The Final Neolithic to Early Bronze Age Transition 201 13 The Final Neolithic (Late Chalcolithic) to Early Bronze Age Transition in Crete and the Southeast Aegean Islands: Changes in Settlement Patterns and Pottery Krzysztof Nowicki Knossos and Phaistos have long been regarded as the most important sites for the reconstruction of Neolithic settlement patterns in Crete (Tomkins 2000: 76). Knossos has yielded a long continuous sequence from the first Neolithic settlers to the beginning of the fourth millennium BC (Evans 1964), but the crucial transition between the Neolithic and the beginning of the Bronze Age (i.e., the second half of the fourth millennium BC) is unfortunately poorly recorded at this site (Evans 1994: 19). Phaistos represents a different part of the story. The settlement on the later palatial hill seems to have been abruptly founded in the latter part of the FN (Vagnetti 1972–3), as was a similar site on the summit of the Gortyn acropolis (Vagnetti 1973). The defensible location and scale of Phaistos indicate that its inhabitants either moved from a low lying large settlement (still unknown) with a history similar to Knossos, or that it was settled by a number of families coming from hamlets and individual farms scattered over this part of the Mesara. But what was the reason for such a change in the local settlement pattern, and was it really a local change? The phenomenon of moving settlements to high hills, during roughly the same period, has been observed elsewhere by Sinclair Hood who explained it with reference to possible security problems (Hood et al. 1964: 51; Hood and Warren 1966: 185). Lucia Vagnetti, however, wanted to see the process as connected with climatic changes (Vagnetti 1972–3: 132), but such a hypothesis does not explain the general changes in settlement pattern seen throughout the entire island. In this paper I will show that the transition between the Neolithic and Early Bronze Age in Crete was not smooth and gradual (Vagnetti 1996: 40) and was not restricted to changes in pottery manufacturing and metalworking as has been often argued. Changes in the second half of the fourth millennium BC are most strongly seen in settlement patterns and social structures (Nowicki 1999). A stricter attitude to territorial rights and a greater definition of identity also probably occurred at this time. It will be argued here that this was not an internal process 202 Krzysztof Nowicki that took place within isolated Cretan communities, but was stimulated by outside influence. It was probably directly attributable to physical intrusion by outsiders and most probably by immigrants arriving in a large number from the east. The links between the changes in settlement organization in Crete and the situation in the region east of it in the fourth millennium BC are supported by fieldwork undertaken during the last few decades by A. Sampson in the islands of the Dodecanese (Sampson 1987; 1988) and by the author in Crete and its offshore islets (Nowicki 2002). These surveys have shown that a short-lasting population peak in the Dodecanese in the latest Chalcolithic period, contemporary with the late FN in Crete and the Cyclades – the phenomenon, first recorded by Simpson and Lazenby (1973), was followed by an abandonment of sites and depopulation. Quite the opposite pattern is found in Crete, where settlement numbers increased substantially during the last few centuries of the fourth millennium, particularly in the coastal zones. Many of these newly founded sites were located on defensible ridges. In Crete this sudden increase in the quantity of sites marked the beginning of the long-lasting development of Bronze Age communities. Such ‘complementary’ changes in two neighbouring areas prompt the conclusion that the processes were related to each other and a plausible scenario is that at least some of the Dodecanesian population may have been responsible for the foundation of the latest FN sites in Crete. This hypothetical migration was probably only one element of a much broader movement of west Anatolian people towards the west. The process probably started earlier, but entered a decisive stage during the second half of the fourth millennium. The most important result of the dynamic interaction that took place at that time, between the technologically more advanced and socially more stratified western Anatolian immigrants and the somewhat conservative south and central Aegean islanders, was the foundation of a completely new form of settlement organization and new communication networks. Both elements were essential for the later emergence of Bronze Age societies in Crete and the Cyclades. The geography of FN and early EM I settlement can only be fully appreciated in the field and the phenomenon will be better understood only when more new evidence is published. The Neolithic settlements at Knossos and Phaistos, however important, cannot answer all the problems related to the transition between the Neolithic and the beginning of the Bronze Age (e.g., Papadatos this volume). Some processes, especially hypothetical migration(s), are better illustrated by the great number of sites that until recently were either unknown or almost completely ignored in general discussions of the period. Archaeological evidence for this process is now much more substantial than it was a few decades ago, when in 1978 Vagnetti and Belli were able to mention only six open air settlements and five ‘undetermined’ open air sites (Vagnetti and Belli 1978). However, this evidence is underestimated and often confused, partly because of the lack of a clearly defined chronology and partly because of the unjustified reservations of some scholars about the value of data from The Final Neolithic to Early Bronze Age Transition Figure 13.1. Map of Crete showing location of FN sites (defensible settlements only apart from Gaidouronisi and Koufonisi). (1–24: see Figure 13.2); 25. Traostalos; 26. Azokeramos Pentalitro; 27. Azokeramos; 28. Karoumes Kastellas; 29. Plakalona; 30. Palaikastro Kastri; 31. Palaikastro Petsofas; 32. Itanos; 33. Cape Mavros; 34. Itanos Alatopatela Site 12; 35. Itanos Site 17; 36. Adravasti Endichti; 37. Magasa Vigli; 38. Palaio Mitato; 39. Kalamafki Kypia; 40. Petras Kefala; 41. Papadiokampos Agios Ioannis; 42. Kavousi Azoria; 43. Monastiraki Katalimata; 44. Vasiliki Kefala; 45. Vainia Stavromenos; 46. Panagia Paplinou Rousso Charakas; 47. Koutsounari Karfi; 48. Gaidouronisi Belegrina Bay; 49. Kataprosopo Bay; 50. Vrokastro; 51. Vathi; 52. Anatoli Pandotinou Korifi; 53. Anatoli Schistra; 54. Myrtos ‘Charakas’; 55. Mythoi Kastello; 56. Faflagos; 57. Tzermiado Kastellos; 58. Krasi Armi; 59. Malia Profitis Elias; 60. Milatos Kastello; 61. Milatos; 62. Drepanon; 63. Dermatos Gorge; 64. Dermatos Kastrokefala; 65. Tsoutsouros; 66. Tsoutsouros Mandalos; 67. Lenda Leontari; 68. Apesokari Vigla; 69. Gortyn; 70. Phaistos; 71. Melidoni; 72. Viran Episkopi; 73. Kerames Tseroni; 74. Drimiskos Agios Georgios; (75–87: see Figure 13.3); 88. Patsianos Kefala; 89. Vraskas Lakoudi; 90. Imbros Gorge; 91. Anydroi Profitis Elias; 92. Palaiochora Nerovolakoi; 93. Chrisoskalitissa; 94. Sfinari Korakas; 95. Phalasarna ‘Acropolis’; 96. Kastelli (Trachilos) Selli; 97. Tsikalaria; 98. Stavros Leras; 99. Limnes Kefali. 203 204 Krzysztof Nowicki Figure 13.2. Map of southeast Crete with FN defensible sites. 1. Zakros Kalyvomouri; 2. Zakros Gorge Kato Kastellas; 3. Kefali Agias Paraskevis; 4. Koufotos; 5. Kokkino Froudi; 6. Voukoliades; 7. Schinokefalo; 8. Kastellas Xerokampias; 9. Trachillas; 10. Xerokampos North; 11. Xerokampos Kastri; 12. Xerokampos Amatou (Paranoma, Kastellakia); 13. Agia Irini Kastri; 14. Livari Katharades; 15. Goudouras Kastri; 16. Goudouras Kastello; 17. Goudouras North-West; 18. Agia Triada Petrokopio; 19. Mesa Apidia; 20. Ziros Rizoviglo; 21. Ziros Patela; 22. Lamnoni ‘Spilia’; 23. Lamnoni (Branigan’s Site 23); 24. Koufonisi. unexcavated sites. Now it is possible to talk of more than 100 well defined open air late FN settlements from just the group of sites with defensible characteristics alone (Figure 13.1; Nowicki 2002; 2004). A great number of these sites are located on the coast and two regions are worthy of special attention for their unusually high concentrations of settlements. The first is the eastern part of the Siteia peninsula (Figure 13.2) and the second is the south coast of the Rethymnon isthmus (Figure 13.3); several smaller clusters extend as far as the western coast between Palaiochora (Figure 13.4) and Phalasarna. A Final Neolithic Refuge Site at Katalimata New evidence relevant to the interpretation of changes in settlement pattern in Crete during the FN was revealed by excavations at Katalimata, located on the isthmus of Ierapetra (Nowicki 2002: 16–20; 2008). There can be no doubt that this inaccessible location must have been chosen for security reasons (Figure 13.5). Katalimata is located on rocky ledges, high on the northern cliff of the Cha gorge (ca. 290 m.a.s.l.), a location that is extremely difficult to access, but with excellent visibility over the Ierapetra plain. Neolithic deposits were recorded at two places on Terrace C. One consisted The Final Neolithic to Early Bronze Age Transition 205 Figure 13.3. Map of the Plakias area with FN defensible sites. 75. Gianniou Plati; 76. Lefkogia Timios Stavros; 77. Lefkogia Modi; 78. Ammoudi Skinias; 79. Damnoni; 80. Plakias Korifi (Paligremnos); 81. Atsipades Korakias; 82. Mirthios Kirimianou; 83. Sellia Kastellos; 84. Sellia Kabana; 85. Sellia Kefala; 86. Sellia Kastri; 87. Argoules Chalepa. Figure 13.4. Map of the Palaiochora-Chrisoskalitissa area. 91. Anydroi Profitis Elias; 92. Palaiochora Nerovolakoi; 93. Chrisoskalitissa. of an undisturbed stratum, 0.10–0.25 m in thickness, lying immediately on the bedrock. The soil was mixed with ash, sherds and animal bones. Chipped stone, ground stone and bone tools were also found. The second deposit was recorded in the eastern part of the terrace. Here, only a few tiny particles of charcoal were 206 Krzysztof Nowicki Figure 13.5. Monastiraki Katalimata from southwest (C: Terrace C with FN deposits). Figure 13.6. FN Pottery from Monastiraki Katalimata. noted, pottery was occasional and neither animal bones nor ground stone tools were found (Nowicki 2008). The pottery from Katalimata represents the native Cretan tradition of the FN, with features dating the site towards the end of that period (Figure 13.6). It is often burnished or polished on the surface, although not so well as the best examples from Knossos and Phaistos, being rather dull and ranging in colour from yellowish brown, through reddish brown, dark red, brown and black. Two sherds are decorated with incised ‘seam pattern’. The core is most often dark olive grey to black. Inclusions consist of fine to medium grits of phyllite, sand and very fine white chalk. The shapes are mainly bowls (open, globular, carinated), cups and larger hole-mouthed jars; a few handles are of the strap type. The closest parallels for the Neolithic pottery from Katalimata regarding surface treatment and form come from Knossos Stratum II, the lower stratum at Phaistos and from the surface of several unexcavated sites in the Ierapetra region, such as the rocky knoll of Pandotinou Korifi, south of Anatoli, and Vainia Stavromenos (the earliest material) (Figure 13.7; Todaro and Di Tonto this volume; Tomkins this volume). Although the Neolithic pottery from Katalimata lacks several distinctive elements of the earlier Phaistos FN assemblage, such as incised decoration, encrustation with red ochre, scoring and pattern burnishing it seems that we should place the Katalimata pottery in the period contemporary with the Phaistos and Gortyn material, close to The Final Neolithic to Early Bronze Age Transition 207 but probably after the end of the old Knossian ‘Late Neolithic’ phase (i.e., postStratum IIB) and earlier than Nerokourou. It would thus belong to the period labelled by Vagnetti as the ‘early stage’ of the traditional Cretan FN period. I have suggested elsewhere that this be labelled as FN I (Nowicki 2002) and according to the revised Cretan Neolithic chronology by P. Tomkins this should be labelled FN III (Tomkins this volume; 2007). The location of Katalimata raises the question of what the reasons were for occupation of such a defensible place in the Neolithic period. Considering the similarities between the Katalimata material and that from Phaistos and Gortyn, we should look for the answer perhaps not in the local situation in the Ierapetra isthmus, but in the circumstances that led to a shift of population to high hills throughout Crete. The Topography of Late Final Neolithic Settlement During FN III, Phaistos and Gortyn in the Mesara, like Katalimata and Pandotinou Korifi in the Ierapetra region, were settlements which continued the Cretan Neolithic tradition. Changes in pottery, however, take place in the next (and final) phase of FN and a non-Cretan origin for some of them must be considered. Among such non-Cretan elements is the so-called ‘cheese-pot’ – a characteristic type of vessel with holes pierced before firing under the rim – that is diagnostic of Late Chalcolithic 3 and 4 and the transition to the Early Bronze Age all over the Aegean (Atkinson et al. 1904: 84; Caskey 1972: 359; Renfrew 1972: 141; Sampson 1984; 1987; Broodbank 2000: 83). The appearance of these vessels in Crete indicates external influence. The type is common at many Cretan coastal sites founded shortly before the beginning of the Bronze Age, that is during the latest FN (Figure 13.8), elsewhere termed FN II (Nowicki 2002) and FN IV according to Tomkins’ revised Cretan Neolithic chronology (see Tomkins this volume). The most characteristic features of these newly founded latest FN (FN IV) settlements in Crete are: (1) their defensibility, (2) their preference for coastal locations and (3) their specific concentration in certain areas, including regions which were marginal for later Bronze Age communities. These may indicate: (1) a serious tension between different groups within the population, (2) a strong orientation towards marine activity, and (3) temporary colonization of the coastal areas without considering their environmental potential to support long-lasting occupation. The latter two factors alone may indicate that the ‘colonization’ took place from the sea. All these arguments, among other factors, suggest a causal relationship between the establishment of the FN III settlements of Phaistos and Katalimata type and the historical circumstances which preceded and led to the foundation of new sites during FN IV, a selection of which will be presented below. The appearance of these new settlements can be traced all over Crete, but is 208 Krzysztof Nowicki Figure 13.7. FN pottery from Anatoli Pandotinou Korifi and Vainia Stavromenos. Figure 13.8. ‘Cheese-pot’ fragments from Cretan FN sites. most characteristic along the coast with the same or similar patterns of site location occurring between the Palaikastro plain in the far east and the Phalasarna plain in the far west. Study of surface material has allowed the recognition of several new pottery groups. The difference between the old Cretan pottery Neolithic tradition and these new groups can be seen when we compare the pottery from Pelekita, Katalimata, and Pandotinou Korifi, on the one hand, with the Red Ware that dominated along the south coast between Zakros and Phalasarna, on the other. Other types of pottery are more common along the northern coast among which two groups are the most distinctive. The first group, more brown than red, with sandy inclusions and organic temper, is harder than the Red Ware and thus similar to the Late Chalcolithic pottery from the northern and central Dodecanese. The second group, with a large amount of calcareous and marble inclusions, seems to be linked to Cycladic FN–EB I pottery. The latter conclusion is further supported by the occasional (but very rare) appearance of the bowl with rolled rim on the northern coast. This shape, however, is absent or extremely rare along the south coast. The largest concentration of latest FN sites (FN IV) has been recorded in the East Siteia region. Here all the available bays and other convenient natural The Final Neolithic to Early Bronze Age Transition 209 Figure 13.9. Plan of Livari Katharades. harbours were taken over by new settlements situated in elevated places. The population of the settlements ranged from two or three households to over twenty households. Vokotopoulos has suggested that an average settlement in the Zakros basin was about 0.2 ha but some sites extend over an area of about 0.9 to 1.0 ha (Vokotopoulos 2000: 130). Among the largest are Zakros Gorge Kato Kastellas (ca. 0.8–1.0 ha), Xerokampos Kastri (ca. 0.6–0.8 ha) and Agia Irini Kastri (0.8–1.0 ha). The distribution of pottery, stone tools and in some cases architectural remains indicates that the full extent of the estimated area of these sites was occupied by domestic houses and their dependencies. However, precise numbers of households and forms of spatial organization are difficult to reconstruct because of a lack of comparanda from excavated sites. Many of these settlements were short lasting and much of their architecture was constructed of perishable material without any stone walls visible on the surface. On the other hand substantial stone architecture is preserved at several sites, such as Livari Katharades (see Schlager 1997: 15; 2001: 160), indicating a rather compact and defensive structure for these sites, with houses attached to each other and only small open areas left between individual units (Figure 13.9). Earlier FN settlement in the coastal region of eastern Crete is characterized by very small sites, representing one or at most a few families, living in more or less temporary houses, as a rule on low hills, coastal terraces and in caves, such as Pelekita (Figure 13.10) on the coast, and Voivoda in an inland valley. The newly founded FN IV sites were larger and as a rule located on defensible ridges on rocky promontories or hills dominating a coastal plain. The natural defensibility of these sites was often complemented with fortification walls (e.g., 210 Krzysztof Nowicki Figure 13.10. Pelekita from south (the cave located where path ends). Figure 13.11. Plan of Kokkino Froudi. Goudouras Kastello, Kokkino Froudi, Xerokampias Kastellas and Alatopatela – Site 12) (Nowicki 1999; Vokotopoulos 2000; Greco et al. 2002). The walls (usually 1.0 to 1.3 m thick) were constructed of large boulders, along one or two sides of a settlement, where the access was the easiest. Other sides were either entirely or partly defended by cliffs and very steep slopes (Figure 13.11). Moving from the north to the south, the following coastal settlements were founded at this time: Palaikastro Kastri, Karoumes Kastellas, Kato Zakros The Final Neolithic to Early Bronze Age Transition 211 Figure 13.12. FN settlement in North Xerokampos. Figure 13.13. Agia Irini Kastri from northwest. Kalyvomouri, Xerokampos North (Figure 13.12), Xerokampos Kastri, Agia Irini Kastri (Figure 13.13) and Livari Katharades. Immediately behind this coastal zone, FN IV settlements were situated in the gorges which offered convenient communication routes to the inland uplands. In the Zakros Gorge a large FN IV settlement was located on the summit of Kato Kastellas and in the Xerokampias Gorge a small settlement occupied the northern part of the Kastellas ridge, which controlled the northern entrance to the gorge. The easiest point of access to Xerokampias Kastellas from the north was defended by a fortification wall. Close similarities in the topographical characteristics of these and other sites in the East Siteian region indicate that they were founded by groups of people representing the same wave of settlers, arriving at the same time and responding to the same historical circumstances. Karoumes Kastellas and Zakros Kalyvo- 212 Krzysztof Nowicki mouri occupied almost identical hills at the entry to the gorges on the edge of small coastal plains with sheltered bays and sandy beaches; Xerokampos North and Xerokampos Kastri lie at the extreme opposite ends of the Xerokampos plain (northeast and southwest), on similar rocky terraces above the gorges; Agia Irini Kastri and Livari Katharades were located on high and steep ridges, above small bays formed by the mouths of gorges. The changes between the latest FN and early EM I period in the East Siteia peninsula were very dynamic and they are not easy to reconstruct solely on the basis of surveys (Branigan 1998). It seems that the coastal settlements expanded quickly to the interior. The next step in this inland expansion was the Epano Zakros basin where a number of FN IV–EM I sites were identified by Vokotopoulos (2000). A great number of sites here were founded around the edges of the basin on defensible ridges, protruding from the neighbouring plateaus, and had fortification walls defending their accessible sides. The similar phenomenon of location of sites on defensible rocky ridges and defending them by walls can be also observed in higher inland plateaus. The largest density of sites has been recorded in the Ziros plateau with the FN IV–EM I settlements on Rizoviglo and Patela, on the hills above Mesa Apidi and near Agia Triada (Schlager 2001: 180; Nowicki 2002: 25). The second cluster has been identified in the Lamnoni plateau (Branigan 1998: 57–58; Nowicki 2002: 25–26) and another between Palaio Mitato and Magasa (Nowicki 2002: 21). The late FN settlement pattern consisted not only of well defined defensive villages, but also of numerous smaller sites representing probably isolated houses and hamlets. Good examples of such sites have been identified by Branigan in the Lamnoni valley (Branigan 1998), and other similar sites were recorded on the road between Ziros and Katelionas, in the valley north of Livari, in the Mesa Apidi plateau, above Mavros Kampos (west of Epano Zakros) and in the Xerolimni-Magasa plateau. The pottery from these latter sites shows more advanced technology (better firing and more careful surface finishing), which points perhaps to an early EM I date. If such a tentative dating is supported by further research, these small dispersed sites might represent the slightly later third phase of settlement expansion (already in early EM I), following upon Phase 1 (initial FN IV coastal ‘colonization’) and Phase 2 (foundation of hilltop inland settlements). In this model Phase 2, with a large number of extensive sites on the hills dominating the East Siteian plateau, may reflect a short-lasting expansion and concentration of FN population under unstable historical conditions. The density of population between the end of FN and early EM I seems to have exceeded the natural resources of the occupied land and this phenomenon (of large settlements in the East Siteia plateaus) was not repeated again during later prehistoric periods. One may wonder, however, why such limitations were not realized by the FN IV settlers? The later, namely EM I–II and MM patterns, show in the plateaus a larger dispersion of people in small hamlets and individual farmsteads which allowed the land to be exploited in a more The Final Neolithic to Early Bronze Age Transition 213 efficient way, while on the coast the population was concentrated in a few settlements. The latter phenomenon was probably related to the increasing role of sea-trade. The main EM I–II centres developed at or around the FN IV settlements of Palaikastro Kastri, Karoumes Kastellas and Kato Zakros Kalyvomouri. Smaller settlements continued below the FN IV hilltop sites of Xerokampos Kastri and Livari, but most of the FN IV defensible settlements, such as Zakros Gorge Kato Kastellas, Xerokampias Kastellas, Xerokampos North, and perhaps Agia Irini Kastri disappeared at the end of the FN IV or in the early EM I period. The same phenomenon is recorded in other regions of Crete, as for example in the Ierapetra Isthmus and in western Crete. Almost nothing is known about FN settlements west of Goudouras; poor evidence of FN presence was recorded around Kalo Nero, but no proper settlement has been yet found in that area. The pattern becomes better visible again in the vicinity of Koutsounari and Ierapetra with two small sites on Karphi above Koutsounari and Rousso Charakas (west of Panagia Paplinou) located according to the same topographical ‘rules’ as those discussed for the East Siteia region. The main FN–EM I settlement in the Ierapetra district, however, was Vainia Stavromenos (Figure 13.14), which covered an area between 1.0 and 1.2 ha. This site may have played a similar role in shaping the FN–EM I settlement system on the south coast, as Petras Kephala and Mochlos did on the northern coast. Vainia Stavromenos was already inhabited in FN III (contemporary with Monastiraki Katalimata and Anatoli Pandotinou Korifi), but this phase was probably spatially restricted to the rocky knoll where the chapel now stands, whereas by the FN IV and early EM I periods the settlement spread down to extensive terraces to the west and southwest. Similar evidence for continuity in occupation through the FN III and FN IV periods at the same site or on other hills in the close vicinity was noticed in the hilly country between Kendri, Anatoli and Myrtos. Farther to the west, FN IV defensible settlement continues along the south coast; Dermatos Kastrokefala and Lenda Leontari being among the most important sites. In general, however, the pattern between Ierapetra and Kali Limenes, along the coastal zone of the south Lasithi and Asterousia mountains, is not very clear yet; the number of sites is much lower than in the East Siteia peninsula and no large defensible settlements (such as Zakros Gorge Kato Kastellas and Xerokampos Kastri) have been identified. Instead, numerous small open air sites were recorded on the hills in the valleys which run across the Asterousia mountains between the coast and the Mesara plain (Vasilakis 1989– 90). The coast between Agia Galini and Preveli is mostly steep without large open plains. The landscape changes, however, in the bay of Plakias. Archaeological reconnaissance here has identified a large number of FN sites and indicates that FN IV settlements followed almost exactly the topographical ‘rules’ as described for the East Siteia peninsula (Figure 13.3): all rocky promontories and 214 Krzysztof Nowicki imposing hills in the Plakias Bay and the Lefkogia valley were occupied by FN IV sites. The strategic promontories of Paligremnos and Kastri, respectively on the eastern and western side of Plakias Bay, were settled (and probably fortified) at the same time as Zakros Gorge Kato Kastellas, Xerokampos Kastri, Agia Irini Kastri, Livari and Goudouras Kastello. The largest settlements in this area – Gianniou Plati (Figure 13.15) and Sellia Kastello – covered between 0.6 and 0.8 ha), but these were complemented by medium-sized settlements and hamlets Figure 13.14. Vainia Stavromenos from north. Figure 13.15. Gianniou Plati from west. Figure 13.16. Palaiochora Nerovolakoi from east. The Final Neolithic to Early Bronze Age Transition 215 (Ammoudi Skinias). FN IV material, identified by Peatfield and Morris on the peak of Atsipades Korakias (Morris and Batten 2000), may represent a specialized site, located on the edge of the initial ‘colonization’ area, beyond the massifs north of the Plakias-Lefkogia valley and looking down on the Agios Vasilios valley. A similar site may have been located on the summit of Kirimianou, above Mirthios, which dominates the western part of the same valley. The site was identified only on the basis of chipped chert stone, similar to those at Korakias, but no pottery has been recorded yet (but a similar situation was observed at Atsipades Korakias before the site was excavated). The westernmost part of Crete is characterized by a few, rather narrow and not very fertile coastal plains. Yet, also here most of the plains and bays were settled by large communities in the same late phase of FN. The largest and probably most important in this part of Crete was Palaiochora Nerovolakoi (ca. 1.0–1.2 ha) (Figures 13.4 and 13.16). The pottery belongs to the Red Ware group, characteristic of the eastern and south coast, but flattened profile wishbone handles, including those of FN III type, seem to be more common here than in eastern Crete (Figure 13.17). The existence of such a large settlement in this area makes little sense considering that arable land is very restricted and poor. The only explanation for the curious foundation of Nerovolakoi might be that the location was chosen by newcomers landing here without much knowledge of the environmental factors. The ridge is the most characteristic landmark when Palaiochora is viewed from the sea. Nerovolakoi, well defended by cliffs on all sides and dominating the coastal strip, would be an ideal place for the first bridgehead after settlers had arrived on the coast, but before they had the chance to learn about the landscape and its resources. Soon after its foundation Nerovolakoi was abandoned and by early EM I its population had become dispersed along the coast to the west and east and into the interior. Never again would a settlement of a similar size as Nerovolakoi be built on the Palaiochora coast during the prehistoric periods. During EM I–II the narrow plain west of Palaiochora and the sheltered valley of Anydroi appeared Figure 13.17. FN Pottery from Palaiochora Nerovolakoi. 216 Krzysztof Nowicki Figure 13.18. Plan of Sfinari Korakas. to have been more conducive to a stable settlement system based on agriculture and herding. Three FN IV citadels, similar to the type known from the Siteia region, were identified on the extreme western coast of Crete. The first, located on the summit of the Chrisoskalitissa ridge, is now almost completely destroyed, but sherds and chipped stone leave no room for doubt about its character and date. The general topography of the place and the later development of the settlement call to mind the location and earliest history of Palakaistro Kastri. The rocky ridge of Chrisoskalitissa, like Palaikastro Kastri, may have been only a small defensible ‘citadel’ that served a larger group of people who had settled on the coastal terrace immediately to the south, on the other side of a little bay. The lower settlement at Chrisoskalitissa developed during the EM I period, probably thanks to its key position in relation to land and sea routes, and went on to become the largest EM and MM site on the western coast (Hood 1965: 101–2). Further to the north a FN IV site was recorded on the south edge of Sfinari Bay, on the rocky promontory of Korakas (ca. 100 m a. s. l.) (Figure 13.18). The summit of the ridge (ca. 100 m by 20–60 m in size) is covered with pottery of the Palaiochora Nerovolakoi type, basically the same Red Ware group (with ‘cheesepot’ fragments) that appears at many similar settlements along the south coast of Crete. The settlement was defended by natural cliffs and a fortification wall, the The Final Neolithic to Early Bronze Age Transition 217 Figure 13.19. Phalasarna East Acropolis from southwest. remains of which can be seen along the northeast and northern edges of the ridge. A similar wall defended the south side of the Phalasarna ridge, where another defensible FN IV settlement was located (Figure 13.19). The Phalasarna East Acropolis exemplifies the topographical characteristics of a FN IV coastal settlement. The ridge is entirely encircled by high cliffs and the site can only be accessed with difficulty from the south. The FN IV pottery, which is of the same type as that recorded at Palaiochora Nerovolakoi and Sfinari Korakas (and similar to the Zakros-Xerokampos group), is still visible along the southern edge of the ridge, despite the fact that the site was intensively used during the Hellenistic period. The same type of pottery as identified at Sfinari Korakas and Phalasarna, perhaps representing the same group of people, was recently identified on a steep slope immediately below the Leras Cave at Stavros (Akrotiri) on the north coast near Chania. This site is at least 60 by 80 m in size and provides evidence for links between the western coast and the area of the north coast in the vicinity of Nerokourou. First Bridgeheads – Offshore Islands The evidence presented above suggests the existence of two main groups of defensible sites with different pottery traditions, Phaistos and Katalimata, on one hand, and Zakros Gorge Kato Kastellas and Palaiochora Nerovolakoi, on the other. The first group probably represents the old Neolithic Cretan population, 218 Krzysztof Nowicki Figure 13.20. Map of Gaidouronisi showing location of the FN II sites. Figure 13.21. FN II site on Gaidouronisi from west. while the second may correspond to immigrants that arrived from beyond Crete and took over all the bays and coastal plains. Unexpected evidence supporting such a reconstruction was found recently on the small islands of Koufonisi and Gaidouronisi (Chrisi) located at a distance of ca. 6 km and 14 km respectively from the south coast of Crete. On Gaidouronisi a FN IV settlement, much eroded by the sea, was identified on a small promontory that bounds Belegrina Bay on the west (Figures 13.20 and 13.21). Erosion has destroyed most of the original surface, but the exposed bedrock is densely covered with sherds (Figure 13.22). Pottery is concentrated in several clusters which may indicate the position of completely eroded houses. In places the number of sherds exceeds 50 per square metre. Despite the erosion stone constructions are visible at the highest point of the promontory and on the western slope. The site extends over an area of ca. 0.6–7 ha (ca. 100–130 m east-west by 60– 80 m north-south). On both sides of the promontory there are sandy beaches that would have allowed boats to land. The southern coast of Crete between Arvi and Koutsounari (with all main landmarks) can clearly be seen from the site; visible also is the coast of the Asterousia to the west and Koufonisi to the east. The pottery is badly eroded by sea water and the original surface has completely worn The Final Neolithic to Early Bronze Age Transition 219 away. The fabric seems to indicate some sort of relationship with the Red Ware group, but there are even closer similarities with the Dodecanesian group. A small piece of Giali obsidian in one of the sherds is the best indicator of direct links between the site on Gaidouronisi and the area of the central Dodecanese, namely Nisiros, Giali and perhaps southern Kos. The presence of large numbers of ‘cheesepots’ further supports the Dodecanesian connection. The second site on Gaidouronisi is located on the eastern coast, on the northern side of the Bay of Kataprosopo. This was a hamlet or a very small settlement, consisting probably of a few houses, which unlike the Belegrina site survived into the early EM I period. FN IV evidence on Koufonisi is of a different character (Figure 13.2). Several closely-spaced clusters of pottery occur along the northern coast Figure 13.22. Pottery from the FN II (IV) site on and the site is either of an enormous Gaidouronisi. size (ca. 500 m by 50–80 m) or alternatively there are two sites with several small clusters of pottery between them. The first large concentration of pottery (mixed with MM and LM sherds) is located on high ground on the northern promontory and spreads out to the southeast for a distance of about 200 m. The second main concentration of sherds is situated about 100 m northwest of the chapel and covers an area of at least 0.5 ha. In the southern part of the site (or cluster of sites) the pottery is almost exclusively of the Red Ware type, whereas in the northern part a substantial amount of the sherds belong to the harder Dodecanesian group. The occupation of this part of Koufonisi can be more easily explained if we reconstruct the original shape of the island as it was in the second half of the fourth millennium BC. Most probably, the two little islets that at present extend a few hundred metres to the north were at that time a crescent shaped peninsula, forming a large bay sheltered from northwestern winds. The large size of the site and the quantity of pottery indicate a very substantial number of people living or temporarily residing here in the FN IV period, at the same time as defensible ridges were being suddenly settled on the opposite side of the narrow strip of sea, separating the island from the Cretan coast. Because 220 Krzysztof Nowicki Koufonisi is an island too small to support such a large community for a longer period we have to assume that the FN IV settlement was only a temporary phenomenon during a time of dramatic change in the East Aegean. On both islands (Gaidouronisi and Koufonisi) FN IV sites were located on low coastal terraces without any natural or artificial defences. This prompts the question of why, in the same period, people should have chosen such different locations for their settlements, namely undefended extensive sites on Gaidouronisi and Koufonisi and defensive settlements on the Cretan coast? The most plausible answer must take into consideration the likely geo-political situation in Crete at the time. Crete was a large island with the substantial native population concentrated in a few large centres such as Knossos and Phaistos and dispersed in smaller groups through all geographical zones. It is hard to see how the arrival of new people, arriving probably in substantial numbers, could have occurred without resistance and tension regarding the exploitation and ownership of land. Both sides of any potential conflict had good reasons to feel insecure. The situation on little islets, such as Koufonisi and Gaidouronisi, situated a few miles off the Cretan shore, would have been different. They could not be defended against an enemy coming from the sea, even if they had been temporarily or seasonally occupied before FN IV, a hypothesis that is not yet proven. On the other hand these islets were perfect bridgeheads which could have been used by newcomers as the first temporary sites before a further expansion onto the Cretan coast. A similar role may have been played by other ‘habitable’ islands around Crete, such as Dionisades, Pseira, Dia and Gavdos. The Dodecanesian Connection? These latest FN coastal settlements in Crete must be seen as representing a phenomenon contemporary with the foundation of many latest FN settlements throughout the Cyclades, of which Agia Irini Period I and Paoura on Keos are among the best known (Figure 13.23). Many of these new Cycladic settlements had topographical characteristics similar to the sites in Crete, as exemplified by a small settlement on a rocky ridge at Kampos Komikias on the western coast of Naxos (Figure 13.24) and Agios Ioannis Kastri on Astypalaia. It is less clear, however, if the hilltop sites such as Minoa on Amorgos and Chora on Ios belonged to the same group of possible immigrants or represented, like Katalimata in Crete, the earlier population of those islands. When analyzing the phenomenon of latest FN settlement in Crete, special attention has to be paid to the islands of the Dodecanese, where a great number of Late Chalcolithic 3 and 4 sites have been identified by Simpson and Lazenby (Simpson and Lazenby 1973) and more recently by Sampson (Sampson 1987) (Figure 13.25). However, the problem of Late Chalcolithic 3–4 (late FN) defensible sites in the Dodecanese requires more field research. The site of Kastri on Astypalea The Final Neolithic to Early Bronze Age Transition 221 Figure 13.23. (above) Map of the South Aegean. 1. Ag. Irini on Kea, 2. Paoura on Kea, 3. Phylakopi on Milos, 4. Chora Kastri on Ios, 5. Kampos Komikianos on Naxos, 6. Minoa on Amorgos, 7. Ag. Ioannis Kastri on Astypalaia. Figure 13.24. (left) Kampos Komikias on Naxos. has topographical characteristics very similar to Cretan coastal sites. The same can be said about three sites on Symi: Pedi Gria (Figure 13.26), Chora Kastro and the site south of and above Panormitis, which is different from that on the northern promontory of the same bay (Sampson 1987: fig. 60). The easiest points of access to Pedi Gria and Panormitis were defended by walls which call to mind the defensive constructions identified at FN IV settlements on Crete. On Tilos a defensible site at Livadia Faneromeni (Figure 13.27), defended by cliffs and 222 Krzysztof Nowicki Figure 13.25. The Dodecanese with the FN sites mentioned in the text. KASOS: 1. The site south of the Airport, 2. Chelatros; KARPATHOS: 1. Finiki, 2. Moulas, 3. Leftoporos, 4. Ag. Theodoros, 5. Gigla, 6. Vouno; TILOS: 1. Livadia Faneromeni, 2. Chora Kastri; SYMI: 1. Panormitis, 2. Pedi Gria, 3. Chora; KOS: 1. The site near Ag. Stefanos; KALYMNOS: 1. Chrysocheroi, 2. Vathy; LEROS: 1. Partheni. probably also partly by a wall, dominated the southern part of the coastal plain. Some sites were located at a slightly greater distance from the sea, but also on hills or rocks dominating their immediate vicinity (e.g., Chrysocheroi on Kalymnos), others were founded on small islands that offered scarcely any agricultural hinterland, such as Kastri on Alimnia and a series of sites on Giali (Sampson 1987). Settlements were also located on more gently-sloping hills (e.g., Astypalaia Vai), low promontories and coastal plains (e.g., Partheni on Leros). This period is also well represented on the two Dodecanesian islands nearest to Crete: Karpathos and Kasos (Melas 1985; pers. comm.). The site on the rocky promontory of Moulas on Kasos (Figure 13.28) (Nowicki 2004: 97) must have been founded at that time, but it is unusual in that occupation continued into the EB I and II periods. Its location calls to mind the topography of Palaikastro Kastri. Some of the FN defensible sites on Karpathos were defended by walls, as is indicated by poor remains of such constructions at Afiartis Gigla and Afiartis Vouno (Melas 1985; pers. comm.). Sites along the western coast of Kasos appear to have a slightly different character. Here the wide coastal strip, stretching for at least 3 km (between the airport and the southern edge of the coastal plain), is densely scattered with Red Ware of a type identical to that characteristic of eastern and southern Crete. The sites must represent extensive but rather short-lasting occupation (individual The Final Neolithic to Early Bronze Age Transition 223 Figure 13.26. Pedi Gria on Symi from northwest. Figure 13.27. Livadia Faneromeni on Tilos from southwest. Figure 13.28. Moulas on Karpathos from northwest. houses, hamlets, small settlements) located close to the sea without any concern about security. The topographical situation is identical with that recorded on the northern coast of Koufonisi (see above). The most plausible interpretation of this choice of site location is, like Koufonisi, that the entire area was temporarily occupied by a large number of people that were on their way somewhere else, most probably to Crete. The site of Trapeza on the Chelatros Bay in the southern part of Kasos (Melas 1985: 46) must have been founded in the Late Chalcolithic (FN) period, but has occupation continuing into EB I. A few sherds from this site exhibit similarities with the material from Vainia Stavromenos. An interesting difference between Crete and the Dodecanese, however, is the fact that, whereas in Crete (and the Cyclades) the appearance of a large number 224 Krzysztof Nowicki Figure 13.29. Pottery from the Dodecanese (D) and Crete (C). D 1: Partheni on Leros (after Sampson 1987, Figure 128); D 2: the site near Ag. Stefanos on Kos; D 3: Panormitis on Symi; D 4 and D 5: Pedi Gria on Symi; D 6 and D 7: the site near Ag. Stefanos on Kos; D 8: Kalythies on Rhodes (after Sampson 1987: fig. 25); C 1: Ag. Paraskevi (Zakros); C 2: Xerokampos Kastri; C 3: Livari Katharades; C 4: Dermatos Kastrokefala; C 5: Xerokampos Kastri; C 6: Ag. Irini. Figure 13.30. Pottery from the Dodecanese (D) and Crete (C). D 1 and D 2: Kastro on Alimnia (after Sampson 1987: fig. 105); D 3: Panormitis on Symi; D 4 and D 5: Pedi Gria on Symi; D 6: Partheni on Leros (after Sampson 1987: fig. 120); D 7: Koumelo on Rhodes (after Sampson 1987: fig. 88); D 8 and D 9: Partheni on Leros (after Sampson 1987: fig.120); D 10 Koumelo on Rhodes (after Sampson 1987: fig. 87). C 1 and C 2: Dermatos Kastrokefala; C 3: Ag. Irini; C 4: Gaidouronisi; C 5: Palaiochora Nerovolakoi; C 6:Ag. Irini; C 7: Livari; C 8 and C 9: Gaidouronisi; C 10: Palaiochora Nerovolakoi. of new FN IV sites marked the beginning of substantial population growth that continued through EB I–II, the Late Chalcolithic 3–4 population peak in the Dodecanese is followed by a sudden drop in the number of settlements, probably indicating considerable depopulation. The situation on Karpathos may have been somewhat different, but here EB I–II sites still have defensible characteristics. This may by another important argument in favour of migration from western Anatolia and the Dodecanese towards Crete. The similarities between the pottery of the Cretan and Dodecanesian sites must be seen as very meaningful in this context (Figures 13.29 and 13.30). The Final Neolithic to Early Bronze Age Transition 225 Summary I have argued elsewhere that the widespread shift of settlement to higher and more defensible hills throughout Crete and the foundation of new sites along the Cretan coast are linked processes (Nowicki 2002). Settlements of the Phaistos and Katalimata types may have anticipated (in FN III) the coastal sites of Xerokampos Kastri and Palaiochora Nerovolakoi type (in FN IV). FN III and FN IV pottery might be seen as partly overlapping and the differences between FN III and FN IV pottery may reflect not only a chronological sequence, but also the different origins of the pottery producers. Sites of the Phaistos and Katalimata type were probably founded by Cretans looking for security at a time when interaction with the outside world, rather than internal problems, forced them to move to better-defended locations. The newcomers that settled in the coastal areas, in particular along the eastern and southern coasts, did not feel safe either and looked for defensible promontories or rocky ‘acropoleis’ above coastal plains and bays. The two types of site represent two stages of a historical process of conflict between the local Cretan Neolithic population and their Chalcolithic neighbours coming from the east. Unstable settlement and signs of conflicts, as recorded on the Dodecanesian islands, are indications that the problems were coming probably from the Anatolian coast. This new population element must have consisted of several different groups, a situation that can be reconstructed from different types of pottery. Southern Crete was dominated by the Red Ware group, the development of which shows direct links with later painted EM I pottery. Similar pottery was recorded in the south Dodecanese, in particular Karpathos and Kasos, but also Rhodes and Symi. However, Red Ware in the south Dodecanese is commonly found together with pottery showing a north Dodecanesian influence (as represented for example by fabrics with glassy quartz). The northern coast of Crete yields more evidence for ceramic links with the north and central Dodecanese and Cyclades. Early Minoan cultural differentiation in Crete may have been due to the complexity of the population structure that emerged during the crucial period of the latest FN (Betancourt 1999; Hayden 2003). Analysis of settlement patterns and how they change together with evidence for pottery differentiation allow four main elements of population to be proposed: (1) a native Neolithic Cretans, (2) a Red Ware group which seems to be particularly strong in eastern and southern Crete, (3) a northern and central Dodecanesian group, well represented now at Kephala Petras and several other coastal regions in northeast Crete. At Petras, however, this latter group may have been eventually replaced by a group close to or identical with the later phase of the Red Ware group (see Papadatos this volume), and (4) a group related to the earliest phase of the Pelos culture in the Cyclades, identifiable in some pockets along the northern coast with occasional links traceable as far as Chrisoskalitissa in the far west. All the groups share some similar characteristics in their material culture and had probably been in contact long before the process of dramatic settlement change 226 Krzysztof Nowicki started in the Aegean. The most enigmatic seems to be the Red Ware group and particularly its interaction with the very strong native element in the Mesara. The number and size of sites suggest that their inhabitants could not just come from the Dodecanese, but had to derive also from the Anatolian mainland. The archaeological evidence presented above, supports Peter Warren’s earlier hypothesis of ‘some movement of people into Crete from the west Anatolian region to join the Late Neolithic population at the end of the fourth millennium and beginning of the third’ (Warren 1973: 43), and Sinclair Hood’s idea that ‘Early Minoan IA reflected the coming of immigrants’ (Hood 1990a; 1990b: 368). In a chronological sense the Cretan Neolithic could be said to end with the FN III period, while FN IV could be seen as the beginning of a new era in a social and economic development that more or less smoothly continued through the EM I period. Although analysis of settlement patterns does not provide a complete picture, it is certainly more relevant to the problem than the hitherto more usual analyses of tombs and their contents. The locations of many of these sites are strongly orientated towards the sea and defensibility was clearly a serious concern. These two facts tell us something about the character of the inhabitants of these settlements. After colonization in FN IV the settlement pattern underwent substantial changes during the early EM I period. This led to a better exploitation of natural resources and better control of the hinterland. Tensions between different groups in Crete and beyond the island continued at least for a time into EM I, but the development of a new form of social organization and a new political system was already under way. This created the conditions for more stable settlement, at least in some areas and particularly along the northern coast. Acknowledgements I would like to thank the organizers of the conference for the invitation to speak and other members of the Department, in particular Umberto Albarella, for their excellent hospitality in Sheffield. I am much obliged to the directors of the Chalasmenos Project, Metaxia Tsipopoulou and the late Wiliam Coulson, for their invitation to participate, which allowed me to excavate at Katalimata. Yiannis Papadatos, Simona Todaro, Peter Tomkins and Manolis Melas kindly shared with me their knowledge of the material from Petras, Phaistos, Knossos and Karpathos; Saro Wallace and Peter Tomkins helped with editing the English. Bibliography Atkinson, T., R. Bosanquet, C. Edgar, A. J. Evans, D. Hogarth, D. Mackenzie, C. Smith and F. Welch 1904 Excavations at Melos. London: Macmillan. The Final Neolithic to Early Bronze Age Transition 227 Betancourt, P. 1999 What is Minoan? FN/EM I in the Gulf of Mirabello region. In P. Betancourt, V. Karageorghis, R. Laffineur and W.-D. Niemeier (eds.), MELETEMATA. Studies in Aegean Archaeology Presented to Malcolm H. Wiener as he Enters his 65th Year (Aegaeum 20): 33–41 Liège: University of Liège. Branigan, K. 1998 Prehistoric and early historic settlement in the Ziros region, eastern Crete. BSA 93: 23–90. Broodbank, C. 2000 An Island Archaeology of the Early Cyclades. Cambridge: Cambridge University Press. Caskey, J. 1972 Investigations in Keos. Part II: A conspectus of the pottery. Hesperia 41: 357–401. Evans, J. D. 1964 Excavations in the Neolithic settlement of Knossos 1957–60: Part I. BSA 59: 132–240. 1994 The early millennia: continuity and change in a farming settlement. In D. Evely, H. Hughes-Brock and N. Momigliano (eds.), Knossos, a Labyrinth of History. Papers Presented in Honour of Sinclair Hood: 1–20. London: British School at Athens. Greco, E., Th. Kalpaxis, N. Papadakis, A. Schnapp and D. Viviers 2002 Travaux menés en collaboration avec l’École Française en 2001. Itanos (Crète orientale). BCH 126: 577–82. Hayden, B. 2003 Final Neolithic–Early Minoan I/IIA settlement in the Vrokastro area, Eastern Crete. AJA 107: 363–412. Hood, M. S. F. 1965 Minoan sites in the far west of Crete. BSA 60: 99–114. 1990a Settlers in Crete c. 3000 B.C. Cretan Studies 2: 151–58. 1990b Autochthons or Settlers ? Evidence for immigration at the beginning of the Early Bronze Age in Crete. Pepragmena tou ST’ Diethnous Kritologikou Synedriou A1: 367–75. Chania: Philologikos Syllogos o Chrysostomos. Hood, M. S. F., P. M. Warren and G. Cadogan 1964 Travels in Crete. BSA 59: 50–99. Hood, M. S. F. and P. M. Warren 1966 Ancient sites in the province of Ayios Vasilios, Crete. BSA 61: 163–91. Melas, E. 1985 The Islands of Karpathos, Saros and Kasos in the Neolithic and Bronze Age (SIMA 68). Göteborg: Paul Äström. Morris, C. and V. Batten 2000 Final Neolithic pottery from the Atsipades peak sanctuary. Pepragmena H’ Diethnous Kritilogikou Synedriou: 373–82. Herakleion: Etaireia Kritikon Istorikon Meleton. Nowicki, K. 1999 The Final Neolithic refugees or the Early Bronze Age newcomers? The problem of defensible sites in Crete in the late fourth millennium BC. In P. Betancourt, V. Karageorghis, R. Laffineur and W.-D. Niemeier (eds.), MELETEMATA. Studies in Aegean Archaeology Presented to Malcolm H. Wiener as he Enters his 65th Year (Aegaeum 20): 575– 81. Liège: University of Liège. 2002 The end of the Neolithic in Crete. Aegean Archaeology 6: 7–72. 2004 Report on investigations in Greece. XI. Studies in 1995–2003. Archeologia Warsaw 55: 75–100. 2008 Monastiraki Katalimata: Excavation of a Cretan Refuge Site, 1993–2000. Philadelphia: INSTAP Academic Press. 228 Krzysztof Nowicki Renfrew, C. 1972 The Emergence of Civilisation. London: Methuen. 1996 Who were the Minoans? Towards the population history of Crete. Cretan Studies 5: 10– 27. Sampson, A. 1984 The Neolithic of the Dodecanese and the Aegean Neolithic culture. BSA 79: 239–49. 1987 H Neolithiki Periodos sta Dodekanisa. Athens: Tameio Archaiologikon Poron kai Apallotrioseon. 1988 H Neolithiki Katoikisi sto Yiali tis Nisyrou. Athens: Evoiki Archaiophilos Etaireia. Schlager, N. 1997 Minoische bis rezente Ruinen im fermen Osten Kretas. Jahresheften des Österreichischen Archäologischen Institutes 66: 2–83. 2001 Plaistozäne, neolitische, bronzezeitliche und rezente Befunde und Ruinen im fernen osten Kretas. Jahresheften des Österreichischen Archäologischen Institutes in Wien 70: 157– 220. Simpson, R. H. and J. F. Lazenby 1973 Notes from the Dodecanese III. BSA 68: 127–70. Tomkins, P. 2000 The Neolithic period. In D. Huxley (ed.), Cretan Quests. British Explorers, Excavators and Historians: 76–85. London: British School at Athens. 2007 Neolithic: Strata IX–VIII, VII–VIB, VIA–V, IV, IIIB, IIIA, IIB, IIA and IC Groups. In N. Momigliano (ed.), Knossos Pottery Handbook: Neolithic and Bronze Age (Minoan): 9–48. London: British School at Athens. Vagnetti, L. 1973 Tracce di due insediamenti neolitici nel territorio dell’antica Gortina. Anticita cretesi. Studi in Onore di Doro Levi:1–9. Catania: Università di Catania, Istituto di Archeologia. 1972–3 L’insediamento neolitico di Festòs. ASA 50–51: 7–138. 1996 The Final Neolithic: Crete enters the wider world. Cretan Studies 5: 29–40. Vagnetti, L. and P. Belli 1978 Characters and problems of the Final Neolithic in Crete. SMEA 19: 125–63. Vasilakis, A. 1989–90 Proistorikes theseis sti Moni Odhigitrias, Kaloi Limenes. Kritiki Estia 3: 11–80. Vokotopoulos, L. 2000 Ochires protominoikes theseis stin periochi Zakrou. Pepragmena H’ Diethnous Kritilogikou Synedriou: 129–46. Herakleion: Etaireia Kritikon Istorikon Meleton. Warren, P. M. 1973 Crete 3000–1400 B.C.: immigration and the archaeological evidence. In R. A. Crossland and A. Birchall (eds.), Bronze Age Migrations in the Aegean. Archaeological and Linguistic Problems in Greek Prehistory (Proceedings of the First International Colloquium on Aegean Prehistory, Sheffield): 41–50. London: Duckworth.