III in date.

¹⁵⁴ Two objects of a rather similar shape appear attached to the ends of (?) cords floating from the top of a Syrian helmet in an Egyptian representation (reproduced in Homer and the Monuments, pl. xvi, fig. 5). It is therefore conceivable that our pendant had something to do with the crest or plume of the helmet.