The Rytha from Akrotiri and Some Preliminary Observations on their Functions in Selected Contexts

Each type is identified and discussed according to its distribution at Akrotiri. Suggestions are made to explain how various types may have worked. Several deposits are then analysed, taking into consideration the architectural setting, the types and number of rhyta that occur, and other material found in association. An attempt is made to distinguish primary from secondary deposition - i.e., rhyta that were left in their usual place, as opposed to those deposited for safe-keeping at the time of the initial earthquake.

It is suggested that rhyta were used in settings that may be interpreted as specifically cultic and ones where cultic and non-cultic activities are intermingled. A role is suggested for rhyta in beer-making and beer-drinking ceremonies, ritual hand-washing, and anointing with perfumed oil.

The site of Akrotiri has yielded an astounding harvest of vessels, notable both for the variety of their shapes and quantity of examples. Not least among the more unusual classes is a group which, for the sake of convenience, we may call rhyta. Vessels classified as such have a single opening at one end and a smaller hole at the opposite end, usually 0.5 cm in diameter, which would allow a liquid to exit in a thin stream (Koehl 1981).

In a comprehensive study of Aegean rhyta, I have separated the vessels by the categories of class, type and group. At Akrotiri, the convex/conical type of the conical class is the most frequent. Besides a single stone example, AKR #1836 (Warren 1979, Fig. 10, Pl. 22c), all are ceramic and most are locally made. A total of sixty examples has been catalogued, although some are still in situ, and, while visible, remain inaccessible for accurate counting and close study (S. Marinatos 1974, Pl. 6b, 7a). The type is separated into three groups primarily by variations in the shapes of their rims: rounded (Fig. 1), ledge (Fig. 2), and flaring bevelled (Fig. 3), which is the most frequent. The decoration of these rhyta is restricted to several systems, the most common being an overall solid wash. Patterned decorations are limited to zones of tortoise-shell ripple, an overall scheme of light on dark reeds, or, less frequently, zones of spirals. There does not seem to be any correspondence between the individual groups based on rim shape and the systems of decoration.

Three convex/conical rhyta are imports from Crete: AKR #1493 (Fig. 4) (S. Marinatos 1972, Pl. 63), AKR #1494 (Fig 5) (S. Marinatos 1972, Pl. 64) and AKR #2131. AKR #1493, with its thickened, slightly concave rim, is painted with zones of spirals and a foliate band near the base. AKR #1494 has an inturned rim with three ridges on its upper surface. Its painted decoration follows a hybrid-unity system with a wide upper body zone of sprays of crocus flowers, and two narrower zones below, with parallel wavy bands and crocus buds. AKR #2131 has a short, thickened flaring rim and a convex upper body which tapers unevenly to a cylindrical tip. The zonal decoration, of tangent spirals, foliate band, and white dots on solid bands, recalls several rhyta from Gournia (Hawkes 1908, Pl. 7.29, 30). It has, in fact, been argued that AKR #1493 and 1494 were made in a potter's workshop at Gournia (Niemeier 1979, esp. 20-22; 1980, 64-65) and the same may be said of AKR #2131.

The globular class is represented by examples from two types. Two rhyta made from ostrich eggs with faience mouth and base fittings, AKR #1853 and 1854, belong to the narrow-necked tipped type (S. Marinatos 1972, Col. Pl. B, Pl. 81-84). These unusual vessels may also have been imports from Crete (Foster 1979, 152). The narrow-necked, flat-based type is represented by two ceramic specimens, AKR #1542 and 1543 (Fig. 6)(Doumas 1983, 327, Fig. 45). They are decorated as a pair, with dark and red bands and patterned body zones of foliate bands, running spirals, and dots. Their fabric and paint would indicate that they, too, are imports from Crete, although I know of no other LM IA examples of this particular type with which to compare them.

Another apparent Minoan import, of Middle Minoan III date, is a tall type of alabastron shaped rhyton, AKR #5128. The type is well-known from examples on Crete, particularly the Mesara Valley sites of Kommos and Kamilari (Betancourt 1985, 105, Fig. 80A; Levi 1976a, Pl. 219 b, c) and as an import, on Kythera (Coldstream et al. 1973, Pl. 79.64). The specimen from Akrotiri is painted in the standard Middle Minoan III light-on-dark technique, with a wide zone of leaves on the body, and red and white parallel wavy lines below, above the base.

The class of piriform rhyta is represented at Akrotiri by two types: the handleless narrow-necked, or 'peg-topped', and the wide mouth rim-handled. The three vessels which are classified as the former are all Minoan imports. Two, AKR #2591 and 3433, have the standard down-sloping rim, cylindrical neck with low neck ridge, semi-globular upper body and tapering tip. They are painted as a pair, with zones of dots, rosette spirals, and foliate bands (cf. Bosanquet et al. 1923, Pl. 17a). The third, AKR #1685, is more unusual, having a globular body and short tip. Its shape and decoration, of tangent spirals, prefigure an LH IIA rhyton from Mycenae (Wace 1932, Pl. 1.10).

The four wide mouth, rim-handled piriform rhyta from Akrotiri, AKR #2600, 3470, 5458 and 5614, are unique in the LC I/LM IA ceramic repertoire and seem to be the products of an innovative local craftsman (Fig. 8). Their distinctive form, with a rim diameter nearly equal to the vessel's height, and long, cylindrical tip, resembles more than any other type of rhyton the shape of a funnel. They are decorated similarly, with light-on-dark sprays of reeds. Only an LH IIA rhyton from Ayia Irini (Cummer and Schofield 1984, Pl. 86.1561) and a handleless LM IB vessel from Zakros (Platon 1971, 108) have a similar form.

The class of cup rhyta (Koehl 1981, Fig. 2a) is represented by a single example, an unusual spouted cup (S. Marinatos 1976, Pl. 45). More common is the type usually called a 'flower pot', classified as an open flat-based rhyton. This class is composed of types not yet found at Akrotiri: the basket, bucket, and miniature pithos rhyta (Koehl 1981, 179, Fig. 2b). All have a small hole in their base, usually placed near the edge. The 'flower pot' rhyton is a cylindrical vessel with two opposing vertical lug handles (Doumas 1983, Fig. 15a; S.Marinatos 1970, Pl. 51, 52; S. Marinatos 1972, Pl. 29, 45; S. Marinatos 1974, Pl. 6). The vessel is frequently unpainted, although specimens with foliate bands or sprays of reeds occur.

The class of closed, flat-based rhyta is represented at Akrotiri by two types: a piriform jar, AKR #1668 (S. Marinatos 1972, Pl. 66b, right) and 2115, and a piriform jar with an internal funnel (S. Marinatos 1974, Pl. 70). AKR #2115 sits on a raised bevelled base and has two horizontal handles on the shoulder. The perforation is in the centre of the base. It has a zone of tangent spirals on the shoulder and bands below. AKR #1688 has three vertical handles; its flat base is pierced off-centre. It is painted with a zone of rosette spirals on the shoulder and a foliate band on the lower body. While these vessels may be imports from Crete, the only parallel that I know of is from Ayia Irini (K.2147).

The piriform jar with an internal funnel is probably also a Minoan import; every other example of the type comes from Crete (Bosanquet and Dawkins 1923, 285, Fig. 4; Dawkins 1903, 253, Fig. 16; Warren 1980, Fig. 35). All have vertical handles on the shoulder, a small air-hole near the upper join of one handle, and a small hole in the base. The one from Akrotiri is decorated with a zone of oblique foliate bands on the shoulder and a wavy foliate band on the lower body.

Among the most appealing classes of rhyta are the animal-head and animal askoid rhyta. There are two types of animal-head rhyta, the horizontal and perpendicular, both of which occur at Akrotiri. On the horizontal type, the front and back of the animal head continue in a straight line. The back is either completely or partially open and there is a small hole through the muzzle, usually in the lower lip. The type begins in MM II (Betancourt 1985, 100), and is represented at Akrotiri by two boar's head rhyta, AKR #3621 (S. Marinatos 1976, Pl. 57a) and 4274 (Doumas 1977, Pl. 204a). These are partially closed at the back, leaving a small circular opening.

The perpendicular type begins in MM III (Pernier 1935, Fig. 225a, b; Creta Antica, Fig. 343). Here, the head and neck join at a right angle, and the back of the neck is closed. There are two openings, one on the top of the neck or at the back of the head, and a small hole in the muzzle. Two perpendicular lioness-head rhyta have been found at Akrotiri, AKR #116 (Fig. 9) (S. Marinatos 1969, 37. 1, 2) and 1855 (S. Marinatos 1972, Pl. 80). The former is made from a dark, rather gritty clay, and lacks any additional surface treatment. It may have been locally made. The latter is from a finer clay; its surface is burnished and enlivened with paint on the ears, eyes, nose, cheeks and mouth. Perhaps it is a Minoan import; it was no doubt inspired by a metal vessel (cf. Karo 1930, Pl. 111, 118.273). The fragment of a clay bull's head is probably also from a rhyton, perhaps of the perpendicular type (S. Marinatos 1971, Pl. 85b).

Animal askoid rhyta, i.e. rhyta in the form of a complete animal (Koehl 1981, 179, Fig. 1) usually stand on all fours and have an opening on the back of the head and a small hole through the muzzle. The bull from Thera, AKR #563 (S. Marinatos 1970, Col. Pl. A.2, Pl. 8.2, 54), compares well with one from Pseira (Seager 1910, Fig. 7). Both are covered with a white slip and painted with a net pattern on the back. The horns of AKR #563 have been cut as, it seems, were those on the vessel from Pseira.

We might consider now how the various classes of rhyta found at Akrotiri were actually used. In the absence of written sources and scarcity of depictions in art, all conclusions are subject to revision. We may be fairly certain that not all rhyta functioned in the same way or were put to the same uses. Several years ago I proposed specific functions for the narrow-necked and conical classes (Koehl 1981, 181-183). The narrow-necked rhyton works like a pipette; when immersed in a liquid it will absorb a fluid through its lower opening, up to the point of immersion, and then will retain its contents as long as its upper opening is covered. When the cover is removed, the contents will flow out in a thin and easily controlled stream. The vessel is held at the neck, between two fingers, and the opening is covered by the thumb. Many of the rhyta from Akrotiri may have worked in such a way, including those of the globular and alabastron-shaped classes, and the narrow-necked type of piriform rhyton. These rhyta could draw liquid from one vessel and transfer it elsewhere, perhaps into vessels with particularly narrow necks, where hardly a drop would be spilled. They may have had other uses as well. Animal-head and animal askoid rhyta were probably filled and emptied in the same way. Submerged with the muzzle down, the liquid would enter through the hole in the lip and remain inside so long as the hole in the neck was held closed.

The conical rhyton seems to have worked as a strainer and filter, as indicated by a rhyton from Ugarit which had a clay strainer attached to its mouth (Koehl 1981, Fig. 5a, b). An unfiltered liquid would be poured into the rhyton; its impurities would remain on the strainer and a clear stream would flow out from the tip. Thus, the horizontal lugs on the earliest convex/conical rhyta, which date to MM III (Levi 1976b, Col. Pl. 80, 81b; Shaw 1980, Pl. 56 b and d), might have helped fasten a strainer which was set on the inturned rim. By LM IA, a less precarious method was developed, whereby the strainer would sit inside the body of the rhyton, although a vestigial inturned rim occurs on a few LM IA rhyta (Fig. 5).

Having explained how several types of rhyta may have worked, other questions might now be considered. What sorts of fluids were rhyta used to draw or filter? On what occasions or under what circumstances were they used? What happened to the fluids when they left the rhyton? There are still several classes of rhyta whose functions require further attention, in particular the open and closed flat-based types. They will now be considered along with the examination of the contexts and associated finds. The following observations and conclusions are based on preliminary published reports and hence must be viewed as provisional.

Our survey of the distribution of rhyta by individual house or building complex and room, begins with Sector Alpha where, in A.1, four convex/conical rhyta were found. A.2 had one convex/conical and one lioness-head (S. Marinatos 1969, Fig. 10), while West Room 2 contained the bull askoid rhyton (S. Marinatos 1970, Pl. 8b).

The House of the Ladies is problematic, since so many of the vessels remain in situ. Nonetheless, from the quadripartite 'treasury' located under the floor of the west side of room 1, a convex/conical rhyton was found in part 2 and another in part 4 (S. Marinatos 1972, Pl. 5; rhyta not visible). The largest deposit of convex/conical rhyta comes from room 7 (S. Marinatos 1974, Pl. 10a). Beginning with the upper storey, a cupboard on the west wall, near the north-west corner, contains a group of convex/conical and cylindrical flat-based 'flower pot' rhyta (S. Marinatos 1974, Pl. 6b). Just below, on the floor, are a cluster of at least five convex/conical and one cylindrical (S. Marinatos 1974, Pl. 6a). On the basement floor, in the same corner, are a cluster of at least three convex/conical (S. Marinatos 1974, Pl. 7a); above, in the fill between the floors are at least five more (S. Marinatos 1974, Pl. 7a). Along the east wall of the basement floor are two convex/conical (S. Marinatos 1974, Pl. 8a) and in the north-east corner, at least one (S. Marinatos 1974, Pl. 8b).

The rhyta from the West House are fewer in number but more varied in type. In room 3, a convex/conical was found in the cupboard in the north wall (S. Marinatos 1974, Pl. 37a). A lioness-head, a narrow-neck piriform, and a wide mouth rim-handled piriform, were found in room 4, while another wide mouth rim-handled, was found in 4a. From the south-west corner of room 5 comes the piriform jar with internal funnel. A third matching wide mouth rim-handled piriform was found in room 6; a cupboard in the adjoining room 7 contains a nest of convex/conical (Doumas 1985, 63). Three or four additional convex/conical were found in the West House but their exact location is uncertain.

The greatest variety of types and perhaps largest number are from Complex Delta. From the cellar of room 3 comes a single convex/conical (S. Marinatos 1971, Pl. 86c), while another is from the upper floor of room 4. A significant number of rhyta was found in room 9. One convex/conical and four cylindrical 'flower pot' rhyta were found on its northernmost window sill (S. Marinatos 1972, Pl. 45). In the north-west corner of the upper floor was found another convex/conical and from the ground floor, two imported Minoan convex/conical rhyta (S. Marinatos 1972, Pl. 48). An important deposit was found on the ground floor of the adjoining room, 9.1 and the well 21, located just to the west. The well yielded four convex/conical (S. Marinatos 1971, Pl. 86b), while room 9.1 produced nine convex/conical, two narrow-necked flat-based globular, and a spouted cup rhyton. The fragmentary clay bull's head, which is probably from a rhyton, seems to be from the debris below the 'squatter's' level in room 14; a single convex/conical was found on the window sill of the 'mill', room 15. Another significant deposit of rhyta comes from room 16, which produced nine rhyta, all but two of which are probably Minoan imports. The two cylindrical 'flower pot' types from the window sill (S. Marinatos 1972, Pl. 29) are local. On the floor of the room were found a stone convex/conical and a piriform jar in the north-west part, a convex/conical, a narrow-neck piriform, and two narrow-necked tipped globular ostrich egg, from the south-west part, and another piriform jar rhyton, in the south-east corner. Finally, a convex/conical and two boar's head rhyta were found in the western part of the floor of room 17 (S. Marinatos 1976, Pl. 17b).

Building Beta seems to have contained only convex/conical rhyta. Five were found on the southern part of the platform magazine in the basement of room 1 (S. Marinatos 1976, Pl. 25a) and two or three come from the floor of room 2. A cylindrical 'flower pot' rhyton, however, was found just outside room 2, by the hearth (S. Marinatos 1976, Pl. 29b). The last building, Xeste 3, also seems to have held only convex/conical rhyta, three or four of which came from room 11; the location of two additional is uncertain.

Based on this overview, several patterns of distribution may be observed. Not only are the convex/conical rhyta the most frequent type, but they occur in every house or building complex that contains rhyta and probably may be expected in every house or building complex at the site. They usually occur in groups or clusters, sometimes in association with other types. In fact, of the twenty-six contexts considered above, seven contained mixtures of more than one type and of these, only West Room 2 from Sector Alpha and room 4 of the West House did not contain any convex/conical. In discussing the distribution patterns of rhyta on Crete, I noted that there, too, they are often found in clusters, sometimes of a single type, as in room 13 at Zakros, or in mixtures of types, as in the 'Treasury Chamber' at Knossos, and that this pattern is characteristic of both palatial and non-palatial contexts (Koehl 1981, 184-186). I also suggested that the rhyta painted in pairs which occur in the 'town repositories', were distributed from their storerooms and carried in processions composed of two lines (Koehl 1981, 186). Does the same pattern occur at Akrotiri, thus enabling us to posit a similar activity? Let us first examine the clusters of rhyta at Akrotiri.

In Sector A, three of the convex/conical rhyta from room A.1 are painted with a solid wash and one has tortoise-shell ripple. The rhyta in room 7 of the House of the Ladies occur on two storeys, with several different clusters in cupboards and on the floors, while the rhyta from 'Treasury' 2 and 4 were covered over by a slab floor, and possibly were not accessible. The cluster of three rhyta from room 4 of the West House are of different types although a second matching wide mouth rim-handled piriform was found in room 4a. In Complex Delta, room 9.1 contained at least thirteen convex/conical, all of which were painted with a solid wash, two matching flat-based globular, and a spouted cup rhyton. On the floor of Delta 16 was a deposit of seven rhyta, all of which seem to be imports from Crete. The only pair are the ostrich egg; the two piriform jar rhyta are different. Of the five convex/conical rhyta from the magazine in Building B, four are painted solid and one bas zones of tortoise-shell ripple; two from room 2 are solid and one has a tortoise-shell ripple pattern. And none of the convex/conical rhyta from Xeste 3 is alike.

Thus, out of all these deposits, the only ones which seem to resemble the deposition pattern of rhyta on Crete are the two from Complex Delta, rooms 9.1 and 16. These deposits are probably not related to each other and, in fact, come from rooms which are oriented in opposite directions, east and west. However, both face onto open areas or courtyards. Thus rhyta could have been distributed at the appropriate times from these rooms for use in the various types of public ceremonies that might have been conducted in these open spaces. In order to gain further insight into the nature of these ceremonies, an examination of the associated finds is necessary. The deposit in Delta 9.1 seems to provide the least ambiguous evidence. Cups are the most frequent type of vessel from this room, with over twenty-four local and imported semi-globular, straight-sided, footed, and bell-shaped cups discovered (S. Marinatos 1976, Pl. 46c, 48, 50). The room also contained, in addition to the rhyton, at least three large amphorae, a nipple ewer, several conical cups, a Canaanite jar and a spouted piriform jar (S. Marinatos 1976, Pl. 18b, 19). Such an assemblage would appear to comprise the equipment necessary for a drinking ceremony: large jars to contain the liquid, smaller spouted vessels for dispensing, and cups from which to drink. Thus we might imagine a drinking ceremony held in the open area outside the room where the equipment was stored. What role the rhyta might have played in such a ceremony, and what drink was served, are related questions. Keeping in mind my suggestion that conical rhyta were used to filter impurities from fluids such as wine, I propose that another liquid which was filtered through a conical rhyton was beer.

The evidence for beer consumption in the Aegean is scarce, although Evans believed it was standard beverage for the Minoans (Evans 1921, 415). There do exist Mycenaean and Philistine jugs which may have been for straining beer (Dothan 1982, 132-155), and perhaps the LM I strainer cups were also meant to filter floating particles of grain (Bosanquet and Dawkins 1923, 72, Fig. 57). Several different methods for brewing beer are known from antiquity, all of which involve the coarse grinding and maceration of grain, usually barley (Lucas and Harris 1962, 10-16). The grain was dried, re-soaked, and pressed through a sieve. The remaining barley was dried, ground again, and kneaded into dough. Yeast was added, and the mixture set aside to ferment. When fermentation began, the mass was squeezed through a coarse wool cloth or a fine sieve, thereby emitting the drinkable beer. Sometimes the dough might be formed into loaves, which were lightly baked, broken up, and set in water. This mixture was heated and poured into a sieve and the liquid allowed to cool for drinking.

Some of the equipment and raw material used in brewing beer is the same as that for bread-making: storage jars, grinding tools, grain and yeast. However, beer-making also requires basins, sieves, storage jars and pouring vessels. I would also add conical or wide mouth piriform rhyta to this list for use in the final straining process. The beer would be poured into the rhyton, a piece of cloth or cleaned wool having already been dropped into place to catch the impurities, and a clear stream would pass through the tip.

The one complex at Akrotiri that indeed contains all of the necessary accoutrements for making beer is Sector Alpha. Besides the four convex/conical rhyta, A1 contained large jars or pithoi, in which remains of barley have been identified (S. Marinatos 1970, 9), a hearth with cooking pots, grinding stones, basins and jars. More vessels which may have been associated with beer-making come from A2, and include pithoi, grinders and pitchers. One of the most important pieces of equipment, a strainer jar, was also found here (S. Marinatos 1970, Col. Pl. A.1). Perhaps the 'bathtub' found west of A1 was also used (S. Marinatos 1969, Pl. 29). The finds from A3 only augment an already complete picture (S. Marinatos 1969, Fig. 18, Pl. 26.1). It seems likely then, that the beer was processed in A1, and decanted for distribution to the residents of the town in A2; A3 stored supplementary material and equipment. Surely other activities took place in the complex as well. However, another activity relating specifically to the lives of the brewers is revealed by the lioness-head rhyton from within A2 and the stone chalice from between A1 and A2. The religious nature of West Room 2 was recognized by Marinatos from the discovery of the tripod table and nippled ewer (S. Marinatos 1969, 31); the bull askoid rhyton was also found there. I think it is no coincidence at all that the nipple ewer and another globular jug from the same room (S. Marinatos 1971, Pl. 73) are decorated with representations of barley. I would suggest that these jugs, and the bull askoid rhyton, were filled with the beer decanted in A2 which was then poured in libations in West Room 2. Is it too much to speculate that the lioness-head rhyton found in A2 was used to fill the narrow-necked jugs?

In Delta 9.1, the convex/conical rhyta may have been used to give a final straining to the beer about to be drunk outdoors; the narrow-necked flat-based globular rhyton could have filled a ewer, from which libations could have been poured. Another vessel from Delta 9.1, the spouted cup-rhyton, is also revealing. It too, is painted with a representation of barley. Perhaps beer was poured from the nippled ewer into this cup rhyton and the liquid allowed to flow out of its base as a libation.

Among the vast numbers of vessels from the floor of Delta 16 were the seven imported rhyta, mentioned above. Marinatos observed that the material from this room may have been gathered from nearby rooms which were more severely damaged, and placed in Delta 16 for safekeeping (S. Marinatos 1972, 20). However, the size of this room could certainly accommodate numerous objects and various activities. Since the objects seem to have been found in clusters, perhaps the individual clusters represent original groupings. As it happens, the two globular ostrich egg and ceramic narrow-necked piriform rhyta were found in the south-west corner, along with a group of stone and bronze vessels and triton shells (S. Marinatos 1972, 21-22). To the north were found the stone convex/conical and piriform jar rhyta. These two clusters of objects comprise some of the most luxurious classes of material from the site and bring to mind the 'Treasuries' or repositories of prestigious vessels from palatial contexts on Crete (Koehl 1981, 184-185). Inasmuch as we can assume that they are indeed special vessels, it is somewhat surprising that they were originally simply stored on a floor, as their intact condition attests. It is more likely that they, along with the other imported rhyta from the room, originally formed part of a repository located elsewhere, perhaps upstairs. Whether these vessels were actually used or were intended for public ceremonial display (Ventris and Chadwick 1973, 284), is uncertain. However, if they were used in ceremonies, various types are suggested by the rhyta and associated objects.

Four classes of rhyta are represented: conical, narrow-necked tipped globular, narrow-necked piriform, and closed flat-base, although the globular and piriform may be considered as functionally equivalent. The presence of rhyta indicates that the ceremonies would involve fluids, and the presence of the conical rhyta suggests fluids that would need straining. As we have identified a beer drinking ceremony on the east side of Complex Delta, which may have involved community or public participation, it is unlikely that the same liquid would have been used as a sacrament here too. The presence of a marble chalice seems to indicate that drinking may have occurred and, if that is the case, then perhaps wine was the drink. It, too, needs filtering, and the conical rhyta could serve that end. However, I would like to propose that a different sort of ceremony was enacted here, that of ritual hand-washing or anointing, which would make use of perfumed oils.

The existence of a perfumed oil industry in the Aegean is well-documented in the Linear B tablets and has been identified in several archaeological contexts (Shelmerdine 1985). The process of manufacturing perfumed oil is fairly simple, and primarily involves the chopping or grinding of astringent herbs which are heated in oil and to which chopped or ground aromatics are added. The mixture is left to steep and a process ensues involving the repeated steeping and straining of new batches of aromatics and concludes when the desired strength of scent is reached. The oil is poured into narrow-necked vessels for use, storage, or shipment. Shelmerdine has argued that wool, which is mentioned in Linear B tablets listing ingredients for unguent-making, was used for straining (Shelmerdine 1985, 19). Certainly the conical or wide mouth rhyta would be ideally suited for such a role in the perfume industry.

The religious associations of perfumed oil are also documented in Linear B (Shelmerdine 1985, 42-43). That perfumed oil may have been used in a ceremony hypothesized to have taken place either within Complex Delta or in the open area to the west, is suggested by the discovery in the same room of eighteen little Minoan askoi that had been stored inside a jar (S. Marinatos 1972, Pl. 31b). We might reconstruct the ritual as follows: the perfume was poured from the askoi into a bowl, perhaps one or more of the stone ones found with the cluster of rhyta (S. Marinatos 1972, Pl. 71b, 74). Then, one of the three narrow-necked rhyta would have been immersed in the bowl, thus drawing out the perfume; the oil could then be released in a controlled stream over the hands or heads of the participants. A miniature marble vat, also from Delta 16, (S. Marinatos 1972, Pl. 69) may be interpreted as a symbolic representation of a vat used in manufacturing perfume. Here it could serve as a libation vessel, with scented oils spilled from its spout.

The last deposit of rhyta whose functions will be considered are those from the West House. The rhyta in the cupboard of room 7 have been associated with the ritual dining thought to have occurred in room 6. They could have either filled the nippled ewers found with them with filtered liquid, or have received liquid from the ewers and filled drinking cups. Rooms 4 and 5 are believed to have been the primary locations for the ritual thought to have taken place in the house (N. Marinatos 1984, 34ff.). The piriform jar rhyton with internal funnel, which was found in the south-west corner of room 5, may well have been used in the performance of a magical ritual, although it may have been made merely as a vessel capable of controlling the flow of liquid out of its bottom. S. Marinatos's description of how the vase worked needs modification (S. Marinatos 1974, 32). While the small air hole on the shoulder is held shut, a liquid poured into the funnel would remain only in the funnel; by removing the finger, the liquid is then released through the hole in the base. The trick then is that a vessel which appears to have a solid base is actually pierced, and the flow of liquid out of the base can be started and stopped at will. Perhaps the liquid which was poured into this vessel was then spilled onto the tripod table found on a window sill in the north-west corner (S. Marinatos 1972, 18).

The rhyta from the adjacent rooms 4 and 4a seem well-suited for the rituals involving offerings and purifications, as reconstructed by N. Marinatos (1984, 48-49). She also suggests that the red pigment found in the room was used for painting parts of the body red (S. Marinatos 1974, Pl. 59c). These rituals are indeed paralleled by those described in the Ugaritic 'Legend of King Keret' who, in preparing to make a sacrifice, washes his arms and puts on red paint (Pritchard 1969, 143). The finds from rooms 4 and 4a would seem to suggest several activities which may have included washing, anointing, drinking and/or pouring libations. The bronze tripod (S. Marinatos 1974, Pl. 63a) could have provided warm water for the tub, although the washing may have been limited to the arms. Perhaps the lioness-head rhyton was immersed in the tub, drawing in water which was then released over the hands of the recipient standing before the 'sink' (S. Marinatos 1974, Pl. 55b). The next step would involve anointing with perfumed oil, whose existence is attested by the askos from room 5. Another rhyton, perhaps now the narrow-necked piriform, would be immersed in a bowl filled with oil, removed, and allowed to drain its contents over the hands or on the head of the recipient, still standing at the 'sink'.

The last ritual using rhyta may have involved drinking. The strainer found in well 23 (room 4) (S. Marinatos 1974, Pl. 77b), the wide mouth piriform rhyta, and assorted jugs, comprise a fairly complete assemblage for filtering and decanting either beer or wine, in preparation for drinks and/or libations.

Addendum

This brief addendum appears in response to some suggestions made at the Congress concerning the liquid that I proposed was served in the public drinking ceremony believed by me to have taken place in Delta 9.1. While I argued that the beverage was beer, Mesdames N. Coldstream, R. Hood and Dr. C. Macdonald thought that barley wine, or 'John Barleycorn', of which only small amounts are normally drunk, would be more appropriate to the narrow-necked jugs found in West Room 2 and Delta 9.1. Mr S. Hood kindly pointed out to me that Evans had made a similar association between the barley decoration on narrow-necked jugs from Knossos and their contents (Evans 1921, 414-415; 1935, 627). While these proposals are convincing, I would only add that the larger spouted jugs and amphorae from Delta 9.1 might have held beer, and thus two beverages, both obtained from barley, might have been served. Would it be too much to speculate that the occasion for such festive activities might have been a 'Harvest Festival' (N. Marinatos 1987; Hallager 1987)?

--------------------------------------

-----------------------------------