1The 2009 and 2010 campaigns in Zone 4 had two major objectives. The first was to define the limits of Building CD. The second was the excavation of its rooms and spaces down to their latest level of occupation (mostly LM IIIB). Deeper tests were also conducted within certain rooms and around the building. The following chapter is divided in two parts. First, the limits and some of the internal spaces of the building will be examined and several tests discussed. Secondly, the excavation of surrounding areas will be presented.

1. The Building on the Summit

2As in 2007-2008, work in the area was divided in two zones. Excavation in Zone 3, under the supervision of Florence Gaignerot-Driessen, was mostly concerned with the central part of the building, namely rooms 3.1 and 3.8 (see above, figs. 5.1-5.5). In Zone 4, work focused on the northern and western facades (walls D55, D61-65), on spaces surrounding the major central room 3.1 to the north (4.5, 4.8, 4.9 and 4.10), west (4.7) and south (4.14), as well as on the south-southwestern parts of the building (4.11-4.13 and 4.15-4.18) (fig. 5.1).

1.1. The façades

3Even if we require a complete architectural study for a detailed description of the walls circumscribing building CD, they nonetheless deserve some preliminary remarks. Most of these walls were already roughly delineated in the two first campaigns (Gaignerot-Driessen & Letesson 2009: 113; Letesson 2009: 130, fig. 6.22, 132, 134 and 136-137) but, in 2009 and 2010, the northern and western modern terrace walls were carefully cleaned to reveal the alignment of Minoan façades.

4The northern façade is ca. 30 m in length and can be subdivided in three parts: D55, D61 and D62 (fig. 5.2). D55 runs on a NE-SW axis and is perpendicular to the fine facade wall D3 to which it is structurally connected. It is especially well preserved in its southwest part where its masonry recalls that of D3 (small to medium-sized stones constituting the inner face of the wall and larger blocks on the outside). Further to the northwest, the alignment of D55 is slightly different and the wall badly preserved. It was probably partly destroyed or ploughed away when the hill was under cultivation. As noted during the two first campaigns, it seems plausible that during the latest phases of the building, the northeast part of Zone 4 constituted an open-air area (Letesson 2009: 132-133), some kind of a terrace delimited or supported by walls D2, D3 and D55. A 2.50 m² test, opened at the corner between D3 and D55, came upon a clear fire destruction layer and underneath, along D55, revealed a shattered storage vessel whose entire lower part still contained burnt seeds, probably lentils (fig. 5.22-23). As in the western part of the street (area 8.1, see below), this fire destruction level is tentatively dated to the Neopalatial period2. D61 forms the northern limit of rooms 4.5 and 4.8. This part of the northern wall is fairly well built with small to medium-sized stones. At the northwest corner of room 4.8 it makes a recess with D62, to the north of rooms 4.9 and 4.10. Both of these features will be discussed below, in the description of the rooms with which they are associated.

5Running on ca 25 m, the western façade, made of conglomerate and sidheropetra3 boulders and stones of various sizes, shows typical recesses. It seems to have been built directly on bedrock. In its northern part (D48 and 49), it is well preserved and carefully laid out. Unfortunately, further south (D63-65), its alignment is somewhat harder to follow and distinguish from the modern terrace wall. The situation is especially confused in the southwest corner of room 4.7 (feature D31). Below the lowest course of D63, a so-called ‘bathroom bowl’, typical of LMIIIA1/2, may have been intentionally placed which could give a terminus post quem for this part of the western façade (fig. 5.24).

6In all likelihood, the southern limit of building CD was contiguous to the large open space 5.10 (see below) and constituted by the walls D51, D52 and D56 (see rooms 4.15-4.18 below for further description).

7Unfortunately, the southeastern area of the building is still unclear and its limit presumably lies under the dump area of the two first seasons, to the east of rooms 4.12 and 4.13.

1.2. Room 4.5

8Room 4.5 (2.90 m E-W x 2.10 m N-S; 6 m²) is located to the east of room 4.8 (figs 5.1-5). This small space is bounded by the north façade (D61), wall D12 to the east, D11 to the south and wall C1 to the west. This area was briefly investigated during the 2008 campaign and a foundation deposit was found between C1 and D11 (Letesson 2009: 133-134, figs 6.26-27). After a thorough cleaning, excavation in this area mostly limited itself to the west part of room 4.5. The roots of two carob trees were chopped off while excavating but their large stumps were left in place. Most of the material that was found concentrated between these stumps seems to have been redeposited, presumably in one single fill4 (fig. 5.25-26).

9Preliminary pottery analysis indicates a MM II date for the material, as exemplified by several drinking vessels: MM II handled conical cups (or ‘tasses tronconiques’, see Poursat & Knappett 2005: 290; 10-04-1667-OB002, 10-04-1667-OB003, 10-04-1667-OB004, 10-04-1667-OB006, 10-04-1671-OB001, 10-04-1671-OB003), hemispherical cups (10-04-1678-OB002, 10-04-1678-OB004) and carinated cups (10-04-1667-OB005, 10-04-1678-OB001, 10-04-1678-OB004), some with a fine lustrous black slip (10-04-1671-OB002, 10-04-1671-OB007).

10A small limestone figurine (ht. 5.6 cm) was discovered close to the northwestern tree stump and is part of the same deposit (fig. 5.25-27). It looks like a roughly worked pebble and resembles two similar figurines found in Zone 2 in 2009 (see above chapter 4, figs 4.8-9). At the same period, in Quartier Mu, a set of highly stylized stone figurines was found fallen from the upper storey in room V5 (Poursat 1992: 33, fig. 22)5. Similar to our example, they only represent the upper part of the body, with the head and torso but are taller (ca. 14 cm) and much more carefully carved.

11Part of a strainer (10-04-1667-OB001), an obsidian blade (10-04-1667-OB007), half of a plate (10-04-1684-OB001), sherds of a large container in a red fabric with double-axes and ropes in relief decoration (10-04-1689-OB002), as well as fragments of bones and shells, were also uncovered in the fill. The latter also produced two ‘champagne cup’ feet, however, and could therefore hardly be earlier than LM II-III. In the southwest corner of the room, some densely packed stones could have formed a small feature (fig. 5.28). Some 30 cm below, in the southwestern quarter of the room, part of a homogeneous MM floor deposit was discovered (figs 5.28-29). It contained two fragmentary MM II handled conical cups (10-04-1723-OB001 and 10-04-1723-OB002), part of a tripod platter (10-04-1723-OB004), an obsidian blade (10-04-1725-OB001), some obsidian flakes (10-04-1720-SA005, 10-04-1723-SA005), as well as a substantial amount of shells, bone and pieces of charcoal.

12Between the upper fill and this floor deposit, only a few small and rolled sherds were found in a sandier yellowish soil. This layer probably resulted from a progressive fill of the earlier floor. At the level reached at the end of the 2010 campaign, wall D11 was still going down. The lower part of C1 was made of small to medium-sized stones, a type of masonry that contrasts with the larger blocks of its higher courses. Strangely enough, in room 4.5, the inner face of wall D61 is only one course high (fig. 5.28). This may imply that it is actually later than D11 and C1 and may even be modern (terrace wall), but further investigation is required to draw a more definite picture. Just to the west of the MM floor deposit, the upper course of a NW-SE wall started to appear (fig. 5.28). Its relation to the aforementioned stone feature in the southwest corner is unclear but its orientation is different from the surrounding walls. Together with the MM floor deposit, this wall shows the existence of a Middle Minoan occupation on the top of the hill6.

1.3. Room 4.7

13Located just to the west of room 3.1, room 4.7 measures some 23 m² (4.10 m E-W x 5.60 m N-S). Its northwest corner is delimited by two façade walls (D49 to the north and D63 to the west). D27 formed its southern limit and D17 its eastern one. Unfortunately, its northeast corner shows a far more complex situation and was probably disturbed by a sounding conducted by C. Davaras in the 60’s (fig. 5.48)7. It is therefore difficult to have a better picture of wall D42 from which the northern part is lost. It could have been a partition wall within room 4.7 or have formed its northeastern limit8. Originally, the room must have been accessible from the south through a 1 m wide opening between D27 and D179. This door was found blocked with small-sized stones and was probably turned into a small niche. Another blocked entrance may have existed in wall D17, leading to the northwest corner of the large room 3.1. As mentioned above, the way D42 was connected to neighboring walls is particularly unclear but it is tempting to propose that it may have formed the western limit of a small transitional space leading either north to room 4.10 or south to room 4.7, or even simply to both. It is worth noting that the only floor level securely identified in the room was precisely located in the area circumscribed by walls D42, D17 and Davaras’s test10. It was made of trodden earth and showed a strange line of flat pebbles carefully laid out along D42 (fig. 5.30). At the centre of the southern part of the room, a stone with two small cavities (D59; ca. 0.6 x 0.5 m) could have been a rudimentary column base (fig. 5.31-35).

14Generally, the excavation of the area and the reading of its stratigraphy were complicated by the presence of a widespread tumble of medium-sized stones mixed with decayed mud-bricks and mud-brick fragments whose density was particularly high in the southern part of the room (figs. 5.30-31). Throughout the room, an unusual amount of plaster fragments, some painted, was found. Usually they appeared in chunks, with several fragments on top of one another and between the stones of the tumble. It is therefore more likely that they came from wall coatings rather than from a plastered floor on the upper storey. The fact that they were found in larger patches along walls D49, D63 and D27 seems to corroborate this hypothesis (figs 5.32a-b).

15Nevertheless, in 2009 a very large plaster surface was also found in the southwest corner of the room (fig. 5.33a). When it was properly excavated the next season, it turned out that it was actually constituted of several layers of plaster – locally up to six – probably stacked there (fig. 5.33b). The nature of such a heap remains uncertain but it is most likely intentional11.

16Once the plaster was removed, a patch of floor surface was discovered that probably corresponded to the level on which D59 was lying (fig. 5.34) as well as to the area east of D42. Unfortunately, this floor level, probably too damaged by the collapse of the neighbouring walls, was not recovered elsewhere. To the west of the plaster stack, the tumble was particularly dense (fig. 5.35). Its eastern side was first considered as the western wall of the room (D31) but plaster fragments were found beneath it (fig. 5.34) corroborating the tumble hypothesis.

17Throughout the room, some 10 cm below the tumble, a harder surface of trodden earth was found. Very few finds were made on this level apart from a smashed pithos (09-04-1646-OB001) and two lithic tools (09-04-1646-OB002-003), just to the west of D42 (fig. 5.36). On the other side of the latter, just above the tumble, a complete stirrup jar (09-04-1627-OB001, fig. 5.38) was also found. The floor level to the east of D42 produced a broken grinding stone (09-04-1645-OB001), a lithic tool (09-04-1645-OB002) and a tripod cooking pot (09-04-1645-OB003) totally crushed by wall collapse (fig. 5.37). It is also worth mentioning that while the topsoils of room 4.7 were excavated, a Neolithic axe (fig. 1.5) was collected.

1.4. Room 4.8

18Room 4.8 (4.50 m E-W x 2.60 m N-S; 11 m²) is located to the north of room 3.1 and to the west of room 4.5 (figs 5.1-5). It is circumscribed by the northern façade D61 (badly preserved in the eastern part of the room), C1 to the east, D22-C3 to the south, and D23 to the west. In 2009, most of the room was excavated (fig. 5.39), and in 2010, the northern baulk and the modern terrace wall were finally removed to reveal the alignment of D61 and get a better picture of the eastern limit of the room.

19Originally, wall D22 was interrupted by a fine sidheropetra threshold (figs. 5.1, 5.17 and 5.21). This doorway between room 4.8 and large room 3.1 was found blocked. It is also possible that in one of the latest phases of the building, an opening existed in D23, connecting room 4.8 to room 4.9 (see below).

20The stratigraphy of room 4.8 was particularly clear and interesting. Below the superficial layers, a silty yellowish soil, quite different from the type of earth usually found on the hill, was encountered throughout the room. The same layer was also identified in rooms 4.9 and 4.10 and was interpreted as a fill12. For that matter, sherds of a pyxis (09-04-0722-OB001) were found in this layer on both sides of wall D23. In room 4.8, this yellowish fill also produced, from top to bottom, the sherds of the same closed vessel (09-04-0732-OB001) indicating that it probably happened as a one-time event, similar to that in room 4.5. In the southwest corner of the room the fill contained an unusual amount of pumice lumps. Further to the north, it also revealed a fragment of a serpentine vessel (10-04-1744-OB001), part of a small bronze object (10-04-1744-OB002), a lithic tool (10-04-1744-OB003) and a fragmentary quern stone (10-04-1744-OB004). Underneath this layer, a floor made of trodden clay appeared. It was probably laid out subsequently to a rough leveling of the progressive accumulation of debris of daily use on the pebble surface that was found below. This clay floor can be associated with a heavily burnt area delimited by some stones – presumably a hearth – in the western part of the room which also produced a loomweight (09-04-0744-OB001). Through flotation, a soil sample of the area revealed some rodent bones and charred plant remains. Along D22 and close to its doorway, a particular feature (D30, for its location see fig. 5.2) was also identified and could be related to the latest occupation of room 4.8. It was constituted by a thick slab (ca 0.40 x 0.43 m), on top of which a quern stone (09-04-0747-OB001) was originally lying upside down together with the lower part of a storage vessel (04-09-0729-OB001) (fig. 5.40). It presumably worked as an industrial installation. Many shells were also found in the area.

21Already protruding at various spots through the clay surface, a pebble floor was finally reached in the whole room (figs 5.39-41). It was covered with a thin layer of ashy earth and was constituted by several layers of pebbles13. A fine column base (max. diam. 0.35 m), embedded in the pebble floor, was found in the middle of the room (fig. 5.39). The pebble floor produced a loomweight (09-04-0751-OB001), two stone vase fragments (09-04-0751-OB002 and 09-04-0795-OB001), a small and fragmentary terracotta piece with one smooth side that may have been part of the inner coating of a hearth or oven (09-04-0779-SA006), a little piece of lead (09-04-0790-OB001), and a pumice artefact (10-04-1753-OB001).

1.5. Room 4.9

22To the west of room 4.8, a small room (2.40 m E-W x 2 m N-S; 4.80 m²) labeled 4.9 turned out to be of great interest. Its limits are the northern façade (D62), wall D23 to the east, wall D22 to the south, and D47 to the west (figs 5.1-5). The room was first touched at the very end of the 2008 campaign and more closely investigated the following year when the north baulk was also removed to reveal its full floor area. In terms of access, room 4.9 underwent several modifications. During its later phase, it seems that it communicated with room 4.10 only. Wall D47 showed indeed a 0.75 m wide opening in its upper courses, just to the north of a good ammouda block D37 (fig. 5.45c)14. Right in front of this first opening, within wall D23, another doorway which connected 4.9 with 4.8 seems to have been blocked. This modest doorway (ca. 0.70 m wide) was probably contemporary with the acme of the building when all the neighbouring rooms were functional and interconnected (see the description of the LM IIIB deposit below). A blocking made of rubble quite distinct from the surrounding masonry was also found in the northern façade wall D62 (discussed below, fig. 5.82). It is hard to tell if this rudimentary opening (no threshold or sill) was a door or a window but it was most likely blocked in the latest phases of the building15.

23Stratigraphically, room 4.9 is relatively similar to 4.8. Once the topsoils were removed, a yellowish and silty layer was encountered. This fill, also covering the upper course of wall D23, produced mixed material with a high concentration of fine and decorated LM I (A)-B-II-IIIA1 pottery (discussed by C. Langohr, this volume). Below, several medium-sized stones, probably representing collapse from the surrounding walls, were removed and revealed a fine destruction deposit resting on a poorly preserved clay floor. With the exception of the material that was unearthed in 2008, it was mainly concentrated in the northern part of the room, along D62 and D23 (figs 5.41-42).

24Along with the finds described from the first excavation campaigns (Letesson 2009: 136-137, figs 6.31-32), it produced, amongst others, fragments of two or three pithoi (09-04-0731-OB001; 09-04-0767-OB008)16, several cups, bowls, and jugs probably fallen from a shelf along D62 (09-04-0739-OB003-004; 09-04-0761-OB002; 09-04-0767-OB003-006; 09-04-0767-OB021), part of an incense burner (09-04-0767-OB007), a bronze needle (09-04-0767-OB001), several lamp handles, a stone weight (09-04-0767-OB001), a whetstone (09-04-0767-OB024), a piece of stalactite (09-04-0761-OB001), and multiple lithic tools (09-04-0767-OB009-012; 09-04-0767-OB014-17; 09-04-0767-OB019-20 and 09-04-0767-OB021). A detailed pottery analysis of this LM IIIB deposit is given elsewhere in this volume (C. Langohr, chapter 8).

25Some of the soil samples from this context were processed through flotation and already provide some exciting preliminary results. In 09-04-0721-SA004, quite a few charred plant remains, including olives, barley, grapes, lentils, and other fruits/nuts were found. In the same sample there were also some bones and shells. Furthermore, in the outlet of the drain flowing from room 4.9 to 4.10 (see description of the latter below), another sample (09-04-1610-SA004) revealed material that was probably washed away from the floor of 4.9. It contained fish bones, crab remains, sea urchins and a variety of charred plant remains, including legumes, such as lentils, grapes and olives. Even if these soil samples and the material that they produced are currently still under study, it can be tentatively argued there is evidence of consumption and/or food preparation in room 4.917.

26Close to the walls surrounding the room the remains of a plaster floor were identified. In some places, the latter also covered the lower part of walls D22 and D23. The same plaster coating was also used in relation with the drain opening under wall D47. This plastered surface is probably all what is left of an earlier floor which was destroyed, cleaned and leveled to create the latest clay floor of the room18. While cleaning the latter, a rudimentary gourna (max. diam. 0.30 m) was also found (fig. 5.44). It was centrally located in the room and embedded in what seems to be a pebble floor that started to appear underneath the clay floor, like in room 4.8. This hypothetical earlier floor showed obvious traces of burnt material and ashes (fig. 5.44) but remains to be investigated further. A terracotta spindle whorl was also found close-by (09-04-1607-OB001).

1.6. Room 4.10

27Because of the test opened in the 60’s which cut through its southern part, the actual dimensions of room 4.10 are difficult to estimate. Located to the west of room 4.9, this 2 m wide room could hardly be longer than 3.50 m (figs 5.1-5). If one takes as a hypothesis that the length of its N-S axis was similar to that of 4.9, room 4.10 would have measured some 4 m². To the north, it was bounded by façade wall D62, to the west by the façade wall D48, and to the east by wall D47. These three walls are all preserved for more than 1 m high (fig. 5.45).

28Just underneath the ammouda block D37, wall D47 showed the outlet of a built-in terracotta drain coming from room 4.9 (fig. 5.45c)19. Some 0.40 m below, the drain flowed into a drainage channel made of thick upright slabs the southern part of which still extends into the baulk (fig. 5.45c and 5.46). This carefully built gutter runs E-W and probably ran through wall D48 (fig. 5.45a)20.

29It seems that the terracotta drain continued to flow into the room at a later stage. At the time, several large pithos sherds may have been used to guide the water flow. At this level, many plaster fragments (probably coming from room 4.9) were found at the outlet of the drain in 4.10. Furthermore, a slight change in the nature of the soil within the room may have indicated the later channel that probably crossed the room diagonally (fig. 5.47) towards the northern end of D48. The latter showed a gap in its masonry that presumably worked as the outlet of the channel (fig. 5.45a)21.

30Stratigraphically, room 4.10 revealed two destruction deposits. Below and partly mixed with the yellowish fill which was also encountered in rooms 4.8 and 4.9, a first deposit was probably constituted by material fallen from an upper floor, a roof terrace and/or shelves along the walls22. This deposit rests on a thick layer of roof or ceiling material (lepidochoma and small fragments of flat stones) mixed with ash and flecks of charcoal. It produced notably half of a stone mortar (09-04-0739-OB002), a triton shell (09-04-0759-OB002), a tripod vessel (09-04-0774-OB001), a large lekane (09-04-0783-OB001), the rim, handle and shoulder of a large storage vessel (09-04-0783-OB002), the upper part of a stirrup jar (09-04-0783-OB003), and the sherds of a painted globular jar (09-04-0789-OB001) (fig. 5.48a-b and 5.49a-b).

31Below the ceiling material, a much harder layer was reached (fig. 5.47). It showed quantities of pebbles and fragments of plaster. Once excavated, it produced a second destruction deposit with many flakes of charcoal23, bone and plaster fragments, little stones, but also a large number of crushed conical cups (09-04-1630-OB001, 09-04-1630-OB006-007, 09-04-1630-OB013, 09-04-1630-OB016-017) in a very loose and silty earth (fig. 5.50). Once removed, this thick layer revealed the aforementioned stone gutter.

1.7. Room 4.11

32Room 4.11, the second largest room in Building CD after room 3.1 also deserves to be identified as a hall, although less monumental than room 3.1, discussed above (figs 5.1-5). This large trapezoidal room (5.25 m E-W x 6.50 m N-S; ca 34 m²) is located to the west of rooms 4.12 and 4.13, in the southern part of the building. It is bounded by wall C11 to the west, C8 to the north, D38 (of which the northern part was using the bedrock) and D34 to the east, as well as D35 to the south. The room had at least two distinct entrances and a more hypothetical third one. The latter, in the southwest part of the room, would have connected 4.11 with 4.1724. The two other entrances are located to the northwest and to the southeast. The first, a 1.45 m wide opening between walls C11 and C8, connected room 4.11 with a transitional space that ultimately led to sanctuary 3.8, discussed above, as well as rooms 3.2, 3.3 and perhaps even 3.6. The second, marked by a fine sidheropetra threshold (0.75 x 0.22 m; ht. 0.20 m), preceded by a step made of smaller stones, opened out to space 4.18 (figs 5.59 and 5.76)25. The most distinctive feature of the room is definitely formed by the two pillar bases (D29 and D32) aligned on its N-S axis (fig. 5.1-5 and 5.51). Both are similar ammouda blocks (0.46 x 0.46 m; ht. 0.40 m) wedged into place on the bedrock by some smaller stones (fig. 5.52). In its southern face, D32 shows two large cavities that seem natural (fig. 5.53). Even if one could hypothesize that they may have been intentionally carved, their potential function would nonetheless remain particularly elusive. Lying on the bedrock to the northeast of the pillar base D29 was a sidheropetra column base (max. diam. 0.38 m; ht. 0.12 m) (fig. 5.55). It is either fallen from the upper floor or simply reused, as a seat or a pot stand, similar to other stone slabs found in the room. Against the north wall C8 is also a fine sidheropetra slab D66, 0.40 x 0.20 m, which at first looks like a threshold stone, but may also be reused as a small work platform or, most likely, as a special seat (fig. 5.51).

33Between the two pillar bases, a roughly triangular area of hard reddish soil was isolated during excavation (figs 5.51 and 5.54). Just around it, the earth contained ashes, flecks of charcoal, many bones, as well as some terracotta fragments that may have been part of the inner coating of a hearth (09-04-1632-SA005). Samples from this hearth/cooking area were taken for archeo-environmental and micromorphological analysis and are currently under study.

34While excavating room 4.11, it was particularly tricky to find a clear floor level throughout the room. Quite quickly, the bedrock, most likely used locally as a floor surface, was reached at various spots in the room. Unfortunately, elsewhere it was especially difficult to define a clear occupation level. Nevertheless, after the removal of a thin but very hard layer of mud-brick fragments, decayed mud-bricks and sherds, a good LM IIIB destruction deposit was discovered over the entire room (discussed in detail in chapter 8, this volume). Most of the material seems to have been crushed on the floor, probably when building CD was destroyed, but some objects, a little bit higher up in the stratigraphy, could also have fallen from an upper floor or from shelves during the same event (figs 5.55-56).

35In the northwest part of the room, close to the opening between C11 and C8, a snake tube (09-04-0742-OB001), relatively similar to those of sanctuary 3.8 was found (see Driessen-Gaignerot, this volume). Next to the latter, sherds of a large pithos (09-04-1625-OB008) that probably originally contained pumice lumps was also unearthed together with a globular painted stirrup jar (09-04-0757-OB001), two fragments of a terracotta drain or gutter (09-04-0742-OB006, 09-04-0762-OB003) and a fragment of a hypothetical circular hearth in terracotta (09-04-0762-OB003), a small bronze needle (09-04-0757-OB003), two obsidian blades (09-04-0742-OB003, 09-04-757-OB002), several stone tools (09-04-0768-OB002-003, 09-04-0770-OB001, 09-04-0770-OB003, 09-04-1606-OB001), a loomweight (09-04-0770-OB002), and the body of an animal figurine (09-04-0768-OB001). In the northeast part of the room, the deposit produced two tripod cooking pots (09-04-1625-OB001 and 09-04-1625-OB006), a globular stirrup jar (09-04-1609-OB002), several lithic tools (09-04-0771-OB001-002; 09-04-1625-OB010), two obsidian blades (09-04-1603-OB001, 09-04-1633-OB002), a little pendant (09-04-0771-OB003) made from a polished and pierced shell (fig. 5.57), and a bronze needle (09-04-1633-OB001).

36Around D32 and close to the hearth, other discoveries comprised a goblet (visible on fig. 5.53) (09-04-0793-OB001), another globular stirrup jar (09-04-1614-OB001), a painted storage stirrup jar (09-04-1625-OB013), a pyxis (09-04-1614-OB001), a lid (09-04-1625-OB005), a whetstone (09-04-0793-OB002), an obsidian arrowhead (fig. 5.58) (09-04-0793-OB003), two lithic tools/stone weights (09-04-1625-OB003-004), two possible lithic tools (09-04-1619-OB001, 09-04-1619-OB003), a piece of lead (09-04-1614-OB003), and an obsidian blade (09-04-1614-OB004). Close to the southeast doorway of room 4.11, the upper part of a pithos (09-04-1625-OB007) was found together with a grinding stone broken in two (09-04-1648-OB001) (fig. 5.59), a stalactite fragment (09-04-1628-OB001), an obsidian blade (09-04-1628-OB004), and a fragmentary stone weight (09-04-1628-OB005).

37In the southwest part of the room, in an attempt to find a clear floor surface, what seems to be an earlier level was touched (compare figs 5.51 and 5.55). It revealed the upper course of an E-W wall (D39)26, to the north of which was found a round stone with a central cavity that may be a gourna. Around the latter, the soil was filled with sherds and could represent the upper layer of the fill of a pit. Amongst these sherds, the base of a stirrup jar (09-04-1624-OB002), the foot of a ‘champagne cup’ (09-04-1624-OB003), and sherds decorated with circles and dots that may represent parts of a large animal figurine (09-04-1624-OB004) were noticed27.

38While working on defining the northeastern limit of room 4.11, a 1 m² area covered with stone slabs, some of which it probably represent discarded grinding stones, was cleaned (figs 5.51-5.60). It was located just west of room 3.2 (Gaignerot-Driessen 2009: 121, fig. 6.11) of which probably formed part. It probably constituted a work platform. Some large pumice lumps and sherds that probably belonged to the upper part of a large pithos, found upside down against the platform in 2008, were also found.

1.8. Room 4.12

39Room 4.12 is located just to the east of room 4.11 from which it is separated by wall D34. It is circumscribed by wall D33 to the south and wall D36 to the north (figs 5.1-5.5 and 5.61).

40As is the case for room 4.13, room 4.12 was only partly investigated and extends further east into the baulk. Its excavated part is about 7 m². Wall D36 is interrupted by a doorway that connects room 4.12 to room 4.13 via a large ammouda threshold (ca 0.80 x 0.30 m) (fig. 5.61). Another opening may have existed in the south wall D33 towards space 4.18. No clear floor surface could be identified in the room even if the lowest level reached was slightly harder than the upper strata. From the southeast corner of 4.12 comes a gourna (09-04-0777-OB002) associated with an odd terracotta object (09-04-0777-OB001) pierced in its base, perhaps a reused vessel (figs 5.61-62).

41At the level where excavation stopped, some scattered finds were made including the sherds of a kylix (09-04-0784-OB001) that could be mended into a profile, a fragment of a lamp (09-04-0784-OB002), part of a fenestrated stand (09-04-0784-OB003), as well as three fragmentary conical cups (09-04-0786-OB001-003).

1.9. Room 4.13

42Similar to room 4.12 to the south, room 4.13 was only partly excavated and extends further east into the baulk (fig. 5.2)28. As mentioned, wall D36 with its ammouda threshold forms its southern limit. To the north, the room is bounded by wall D57 and to the west by wall D38 of which the northern part is actually formed by the roughly worked bedrock. Three installations were found in the room. The first one (D44, see fig. 5.2) is located in the corner between D36 and D38 and is bordered to the north by an upright sidheropetra block (fig. 5.63). In this little closet-like space were stored a terracotta slab and a collection of lithic tools of various size and textures (09-04-1613-OB002-003; 09-04-1613-OB005-006) hidden beneath an upside down grinding stone (09-04-1613-OB004). D44 probably worked as a storage space for objects related to grinding activities. Some other lithic tools (09-04-775-OB001; 09-04-1616-OB001) and a particular stone weight (09-04-1613-OB001) were also found nearby. The area also produced a small quantity of shells, including at least two or three murex shells. A little bit further to the north, also along D38, the pierced lower part of a large vessel (09-04-1658-OB001) was found partially surrounded by medium-sized stones (D60) at a slightly higher level than the floor surface (fig. 5.64). The function of such an installation is unclear but the stones may have been placed there in an attempt to stabilize the vessel. The third feature (D40), in the northeast part of the room, is a thick ammouda slab (ca 0.45 x 0.30 m; ht. 0.12 m) which formed some kind of platform (fig. 5.65). The area around D40 bore traces of destruction that were not found elsewhere in the room: charcoal, collapsed small to medium-sized stones, mud-brick and plaster fragments, as well as pieces of tarazza29.

43Apart from these features, there was no substantial deposit on the floor of the room. Nevertheless, some finds that are worth mentioning were made in the proximity of the ammouda platform D40: a small stone lid (09-04-1608-OB001) and two loomweights (09-04-1618-OB001; 09-04-1623-OB003). Some diagnostic sherds were also objectified in the apothiki after being mended: the foot and lower part of a red painted kylix with horizontal thin lines (09-04-1608-OB002), the rim and handle of a kylix with red painted spiral decoration (09-04-1608-OB003) and the rim and upper body of a bell krater with octopus decoration (09-04-1608-OB002).

44During the 2010 campaign, a sounding was also opened in room 4.13 in an attempt to localize the assumed continuation of the street level found to the northeast (Letesson 2009: 137-138, fig. 6.33; see also the discussion on space 8.1, below, by T. Gomrée). This 1.40 x 3.00 m test was opened within the room and through its eastern baulk (fig. 5.66). The latter was first dug until the level reached in 4.13 and revealed some large stones that may have constituted the eastern limit of the room. In its western half, the test revealed a layer made of densely packed small and rolled sherds that undoubtedly constituted the fill/leveling required for the laying out of the floor surface reached throughout the room. This layer also produced fragments of mud-bricks and plaster, small pumice lumps, pebbles, bone, a boar’s tooth, shells, two cups (10-04-1741-OB001; 10-04-1746-OB002), a small incense burner or firebox (10-04-1745-OB002) (fig. 5.67), a loomweight (10-04-1746-OB003), and the base of a terracotta beehive (10-04-1746-OB001).

1.10. Room 4.14

45Located between the sanctuary (3.8) and room 4.6, just to the south of the hall (3.1), room 4.14 (2.80 m E-W x 3.50 m N-S; ca 10 m²) was mostly excavated in 2010 (figs 5.1-5). It is bounded to the east by wall D16, wall C7 to the south, C9 to the west, and C18 to the north. A small gap between C9 and C7, in the southwest corner of the room, may have constituted a small niche30. On the other side of the room, bordering room 3.1, it seems plausible that an opening in wall C18 was blocked with rubble, therefore isolating 4.14 from the hall with which it originally communicated. In its northwest corner, the room had two steps (D41) that led up to the doorless spaces 3.10 and 3.9, flanking the shrine 3.8 (see F. Gaignerot-Driessen, above) (fig. 5.68). The lowest step was formed by a large ammouda block (ca. 0.65 x 0.35m; ht. 0.25 m) and packed earth, the second one was built with two flat stones roughly aligned (ca. 0.70 x 0.20 m; ht. 0.18 m). Just to the west of the second rudimentary step, a small area could have formed some sort of a landing. There, along wall C18, a stone with a carved depression – probably a pivot hole – was discovered. It is hard to tell if this stone was in situ but it could attest to the presence of a door. The two steps probably constituted the first flight of a stairway that could have extended above the doorless spaces 3.9 and 3.10 or made a return towards the south, using the thick wall C9 as foundation31.

46Throughout the room, a floor surface made of trodden earth was quickly reached. Within the floor, some stones appear which were afterwards identified as the upper course of two earlier walls (D45 and D46) (see below). D46 only showed the upper face of one of its blocks, close to D41, and may have been reused as a slab in front of the stairs. On the contrary, more of D45 was already visible and it seems plausible that this part of the wall was used as a work platform contemporaneously with the trodden earth floor surface (fig. 5.69). Close-by, quite a few lithic tools were found (09-04-1653-OB001-006, 10-04-1663-OB002-005). Oddly enough, most of the other finds were concentrated in the east part of the room, along wall D16 while almost nothing was found to the west (fig. 5.69).

47This deposit produced an unusual number of lithic tools (09-04-1637-OB001, 09-04-1638-OB002-003, 09-04-1650-OB002, 09-04-1654-OB005-006, 10-04-1681-OB001-005, 10-04-1681-OB007, 10-04-1698-OB001-003), a small pendant in semi-precious stone (09-04-1638/OB001), some loomweights (09-04-1644-OB001, 09-04-1654-OB003, 10-04-1661-OB001), the sherds of at least two small stirrup jars (09-04-1650-OB001; 09-04-1653-OB007), a broken grinding stone (09-04-1654-OB002), part of a triton shell (09-04-1654-OB004), a whetstone (10-04-1661-OB002), a possible bone tool (10-04-1681-OB008), a stalactite fragment (10-04-1681-OB006), as well as some feet of tripod cooking pots.

48A 2 m² sounding was also opened in the northwest part of the room, between D41, D45, D46 and C27 (fig. 5.71). Quite quickly, it appeared that D45 and D46 were the remains of earlier perpendicular walls, while the first ammouda step of D41 and C27, already floating, had to be associated with the later trodden earth floor. In the highest layers of the test, the sherds of a low MM II carinated cup (10-04-1679-OB001) were found. The sounding also revealed quite a number of medium-sized stones. Below, a small destruction deposit was reached. It contained the pierced base of an amphora (10-04-1703-OB001) and a large open vessel (10-04-1703-OB002 and 10-04-1711-OB001) crushed beneath the latter (figs 5.70a-b).

49In all likelihood, this destruction deposit, as well as the walls D45 and D46, testifies to the existence of an earlier occupation on the top of the hill32.

1.11. Rooms 4.15-17 and space 4.18

50These rooms constitute the southernmost wing of building CD (figs 5.1-5). They were only superficially cleaned during the 2010 campaign and will therefore only be briefly described here.

51Room 4.15 (2.50 m E-W x 2.50 m N-S; 6.25 m²) is located just to the south of room 3.5. It is bounded by the poorly preserved west façade wall D65, wall C10 to the north, D53 to the east and D51 to the south (fig. 5.72). An opening between D51 and D52 may have formed an entrance. If exact, it would have connected building CD with the large open area that extends to the north of building 5 (discussed in chapter 6, below). In the northeast corner of the room, close to some kind of niche in wall C10, a sidheropetra slab that may have been a threshold was found in the alignment of D53. Around it is a concentration of medium-sized stones, perhaps tumble. Close to wall D53 and especially D51, the outline of several large vases can be traced, which will be explored during the 2011 campaign.

52Just to the east of 4.15 is room 4.16 (2.20 m E-W x 2.60 m N-S; 5.70 m²) which also revealed a large number of sherds, bone and plaster fragments when the topsoil was removed. This little room is circumscribed by walls D53 (west), D52 (south), C10 (north), and D54 of which the northern part is constituted of roughly worked bedrock (east) (fig. 5.73). Apart from a hypothetical opening towards room 4.15, a doorway connected room 4.16 to room 3.3 to the north. This opening may have been preceded by a step of which the upper part is already visible (fig. 5.73).

53The surface reached beneath the topsoil of room 4.17 (2.70 m E-W x 2.20 m N-S; ca 6 m²) is literally covered with sherds. The space is bounded to the west by wall D54, to the south by D52 and D50 to the west (fig. 5.74). To the north, room 4.17 opens onto niche 3.7 where several small jugs were discovered in 2008 (Gaignerot-Driessen 2009: 122-123, fig. 6.14). Further to the east, 4.17 perhaps also gave access to room 4.11.

54To the east of room 4.17 and south of rooms 4.11 and 4.12, space 4.18 is less clearly defined than the three already mentioned rooms (fig. 5.75). To the west, it is bounded by wall D50, to the north by walls D35 and 33.

55At the eastern end of D35, a fine sidheropetra threshold gives access to room 4.11. Right next to it, sherds of a pithos with rope decoration and of an open vessel with octopus decoration, perhaps a krater (09-04-0785-OB001) (figs 5.76-77) were found together with a lithic tool (09-04-0785-OB002) and a pumice artefact (09-04-0785-OB003).

56Further east, just before the baulk, an opening may also have existed towards room 4.12. Except for a stretch of wall formed by D56, space 4.18 could have been largely open towards the south. The area could thus have formed some sort of porch, and therefore one of the potential entrances of building CD. The cleaning of the area produced a grinding stone (10-04-1734-OB001), a stone vase fragment (10-04-1734-OB002), and part of a terracotta bull figurine (10-04-1734-OB003).

2. Building CD: Surrounding Areas

57Thibaut Gomrée33

58In order to define the outer limits of Building CD, a series of tests was undertaken at several, specific locations around the building. These have all given the label Zone 8, corresponding to the outside areas of zone 4 (figs 5.1-5).

2.1. Area 8.1

59Initially intended as a continuation of the excavation of a paved area discovered in zone 4 in 2008 (Letesson 2009: 138), area 8.1 corresponds to the street located on the southeastern part of zone 4 (figs 5.1). A simple cleaning led to the discovery of some complementary slabs and the existence of a drain longing the south side of the street. The connection between the street and façade wall D2 has also been confirmed: despite the fact that not much attention was given to the alignment of some slabs along D2, it turned out that the street and the wall functioned together. This said, it seems that this last aspect was not essential for the builders, as some slabs are well aligned along wall D2, while on other spots, vacant spaces between the two structures were simply filled with earth and small stones. This fact may indicate that wall D2 predates the street. We were, however, unable to define the connection between the street and the entire east part of zone 4 and more tests are needed.

60The excavation of the western part of the street began with the removal of a layer of tumble that had already been revealed in 2008. This tumble represents a destruction level with traces of a violent fire. It was composed of roughly-hewn blocks, large stones and much mud-brick. It also contained a fairly well preserved charred beam (fig. 5.78), ca 0.60 x 0.15 m (10-08-5010-SA002). This destruction level rested against walls D14 and D15. The material contained in the tumble dates to MM II-LM I and suggests a Neopalatial destruction. As the surrounding walls D14 and D15 form part of the LM III building, it appears that this tumble derives from another structure (unless this part of the building was a reoccupation of an earlier building?). Beneath the layer of tumble, a gap in the pavement along walls D14 and D15 exists, and at a lower level, some coarse pottery was found which actually passes beneath wall D15. This material has not been studied yet but should predate the façade here.

61The pavement in the western part of the street is more irregular and in the southwestern part of the trench, it clearly seems to curve towards the south. It appears then that at this point, the street changes direction, passing either under a later floor of space 4.13, or running alongside the east limit of this latter space. It would not be too daring to think that the western part of the street represented an open area of some form, either a small plateia or an area for outdoor activities.

62The drain runs alongside the pavement in an E-W direction. Its layout is quite irregular, composed of stretches that widen and narrow (max. width: 0.40 m; min. width: 0.20 m) (figs 5.79-80). The material recovered from the drain is quite homogenous and dates to the MM II-LM I periods. Since the LM III period is almost absent from the surface of the street, it may be suggested that the street was built in the Protopalatial period, MM II at the latest, then used during the LM I period, but going out of use in LM III; alternatively, the MM II material from the drain might be residual, whereby the street would be Neopalatial. I would incline towards the first hypothesis. Anyway, the street was related to wall D2 further to the north, but, as noted, this wall is earlier than the setting of the street, which means that it could be relatively ancient. Ultimately further work is needed to clarify these chronological phases, particularly in the eastern area of zone 4.

63In the middle part of the street the drain runs along a course of blocks belonging to a structure which is not yet fully understood (H1, on top and left of fig. 5.80); before widening further to the east, it first narrows considerably while the drain runs alongside a line of blocks at a lower level. Here, both limits are almost contiguous, leaving only a gap of a few centimeters. Although this gap is sufficient for the flow of water, the area is still difficult to interpret. The possibility that structure H1 represents the remains of a wall cannot be excluded; as for the blocks at a lower level, it is possible that they represent the remains of a threshold or a passage, of which only the lower course is preserved. This would also explain the projection of the pavement in the general layout of the street in this location. To confirm this hypothesis, a structure delimiting this presumed ‘passage’ in the eastern part needs to be found. The answer to these questions probably lays under the southern part of the area, which is still covering the southern edge of the street – or its continuation if an open area existed on this part of the hill…

64The work conducted in the eastern part of the street was not particularly conclusive: the pavement was rapidly lost down the slope where the bedrock appears directly under the topsoil. However, part of the pottery collected from this area dates to the Early Minoan period.

2.2. Area 8.2

65Area 8.2 (ca 11.3 x 2 m) corresponds to a test opened along the eastern façade of Zone 4 (wall D3) and the extension of street 8.1. The objective was to identify the nature of the outside area along the façade, and secondly to check if the street unearthed in 8.1 could be found again in any form further to the east or northeast. Unfortunately, no proper archaeological layers were identified but the test led to the clearing of the lower course of the eastern façade D3. It turned out that this wall was directly built on natural bedrock (a wedging stone is also visible under the corner block of walls D2 and D3). If any surface existed here originally, it is no longer preserved. Still, it is interesting to note that the pottery collected from the lower levels against the façade comprises especially Protopalatial or earlier sherds.

66The work in this test also included the dismantling of the modern terrace wall in the northern part of the area, which allowed us to expose the northeast corner of the building. This corner is formed by a substantial roughly trimmed block, almost identical to the one sitting immediately south of it. In front of the corner, beneath the terrace wall, a densely packed level of small stones was reached, more or less at the expected level of circulation. It is not clear whether this represents tumble, a leveling layer or the foundation for a floor surface that is no longer preserved.

2.3. Area 8.3

67The north façade of building CD, mostly build of large blocks, shows a stretch north of room 4.9 where the wall is very thin and may present evidence for a blocked door or window (visible in fig. 5.82). Thus far the blocking has not been removed but a test of ca 2.8 m ² was made (area 8.3) in front of it to explore the possible existence of a street or court level here. The topsoil of the test contained mainly LM III sherds, some of fine quality. This rested immediately on an archaeological layer which comprised a MM II deposit, with vases broken in situ. One of these vases (fig. 5.81, left) was resting on two stones which turned out to belong to an earlier E-W wall, itself connected with a N-S wall (fig. 5.82). This vase was filled with residues comprising a large number of small bones, sampled for further analysis.

68While the excavation needs to be continued, it already shows the importance of the occupation of this area to the north of the building during the Protopalatial period, and this beyond the limit of the LM III complex. It could be seen that this LM III façade actually cut through the Protopalatial deposit which continues beneath, as illustrated by an inverted storage jar and other pots found directly under the façade wall (fig. 5.83).

69The deposit unearthed in 8.3 also comprised a miniature jar (10-08-5041-OB001), some conical cups (10-08-5043-OB001 et OB002), two cooking pots (10-08-5043-OB003, 10-08-5043-OB007), a cup (10-08-5043-OB009), a carinated cup (10-08-5045-OB003), and a fragment of decorated pithos (10-08-5049-OB001).