Abbreviations other than those in standard use

PK I–VII. ‘Excavations at Palaikastro I to VII’. I in BSA 8 (1901–2) 286–316; II in BSA 9 (1902–3) 274–387; III in BSA 10 (1903–4) 192–321; IV in BSA 11 (1904–5) 258–308; V in BSA 12 (1905–6) 1–8; VI in BSA 60 (1965) 248–315; VII in BSA 65 (1970) 203–42. PKU. The Unpublished Objects from the Palaikastro Excavations 1902–6 (BSA Supplementary Paper 1), 1923. PKU II. ‘Unpublished objects from Palaikastro and Praisos’ BSA 40 (1939–40) 38–59. PK Survey. ‘An Archaeological Survey of the Roussolakkos Area at Palaikastro’ BSA 79 (1984) 129–59.

TPHC. Gesell, G. C., Town, Palace and House Cult in Minoan Crete (SIMA LXVII), Goteborg 1985.Google Scholar

¹ The land was purchased for the Greek State by the British School at Athens with a generous grant from the Institute for Aegean Prehistory. The preparatory and legal work was undertaken by N. Kotsiphakis who devoted much time to negotiating with the landowners. Dr I. Tzedhakis has given constant support for the new work and much assistance with the land purchase and securing permission to excavate. The Ephor, Dr N. Papadakis, has enthusiastically shared and supported our interest in the site and his representative, K. Kotzageorgi, was a helpful and interested participant in the excavation. Without the help of all of the above the first season could not have taken place. We are deeply grateful.

² The excavations were conducted under the auspices of the British School at Athens and we are grateful to the Managing Committee for permission and encouragement. The work was funded by the British School at Athens, the Institute for Aegean Prehistory, and the William and Mary Greve Foundation. Excavations were directed by L. H. Sackett and J. A. MacGillivray with the advice of M. S. F. Hood. The architect was J. Driessen, the surveyor was D. Smyth, and A. Thomas acted as Administrative Assistant. Assistants in the field were A. Moore, S. Thome, K. Skourtopoulou, J. Regan, W. Müller, O. Matzari, M. Prent, and C. Crowther with part-time assistance donated by S. Townsend and P. Cronin. The Apotheke was run by C. MacDonald, assisted by L. Beaumont, M. MacGillivray, and the School's vase-mender, P. Petrakis. The programmes were set up and computer operated by N. Winder who was also responsible for the study of the bone material assisted by his wife Heather and S. Wall. The soil sampling and water separation was the responsibility of A. Sarpakis and supervised in the field by C. Rushe and S. Wall with the help of E. McGovern. An average of twelve workmen each day were supervised by the School's foreman, N. Dhaskalakis, and two washerwomen were overseen by E. Petrakis. This report was written by MacGillivray incorporating texts by Sackett and Driessen and reports by C. MacDonald on the pottery, and C. Rushe and S. Wall on the environmental research. The plans are the work of D. Smyth and J. Driessen, the pottery drawings are by C. MacDonald and the terracotta head in FIG.5 is by E. Cassese. The photography is the work of L. H. Sackett, with the exception of PLATE 21 c by W. Müller.

³ Building 1 is orientated 35 degrees to the West of North and seems to point to Petsopha. All other buildings and town blocks at Roussolakkos have an orientation East of North, perhaps pointing to the harbour or because of the local contours of the bedrock.

⁴ Streets in Roussolakkos had varying widths of 1.40 to 4.00 m and most seem to have measured about 2.00 to 2.50 m (PK II, 278; PK IV, 259). For Minoan roads elsewhere see Shaw in Hesperia 51 (1982) 193.

⁵ A stepped street with a gutter alongside it is discussed in PK IV, 259 and illustrated in plate xivb. A similar feature may reveal itself south of Building 1. Drainage problems were, and still are, one of the main civic concerns of the Roussolakkos townfolk and the regular cleaning of the street was a task which may have ceased to have been performed after the destruction in LM IB, when the ‘Harbour Road’ was allowed gradually to silt up to a level above the highest step of the staircase leading up to the entrance to Building I. For drainage problems at Palaikastro see PK VI, 255 and n. 21.

⁶ The rise of the steps from east to west is 0.12, 0.08, 0.15, 0.14 m; the width varies from 0.90 to 0.70 to 0.45 m, the depth of each step being somewhere between 0.40 and 0.50 m. A fine slab 2.00 m long and c. 0.40 m wide is set in front of the threshold. The slab is broken in several places.

⁷ Thresholds with pivot-holes have also been observed in Blocks Gamma and Kappa, see PK II, 290; III, 203.

⁸ The blocking wall, which is sitting half on the threshold and half on the slab in front of the entrance, consists of a few cut ammoudha blocks set on an earth foundation. The wall has a stepped profile. It is for the moment not clear whether the entire entrance was blocked or only part of it. Moreover, it is not impossible that the blocking wall served as a window sill, suggested by the even upper surface of the construction.

⁹ In this case, we could expect another doorway to the west of the present one on the other side of the column. The effect would be that of a small columnar portico similar to that at the entrance to Block X, see PK IV, 284. The threshold also has a fine cutting parallel to the long side of the block on the interior. This may indicate the place against which the door shut.

¹⁰ In Block X also the denudation of the site left only the foundation walls in part of the mansion, see PK IV, 278.

¹¹ Preziosi, , Minoan Architectural Design (1983) 19Google Scholar, the ‘Square-within-a-square’ pattern.

¹² For this type of terrain preparation and plinth construction see Shaw, , Minoan Architecture: Materials and Techniques (ASAtene NS 33, 1971 (1973)) 75–7, 92–3.Google Scholar

¹³ For a summary of building materials used at Roussolakkos see PK Survey, 143–4.

¹⁴ Both gypsum, several blocks of which were present on the surface in this area, and the Cape Sidhero slabs had to be transported from the peninsula over 25 km to the north of Roussolakkos, see PK Survey, nn. 25 and 27.

¹⁵ Block N also preserved some plaster in position on its outer façade, see PK VI, 253.

¹⁶ One of the ammoudha blocks deserves special mention. Although it is broken on one side, its dimensions are still substantial: 0.80 × 0.45 × 0.25 m. Its special interest lies in the fact that it shows a cutting along the short side, comparable to the blocks discussed by Shaw, op. cit. (n. 12 above), 172 fig. 203a. In contrast to these, the present block does not preserve dowel-holes. We thank V. Fotou for the suggestion that it may have been used as support for the bottom step in a staircase.

¹⁷ This may be seen in an unpublished photograph in the Archives of the British School at Athens.

¹⁸ Platon, , The Palace of Zakros (1971) colour plate 119.Google Scholar

¹⁹ We are still uncertain as to the interpretation of the very large and fine Cape Sidhero slabs in EQ 94, apparently used as paving. Since one of these slabs overlies part of the west plinth of Building I, PLATE 21a, they would appear to be of later construction belonging to a floor laid down when Building 1 was in ruins. However, the quality and appearance of the paving slabs suggest that they should somehow be related to the original construction of Building 1. It may be that the slabs were laid out on top of the plinth but against the ashlar wall set on the plinth. If the latter, however, it would be a unique feature in Minoan architecture. Further stratigraphic testing is necessary to elucidate this important point.

²⁰ See Popham, , The Destruction ofthe Palace at Knossos (SIMA 12) (Goteborg 1970)Google Scholar fig. 3 nos. 1–6, 8, 9, plate 1 for cups similar to PK/86/131.

²¹ For Kommos see Shaw, , Hesperia 48 (1979) 170–1CrossRefGoogle Scholar; for Chania see Tzedhakis, and Hallager, , AAA 6 (1973) 440–2Google Scholar; 15 (1982) 22; 16 (1983) 7; 17 (1984) 14–17.

²² Areas 6 and 7 have been provisionally interpreted as courtyards, the first because of the massive slab pavement, the second because of its series of fine pebble floors built one on top the other (winter rains may have caused the periodical renewal of the pebble floor).

²³ There are as yet no clear indications of re-occupation to the east of Corridor 1, except perhaps for the rectangular construction in ET95. Measuring 0.70 × 0.55 m it is set over the top of the LM IIIA2 rubbish layer and so may be contemporary with the new construction over Building 1, otherwise the area was devoid of building after LM IIIA2.

²⁴ See PLATE 22b for a view of head in position from the west.

²⁵ There is a slight difference in level between the floors of areas 1 and 4 which is overcome by a step in the connecting doorway.

²⁶ It seems likely that the north-west wall of Room 4 had three phases: in the first Room 5 may have acted as a corridor leading to 8, in the second an oblique wall was put across the corridor forming a narrow doorway with the north-west wall of Room 5, in the third phase the doorway was blocked.

²⁷ Melas, , The Islands of Karpathos, Saros, and Kasos in the Neolithic and Bronze Age (SIMA 68) (Goteborg 1985) 104 figs. 39.1034Google Scholar, 102.1033–6.

²⁸ For Minoan ovens see, Hallager, and Tzedhakis, , AAA 17, (1984) 17 fig. 14Google Scholar and Shaw, , Hesperia 48 (1979) 157 fig. 3.CrossRefGoogle Scholar

²⁹ By the time of the new construction in LM III Building I must already have been stripped of its ashlar since no substantial blocks were re-used in the LM III construction. Only the plinth seems to have survived. It is possible that the LM IIIA occupants of Roussolakkos are to blame for this removal. The earlier excavators regarded ashlar construction as characteristic of LM III Roussolakkos (PK IV, 281–5) and LM IIIA seems to have been a period of great prosperity for the town, see Kanta, , The Late Minoan III Period in Crete (SIMA LVIII) (Goteborg 1980) 189–92.Google Scholar

³⁰ A modern trackway prevented any extension of the excavation to the south. The top of the walls between Rooms 4 and 5 and 5 and 8 was visible before excavation.

³¹ The plaster floor in Room 7 was found about 0.20 m below the surface. Fortunately, this area seems not to have been used for agricultural activities which explains why the paving in Room 5 was preserved intact.

³² McEnroe, , AJA 86 (1982) 14 classifies the houses at Roussolakkos in three categoriesCrossRefGoogle Scholar: type 2a, houses which combine residential and storage/industrial rooms within the same perimeter on the ground floor and adopt certain palatial architectural features; while those of type 2b do not; type 3 houses are very similar to the previous but are much smaller. McEnroe counts at least 3 type 2a, 3 type 2b, and 15 type 3 houses at Roussolakkos; the smaller houses were confined to certain Blocks (Ksi and Pi) or were squeezed into spaces left among the larger buildings. Our Building 2 would, because of its light-well, be a type 2a house in McEnroe's classification.

³³ In contrast to Building 1, the plinth only comprises one course. The wall is a maximum of 1.00 m thick, the largest slab being 0.90 m long. Again in contrast to the slabs in the plinth of Building 1, the slabs of this building are set with their long sides along the wall and not across.

³⁴ For another example of light-well with drain at Roussolakkos see PK II, 279, 293.

³⁵ See, Astrom, and Holmberg, (CVA Sweden 3, 1985) 31–2 pl. 9.16, 17Google Scholar; and Forsdyke, , Prehistoric Aegean Pottery (1925) 82 A478.Google Scholar

³⁶ This piece is similar to a bowl from Block X (PKU, 27 fig. 17) and is exactly similar in decoration to an example from the Zakros Pits, see JHS 23 (1903) 251 fig. 14.

³⁷ JHS 23 (1903) 249 figs. 1, 3; BSA 62 (1967), pl. 78a.

³⁸ Paton, and Myers, , JHS 18 (1898) 209–17 discuss early oil manufacture.CrossRefGoogle Scholar This hypothesis is supported by the fact that the area where the vat was discovered was the only one to produce a thick oil-slick on the surface of the trench after it had filled with rainwater one night.

³⁹ Above n. 32.

⁴⁰ Lamb (CVA Great Britain 2, Cambridge 2) (1936) 7 pl.iii. 25; Forsdyke, op. cit. (n. 35 above) A593.

⁴¹ MacGillivray, , ‘Cycladic Jars in MM III Contexts at Knossos’, in Hagg, and Marinatos, (eds.), The Minoan Thalassocracy: Myth or Reality (Stockholm 1984) 154 fig. 2.Google Scholar

⁴² It was found by Pat Cronin who very kindly brought it to our attention. It was studied in Palaikastro by J. Driessen, who wrote the following comments, and then sent to the Siteia Museum.

⁴³ See Olivier, J.-P., Le Monde Grec. Hommages à C. Préaux (Brussels 1975) 441–9Google Scholar, who is able to read ten Linear A signs (DA, I, JA, KI, PI, RI, RO, TA, TE, and SU), whereas Godart, L. in Aux engines de l' Hellénisme. Hommages à H. van Effenterre (Paris 1984), 121–8Google Scholar, accepts thirteen signs (the same as Olivier plus PA, SE, and TO). These correspondences are based on the presence of homomorphic groups in Linear A and B. The present sherd is the third document inscribed in Linear A from Roussolakkos and forms an addition to Godart, and Olivier, , Recueil des inscriptions en linéaire A (Paris 1983)Google Scholar (Gorila 4). See also Gorila 5, 110. From Block B came a tablet (PK II, 284–7; PKU, 145 fig. 127) and there is also a fragment of a cup with painted inscription, see Boardman, , BICS 5 (1958), 12.Google Scholar

⁴⁴ See McArthur, J., A Tentative Lexicon of Mycenaean Place-Names I, Annex to Minos 19(1985) 90–2.Google Scholar It is now convincingly argued that no east Cretan place names are mentioned in the Knossos archives and that setoija, which was initially equated with Siteia (e.g. Chadwick, and Baumbach, , Glotta, 41 (1963) 243)Google Scholar, may actually have been Malia (see Bennet, in AJA 99 (1985) 243.Google Scholar