The Pylos Regional Archaeological Project: Preliminary Report on the 1992 Season


Principal Investigator: Jack L. Davis

Susan Alcock, University of Michigan
John Bennet, University of Wisconsin, Madison
Yannos Lolos, Guides' School, Athens
Cynthia Shelmerdine, University of Texas at Austin
Eberhard Zangger, University of Heidelberg

With an appendix by Sergei Yazvenko

May 15, 1993


The first season of fieldwork sponsored by the Pylos Regional Archaeological Project was conducted between June 22 and August 16, 1992.[1] In total, over forty archaeologists and students of archaeology participated in PRAP.[2]

There have been no significant changes in workplans as described in our original proposal to NEH. All major components of the project are now well underway, and an oral report on the results of the 1992 season was delivered at the 1992 annual meeting of the Archaeological Institute of America (abstract American Journal of Archaeology 97 [1993] 330-31). After one full season in the field we now have a fairly accurate impression of the types of material we will be recovering in the future and the expertise needed to study it in the museum. We have, therefore, added three new members to our museum staff as Museum Consultants: Ann Harrison of McMaster University, Sharon Gerstel of Dunbarton Oaks, and Kathleen Slane of the University of Missouri, Columbia. Harrison will take responsibility for study and publication of Archaic through Hellenistic finds; Slane, of Roman; and Gerstel, of Byzantine and medieval.

PRAP continues to avail itself of current electronic technology whenever possible. Development of this aspect of our project has been considerably facilitated through grants from the University of Illinois and from the University of Michigan. In 1993 we will be able for the first time to plot spatial data daily in the field with a computerized mapping routine (see Appendix 1).

Raising of matching money has thus far progressed without difficulty. Thus far PRAP has received as contributions toward matching NEH funds:

1992 National Geographic Society $20,374

1993 National Geographic Society $23,865

1992 Institute for Aegean Prehistory $5,628

1993 Institute for Aegean Prehistory $6515

Total $56,382

In response to our applications both this year and last, the Institute for Aegean Prehistory expressed its willingness to make available to PRAP additional funds, had our applications to NGS been unsuccessful. We have every reason to believe that next year we will enjoy continuing support from INSTAP at this same level.


In 1992, substantial progress was made toward addressing several problems central to the research design of the project. With the completion of surface survey of the entirety of the Englianos ridge, it is now possible to estimate for the first time the total extent of the prehistoric community around the Bronze Age Palace of Nestor; the so-called "lower town" investigated by Blegen's team should now be seen as part of a much more extensive settlement. Palaeo botanical researches underway by members of our team promise to provide environmental data of high quality. Geoarchaeological investigations are yielding new clues to the location of the ancient port associated with the Palace of Nestor. Finally several intrinsically important new finds included a Mycenaean sealstone and two new Linear B fragments--a tablet and a nodule--studied by members of PRAP this past summer.

Goals of the Project

The general goal of the Pylos Regional Archaeological Project is the systematic and intensive documentation of all remains of past activity in an area of southwestern Messenia (Greece) centered on the Palace of Nestor, excavated by C.W. Blegen in 1939-69. Our primary objective is to reconstruct and interpret changes in settlement patterns and land use in the region throughout the Bronze Age and historical period, through the integration of archaeological and textual data. An important aspect of this work is the examination of the archaeologically-visible consequences of the emergence of the Palace of Nestor, of its operation as a center, and of its decline. A second important component of our work is an investigation of the material consequences of Sparta's subjugation of Messenia and the subsequent development of free city-states by the local population, when liberated in the 4th c. B.C.

By the 1960s, archaeological research in Messenia (particularly that of McDonald, Hope Simpson and Marinatos) had produced much information essential to the study of past settlement patterns and land use. More recently, George Korres has built on that research, reinvestigating a number of known excavations and working at new locations. But, although a large number of sites are known in Messenia, very few settlement sites have been excavated. Sites discovered were also investigated in years before the deployment of techniques of systematic intensive survey had become regular in Greece and before geophysical methods of subsurface investigation had become commonplace in archaeological projects. No systematic surface collections of artifacts were ever made at the palace itself. Our research will fill very significant gaps in our understanding of the archaeological record and is, we hope, providing a framework in which previous research can for the first time confidently be viewed as part of a coherent regional whole.[3]

Fieldwork in 1992

On the basis of preliminary fieldwork undertaken in 1991, we selected a total study area of ca. 250 km2 and received permission from the central archaeological council of the Greek Ministry of Culture (KAS) to study it in its entirety. While a smaller area would alone have provided us with a sample of sites situated at the lower end of settlement hierarchies, the selection of this larger study area allows us to include significant regional centers of all periods of the past (the Palace of Nestor for the Late Bronze Age, Koryphasion for the C-HL, Paleonavarino and Metamorphosis for the Medieval period), in addition to habitations in their hinterlands.

Figure 1: Areas and Sites Investigated by PRAP in 1992. (RJR)

Archaeological fieldwork in our first full field season took two forms. First, intensive survey of ca. 12 km2 provided detailed information about the surface distribution of artifacts in two principal areas: the entire Englianos ridge, including the immediate area of the Palace of Nestor; and a sizable area in the probable boundary region between the Hither and Further Provinces of the Bronze Age Pylian state. Second, an inventory survey team conducted intensive surveys of limited extent over and around already known sites, primarily in the coastal region between Tragana, Koryfasion and Petrohori, and also in the vicinity of Koukkounara.

The Palace of Nestor Settlement

As a result of intensive work on the Englianos ridge by PRAP teams, we have gathered significant new information about the extent of settlement around the Palace. The Palace sits on one of several long ridges that run toward the sea from the foothills of the lofty range of the Aigaleon mountains, widely accepted as marking the boundary defined in the Linear B texts between the two provinces of the kingdom of Nestor. To the north of the Bay of Navarino, these ridges converge on the broad lowlands that support the modern towns of Romanou, Petrochori, and Yialova. The Englianos ridge on which the Palace was situated, carries on its back the major modern route of communication through Triphylia, an asphalt two-lane highway that today links the towns around the Bay of Navarino with settlements near Kyparissia and in the Soulima valley, areas where, among others, the Mycenaean sites of Mouriatada, Peristeria, and Dorion-Malthi are located. At its westernmost extent the ridge leaves the valley floor just north of the modern town of Koryfasion, formerly Osmanaga. From here it climbs steadily, bordered both on the north and south by steep ravines, finally converging with several minor ridges and the uplands adjacent to Hora on the south. Almost half-way between the coastal plain and modern Hora, the ridge rises more steeply, this ascent defining the border between the Kato and Ano Englianos areas. About a kilometer further to the northeast, the ridge narrows appreciably and is tied to the uplands of Hora only by a narrow neck.

The Palace of Nestor sat on the highest of a series of four knoll-tops along the upper ridge and enjoyed a commanding view over its approaches in all directions, as well as itself dominating much of the territory both inland and toward the coast. It has been clear since Blegen's excavations that the Palace complex itself and its immediately related buildings, themselves approximately 1.5 ha. in size, formed only a fraction of the total settlement on the upper Englianos ridge. In the report of the Minnesota Messenia Expedition, McDonald and Hope Simpson estimated the total extent of this settlement as "at least 325 NE-SW x 200 m." for a total minimum area of 6.5 ha. By test-trenching, Blegen's team explored parts of this "lower town" (as it came to be known since it lay on terraces lower than the citadel of the Palace itself). In 1992, an important item on our agenda was to survey the entirety of the area that might have been covered by this town and, on the basis of surface collections, to provide a more accurate estimate of its extent during various periods.

For much of the season, two intensive survey teams operated on the ridge, separated by the modern asphalt road--a convenient border. Some 125 individual tracts (units of the landscape about one to two ha. in size and homogeneous in vegetation) were studied by these two teams of 6-7 individuals. Each tract was walked in parallel linear transects, spaced 15 meters apart. Densities of artifacts observed were recorded for each 100 m. walked and all potentially diagnostic pottery observed while walking transects was collected and brought to the museum. Because a substantial quantity of finds was collected and because much of the fieldwork was only finished in the closing days of the season, it is not yet possible to present definitive conclusions. Nevertheless, examination of the distributions of densities of artifacts suggests already that it will be necessary to revise some of our ideas about this important prehistoric community.

Figure 2: Pottery Densities on the Englianos Ridge. (JLD/TJM)

For one thing, surface artifacts are not only dense on the terraces immediately west of the Palace, the location of most of Blegen's explorations in the so-called "lower town". They are as dense, or nearly so, to the south, southeast, and also in the direction of Hora, much farther to the northeast than the well-known Tholos Tomb IV, excavated by Blegen's team; this tholos must have lain within the area of the town here at Pylos, as, of course, did many of the tholoi at Mycenae. On the basis of these first approximations, it seems to us likely that that the settlement was once strung out along the ridge for as much as 1000 m. NE-SW while the maximum dimension of the ridge across its back NW-SE varies from 200-300 m. or even greater. The total area of the site may, therefore, have been some 20-30 ha. This estimate is of course only approximate and one of our goals in the next two years is to refine these results. In 1993, for example, we plan to establish a grid over parts of this larger town site. More detailed surface collections will be made according to the grid; geomorphological coring will test the depths of deposit and evaluate the likelihood that they are in situ; and geophysical prospection in selected areas will, we hope, supply further information about the density of occupation on the site.[4] Meanwhile, continuing analysis in the museum of finds from the 1992 collections will permit us to speak with more precision about variations in the size of the site through time.

Already it is possible, however, to offer a few preliminary observations in this regard. In 1992 for purposes of rough estimates we divided the town into four quadrants around the fenced archaeological zone. Preliminary analysis of pottery led to the following results. While all four yielded MH-LH III material, the most closely datable LH IIIA-B came from the northeast, an area which also produced EH, suggesting early occupation in this region. Geometric and possible Geometric material was present in the southwest and southeast quadrants, and A-C in the northeast, further strengthening the idea that occupation continued on Ano Englianos in the Dark Ages. The most recent material--Roman through Byzantine--is at present confined to the southwest sector. Other artifacts discovered include chert tools and flakes, a projectile point, a Late Helladic III animal figurine, and a seated human figurine.

Work in 1992 also produced new information about the cemeteries associated with this settlement. Blegen had already excavated chamber tombs (E-1 to E-10) at the location called by him the Tsakalis hill, which, we can now see, lay at or just outside the western edge of the prehistoric settlement, as did Blegen's Tomb K-1. Not far to the south, at our site B5, however, we found evidence of additional chamber tombs, perhaps part of a large cemetery now almost destroyed by terracing, on the southeast slope of the ridge overlooking the deep ravine that divides the Englianos ridge from that of Vayies, on which the village of Myrsinochori sits to the southeast. Also not far outside the limits of the settlement, but to the northeast, at our site B4, we recognized the remains of more destroyed chamber tombs with what may well be the remains of a looted prehistoric grave cut into the top of a tumulus-like knoll not far away (POSI B3) and in a position that affords a commanding view down on the Palace of Nestor from the northeast. Once our work in the vicinity is concluded, it is realistic to expect an unparalleled picture of a Bronze Age palace in its overall settlement context.

The Boundary Between the Hither and Further Provinces

Survey in the regions that are likely to have belonged to the Hither-Further province boundary region focussed on the narrow valley south of the modern village of Metaxada, east of modern Hora on the other side of Mt. Aigaleon, reached by a road that departs Hora through the modern suburb of Volimidia, where many Mycenaean tombs and a Bronze Age settlement have been excavated by Marinatos and Korres. Here our discoveries were quite unlike patterns of prehistoric settlement that have been reconstructed by surveys in the eastern Peloponnese. Much to our surprise, we did not recognize evidence for small Neolithic, Early Bronze or Late Bronze Age sites of the sort so common in the Argolid. Indeed, all prehistoric habitation in the valley appears to have been located at the single site of Kalopsana, previously known from chance finds of bronze double axes and a sword, now on display in the Hora Museum. McDonald and Hope Simpson's sherding suggested LH III habitation with the possibility of MH and early Mycenaean material. We have not yet been able to locate and restudy the material from their collection, but in our own extensive collections there was no trace of LH III occupation present. Instead, other than scant traces of Archaic-Hellenistic material, all of our finds appear to date to the Middle and early Late Bronze Age. There is, therefore, as yet no evidence that this valley was inhabited at all after the emergence of the Palace of Nestor. It will be interesting to see if a similar phenomenon is observable in other parts of the survey area, and if so, the formation of such no-man's lands may have been directly related to the the formation of central places in the palatial period that attracted populations from more marginal boundary areas (cf. the Agiofarango valley in Crete, almost deserted after EM, when Phaistos emerges as palace).

The Koryfasio-Romanou-Tragana Triangle

Our second principal focus of research in 1992 lay in and around the coastal plains southwest of the Englianos ridge and included areas around Koryfasion, Romanou, and Tragana. Here an "Inventory" team conducted intensive surveys of limited extent around already known sites. Investigations revealed the existence of an enormous site around the modern village of Romanou, over 1 km2 in extent (I.5 and I.6). Finds there were almost entirely of Greco-Roman date but there are also traces of the early and later Mycenaean periods. Of wider significance is the fact that our work this summer in general suggests that this Tragana-Romanou-Koryfasio triangle was an important focus of prehistoric activity at times contemporary with the operation of the Palace of Nestor. Koryfasio Beylerbey (Site I.1), a large prehistoric and historical site just to the southeast of the Osmanaga tholos tomb was a target of special investigation: plentiful Mycenaean surface finds were collected from 300 20x20 m. grid squares--kindly established for us by

Figure 3: Late Mycenaean Sealstone from Koryfasion: Beylerbey. (LBP)

Professor Frederick Cooper and his colleagues over an area of 12 ha.[5] Finds included a Late Mycenaean sealstone (above) with a representation of a bull assigned by Professor John Younger to his "Mainland Popular Group" of LH IIIA on the basis of examination of drawings and photographs. Clearly Beylerbey was a settlement of some consequence.

At the same time, intensive survey defined the limits of a contemporary 2-ha. settlement at Koryfasio Portes, nearly adjacent to the Osmanaga Tholos on the northwest. Furthermore, geomorphological investigations in the area immediately southwest of the tholos tombs at Tragana (Site I.6) seem to confirm a theory developed by Kraft, Rapp, and Aschenbrenner that the course of the river bordering the Englianos ridge on the northwest had been artificially diverted in antiquity. Zangger's analysis of air-photos and microtopography suggests that the purpose of this diversion may have been to lead the stream into a now in-filled port basin lying immediately north of Romanou, that once was connected to the sea. If this hypothesis is supported by work next year, it is possible that we will have discovered the port and port communities associated with the Palace of Nestor.

Linear Scripts

Finally, it is appropriate to make brief mention here of two new Linear B inscriptions recently found at the Palace of Nestor which Cynthia Shelmerdine and John Bennet have studied in association with our general work and will be publishing shortly. We are grateful once again to the Eforeia for permission to publish these finds, and to Frederick Cooper for offering us the tablet fragment to publish.

The first, a fragment of a small page-shaped tablet provisionally assigned the number Xn 1479, was discovered in 1991 in the course of cleaning carried out by the Minnesota Archaeological Research in the Western Peloponnese (MARWP) team as part of their project to produce a final state plan of Blegen's excavations at Bronze Age Pylos. It came from fill left by Blegen over a wall in the area of the Northeast Workshop. It is broken at top, bottom and right, but complete at left. The lines of text are separated by rulings. The two best-preserved lines appear to contain personal names: qo-wi-ro (probably complete) on the upper and o-ma [or o-pe(surely incomplete)] on the lower. Neither name is attested elsewhere in the corpus. There are traces of signs at the top, and a third ruling at the bottom, suggesting the presence of further text. The shape of some signs is distinctive, and may connect the scribal hand with one already attested in the Pylos archive. Also interesting is the verso of the piece, which preserves a deep punch made from right to left diagonally into the tablet, presumably by a stylus. Since there is some controversy among Linear B epigraphists about the normal form of a stylus, investigation of this mark may shed some important new light on this matter.

The second piece, a nodule provisionally assigned the number Wr 1480, was discovered by Mr Yiannis Rokanas, a guard at the palace, in 1988 in a flower-bed just inside the entrance to the site. Nodules are of particular interest to those studying the Linear B administration, since they are the only class of Linear B documents which--by definition--represent transactions between two parties, probably the palace and an individual or producer. They generally involve the impression of a seal, indicating identity, onto a lump of clay which is pressed around a cord. A palace scribe adds varying amounts of textual information concerning the transaction. Our nodule has the usual three faces, each bearing one or more Linear B signs, the whole originally formed around a cord. It is unusual, however, in that the single sign wa, elsewhere understood as an abbreviation for the adjective wanakteron, "royal," appears on side alpha apparently in lieu of a seal impression. Also absent is the ideogram which frequently appears over the seal impression. The subject-matter of the transaction is apparently contained on side beta, which reads pa-ta-jo, probably "javelins" or "spears". The nodule was found not in excavation fill, but a short distance downhill from the Northeast Workshop, an area of the Palace already known to be associated with arms-related production. It may therefore represent the delivery (or conceivably the issue) of javelins under direct control of the palace.

In sum the first season of PRAP has already made important headway toward answering major questions associated with Late Bronze Age settlement in the region of the Palace of Nestor and to the east. Future work will build on these beginnings, providing an unparalleled view of a Late Bronze Age center in operation. In addition, next season will see a new focus emerge in our work as detailed investigations of Classical Koryphasion are initiated. Investigations in 1992 already have made a substantial contribution to our understanding of the historical periods, when the settlement pattern in southwestern Messenia appears to have been rather different than in other parts of Greece. We have found relatively little evidence that in the Classical period the countryside was dotted with small farmsteads, a type of site that has now been found in abundance by other intensive surface surveys in many other regions; instead, evidence from 1991 suggests that the population was nucleated in large towns, largely in coastal locations, like that at Romanou mentioned above. Later Roman and Medieval settlement likewise appears to have been focussed on large centralized communities and we have yet to recognize evidence for the expansion of small sites that is a feature of Late Roman and Middle Byzantine settlement patterns elsewhere in southern Greece. No doubt work in 1993 will greatly improve our understanding of Messenian settlement patterns in these later periods, as well as in prehistoric times.

Appendix 1: Computerized data storage and analysis

Deborah Harlan at the University of Wisconsin maintains the central archive of PRAP and dispenses copies of records as needed by project members. All data collected in the field or in the museum are entered daily into electronic data bases in Microsoft Excel or Filemaker Pro. Other data bases contain a list of all sites and other findspots of antiquities reported prior to the start of our own research project, a regional bibliography, and an on-line summary of museum collections on display in local museums. In the field, data is stored in MacIntosh computers (two Powerbooks and a Classic), supplied to the project by the University of Illinois at Chicago and by the University of Michigan. A Personal Laserwriter, donated by the University of Illinois at Chicago, allows high-quality printed copy. Near instaneous plotting of spatial data in the field can be achieved with Foxbase. We plan to conduct much of our final analysis of spatial data by means of a Geographical Information System based on a Sun workstation, bought by the University of Michigan in support of PRAP research by Susan Alcock and Sebastian Heath. Already a Digital Elevation Model for the entire survey area has been completed. In the near future a Global Positioning Station will be employed to gather UTM coordinates for all sites in the area.

Several samples of PRAP data forms are included on the following pages to illustrate how information is handled electronically at various stages of collection and analysis.


19 PRAP Tract Data Entry Form. Sample contains basic information on artifact density collected by fieldwalkers in first examination of a tract (B-92-30).

20 PRAP POSI Record Form. Sample contains basic information concerning gridded

collection of artifacts in and around field B-92-30, now designated as POSI (Place of Special Interest) B1. After inspection of data from initial fieldwalking, it was observed that artifact densities in Tract B-92-30 and its vicinity were particularly dense and should be investigated in greater detail.

21-22 PRAP POSI Museum Feedback Form. Report from the PRAP museum staff to field teams, summarizing the analysis of finds from the gridded collection of POSI B1.

23 Geographical Information System Digital Elevation Model for PRAP Study Area. Compiled by Vince Gaffney.

24 Density of Pottery in Tracts on Englianos Ridge. Compiled in Foxbase by Sebastian Heath. Palace of Nestor in center of map.

Appendix 2: Palynological Analysis (by Sergei Yazvenko)

The study of the dynamics of the ancient environments is now an important component in archaeological research programs such as PRAP. One of the key aspects of the environment is, of course, its vegetation and the principal method at our disposal for reconstructing past vegetational histories is pollen analysis. But past pollen assemblages, when recovered from marine, lacustrine, or wetland sediments, must be realistically interpreted. The accuracy of interpretation depends on the depth of our understanding of pollen morphology, plant ecology, basin geochemistry, and the relationships between modern pollen and its source vegetation.

Objectives and Research Strategies

I. To provide an accurate and detailed record of past vegetational changes in the PRAP survey area, specifically in the neighborhoods of Tragana, Romanou, and Koryfasion. Principal questions to be answered include:

What was the structure of natural vegetation within the landscape?

Was the primeval coastal vegetation mainly pine or oak dominated?

Can past vegetation be quantitatively estimated?

II. To provide information about long-term changes and patterns of human impact on the local vegetation. Our study focusses on the following specific questions:

When did a human population first have a major impact on the environment of the region and in what way was the environment effected?

What were the main agricultural crops in the past? How did they change through time?

Can the proporation of land under cultivation at various times in the past be estimated from pollen data?

Does the pollen data support the hypothesis that olive cultivation played a minor role in Late Bronze Age Messenia, and increased dramatically during the Dark Ages (Wright 1972)?

III. To establish a detailed and precise isotope record for the pollen record of the area.

Absolute dates for Aegean palynological records are few and far between, and there are considerable obstacles to improving chronological control, since sediments are mostly inorganic. In addition, it is necessary to cope with a probable "hard water effect": 14C dates of sediments in calcareous terrain or in proximity to pre-Quaternary carbonates can be affected by old carbon and, if so, dates will be produced that are too old (Shotten 1972; Fritz 1984; McDonald et al. 1991). There is as yet no reliable and universal technique for checking the validity of radiocarbon dates for such types of sediments.

The pollen sequence studied by PRAP in 1992-93 has been obtained from the brackish coastal "Osmanaga" lagoon, the entire area of which is dominated by carbonaceous deposits. The lagoon sediments are also calcareous and most organic matter in them seems to originate from submerged aquatic macrophytes, a situation that invites a "hard water effect". Submerged plants, unlike terrestrial or partially submerged plants often draw carbon for photosynthesis from hydrocarbonate ions in the water (possibly "old" carbon) together with with atmospheric carbon dioxide.

Dates affected by old carbon can be recognized by measuring the 13C content of the sample dated (Aravena et al. 1992). The 13C content of terrestrial plants and partly submerged aquatic plants (like reeds) with a C (3) photosynthetic cycle normally ranges between -24 and -30 per mille in delta 13C values. Most submerged plants are enriched in 13C to values between -12 and -20 per mille (Hakansson 1985). A second indication of the presence of a hard water effects is the 14C content of the recently submerged plants which are presumably the main source of carbon throughout the core.

It is possible to compensate for hard water effects by trying to find terrestrial plant remains in the core sediments. This will be attempted in the coming field season. If such remains can be found, AMS will be necessary to date them since they are likely to be too small to be studied by conventional techniques. We will also core in the eastern marshy part of the lagoon in order to obtain conventional dates from organic matter dominated by reeds (i.e., terrestrial plants). These cores will be analyzed and their pollen curves correlated with those collected in 1992.


Work thus far has concentrated on three activities: pollen coring and analysis; the study of modern vegetation and pollen; and the establishment of an isotope chronology.

The best site in the PRAP survey area is Osmanaga Lagoon. Three cores were collected from the western part of this lagoon in 1992. Eighty-six pollen samples have been extracted from the longest core (420 cm.) and are now being processed. The preservation of the pollen in them ranges from excellent to reasonable. Pollen content in the samples is low, but it has been possible experimentally to increase the pollen concentration tenfold. In 1993 one core will be extracted from the eastern marshy part of the lagoon. It will be cursorily analysed, then dated and correlated with the western core so that isotope chronology for the latter may be improved.

The examination of modern pollen-vegetation relationships can offer important clues to the real meaning of the fossil pollen record. Our research in this regard includes:

a) a study of contemporary flora, in collaboration with Professor T. Georgiadis, Institute of Botany, University of Patras;

b) a compilation of a database on the ecologies of mediterranean plants;

c) the collection in 1993 for a reference collection of pollen from about 200 of the most abundant plant species that currently are growing in the area;

d) the collection of surface pollen samples from the whole range of extant plant communities in the modern landscape.

Some progress has already been made toward establishing an isotope chronology for the past vegetational history of the area. In 1992 twelve samples were collected for 14C dating from Osmanaga Lagoon, five of which were processed in Moscow. Organic content unfortunately appears to be extremely low, but one date has been obtained (see pollen diagram). Other aspects of our study of chronology have been directed toward the measurement of the 13C content of samples in order that the degree of isotope fractionation may be be estimated and so that the source and stability of carbon input over time may be traced; eight samples are currently being examined at the laboratory of geochemistry, University of Waterloo, and results are immanent. An attempt is also being made to clarify the source of carbon (whether aquatic or terrestrial plants). This is requiring the botanical identification of macroscopic plant remains in all 14C dated samples as well as the measurement of total carbonate content compared to the content of organic matter.


Pollen Cores

This year our efforts concentrated on the analysis of pollen from the longest core from the Osmanaga Lagoon (p. 34). Very preliminary results are shown on the attached pollen diagram, with the vertical scale representing depth, and thus time, and the horizontal scale indicating the percentage of various pollen types. Only 35 of a total of 120 types have been selected for representation here. The core may be divided into three zones. These are visually observable and are confirmed by cluster analysis (cladogram at the right of the diagram).


In this zone human presence is hardly detectable. The little Cerealia pollen present may well come from wild grasses. Other features of the zone may be characterized as follows:

A dominance of Pinus (indentifiable pollen indicates Pinus halepensis) and deciduous oaks (Quercus pubescens-type). Pine pollen drops rapidly in the upper part of the zone.

An appreciable proportion of steppe plants, such as Artemesia, Asphodelus albus, and phrygana shrubs (Cistus, Ononis, and Labiatae)

An abundance of redeposited pre-Quaternary spores indicates a substantial influx of terrestrial material, transported by streams.


There are dramatic changes between Zones A and B. A prominent peak in Olea (olive) pollen (35%), and many independent signs of vegetational change, should be indicative of widespread cultivation and greatly increased human-induced alterations in plant cover. Principal features of this zone are as follows:

New cultivated plants appear, including Platanus (plane tree) and Juglans regia (walnut).

Pinus (pine) pollen almost totally disappears. The most probable cause is human alteration of the landscape although other possibilities, such as climatic change, cannot be excluded.

Deciduous oaks are surprisingly not adversely effected.

Evergreen oaks (Quercus ilex-type; includes Q. ilex and Q. coccifera) become more abundant, one common result of increased human impact on vegetation.

There are other signs that macchia is present, including Pistacia pollen and some Ceratonia siliqua pollen.

A particularly exciting discovery was a few grains of Secale cereale (rye). It remains to be determined whether rye was cultivated for its own sake or was a weed in wheat and barley fields.


At the transition from Zone B to Zone C, olive pollen drops, perhaps representing a decrease in the extent of cultivation, but its percentage is never so low as in Zone A and it is unlikely that cultivation ever totally ceased. Higher in Zone C, the percentage of Olea pollen consistently increases. The evidence for human influence on the vegetation is greatest for this zone.

Deciduous oak steadily decreases to zero in very recent times, a finding that confirms observations in other parts of Greece and Turkey (e.g., at Lake Lerna [Jahns 1991]; in Crete [Bottema 1980]; and, to some extent, also in southwestern Turkey [van Zeist et al. 1975]). It is especially exciting that deciduous oak now disappears at a time of extensive olive cultivation, while in Zone B, when Olea pollen also suggests widespread olives, it was not significantly effected. It is as yet unclear whether this difference reflects climatic variation or variable patterns in human exploitation of the land.

Pollen from plants of degraded macchia, such as Pistacia, Erica, Arbutus, and Juniperus, shows a sharp increase, as does that of phrygana components like Sarcopoterium.

Vitis (vine) pollen appears, but because of the low pollen productivity of Vitis it is likely that it had been cultivated earlier.

The uppermost part of the diagram shows evidence of such non-European newcomers as Zea mays (corn), Ambrosia (ragweed) and Eucalyptus.

Modern Pollen

Many surface samples (ca. 70) of modern pollen were collected during the field season of 1992 along transects, and along the transects modern vegetation was quantitatively estimated in more than 200 plots, each 100 square meters in area. Such data will allow us to establish quantitative relationships between the nature and extent of modern pollen rain and its source vegetation.

Very preliminary conclusions are already possible on the basis of the analysis of several samples. Olea pollen is by far the dominant pollen type in most samples. Olea is both productive of pollen and its pollen is readily transported: even fairly far from the trees themselves, Olea pollen still dominates regional pollen rain. Among the oaks, evergreens are dominant in the pollen rain as they are in the uppermost levels of the samples from the Osmanaga Lagoon deposits. Most taxa represented in macchia and particularly in phrygana (Sarcopoterium spinosum, Phlomis fruticosa, other Labiatae, Calycotome villosa etc.) are heavily under-represented in the pollen rain, logically so, since most of these taxa are insect-pollinated and do not need to produce massive quantities of pollen. Very careful analysis of pollen from cores is thus necessary because even solitary pollen grains of phrygana and some macchia plants may indicate that such types of vegetation were significant components in the whole plant cover of the region.

Isotope Chronology

We have been successful in obtaining one 14C date thus far. It lies at a depth of 105-110 cm. in the longest Osmanaga core and is dated to 1240 yrs. B.P. +/- 160 yrs. (uncalibrated). Calibration by the high precision tables of Stuiver and Pearson (1986) would put it about 130 years earlier. Such a date is not unrealistic for deposits of ca. 4 m. deep, if it is assumed that sedimentation in Osmanaga Lagoon probably began in the second millennium B.C. More dates are, however, sorely needed, as is a detailed analysis of the hard water effect and the carbon source.


Aravena, R., B.G. Warner, G.M. MacDonald, and K.I. Hanf 1992. "Carbon Isotope Composition of Lake Sediments in Relation to Lake Productivity and Radiocarbon Dating," Quaternary Research 37 (1992) 333-45.

Bottema, S. 1980. "Palynological Investigations on Crete". Review of Palaeobotany and Palynology 31: 193-217.

Filina, N.I. and S.B. Yazvenko 1989. "Quantitative Birch (Betula L.) Pollen Separation by the Analysis of Canonical Variates". In The 6th All-Union Conference on Palynology (abstract), December 18-22, 1989, Minsk, USSR, pp. 294-95 (in Russian).

Fritz, P. 1984. "Radiocarbon Dating of Marl Deposits. Can We Assess the Hardwater Effect?". In W.C. Mahaney ed., Correlation of Quaternary Chronologies (Norwich CT): 15-21.

Hakansson, D. 1985. "A Review of Various Factors Influencing the Stable Carbon Isotope Ratio of Organic Lake Sediments by the Change from Glacial to Post-glacial Environmental Conditions." Quaternary Science Review 4: 135-46.

Jahns, S. 1990. "Preliminary Notes on Human Influence and the History of Vegetation in S. Dalmatia and S. Greece," in S. Bottema, G. Entjes-Nieborg, and W. van Zeist eds., Man's Role in the Shaping of the Eastern Mediterranean Landscape (Rotterdam): 333-40.

MacDonald, G.M., R.P. Beuken, and W.E. Kieser 1991. "Radiocarbon Dating of Limnic Sediments: A Comparative Analysis and Discussion." Ecology 72: 1150-1155.

Pons, A. and P. Quezel. 1985. "The History of the Flora and Vegetation and Past and Present Human Disturbance in the Mediterranean Region." In C. Gomez-Campo ed., Plant Conservation in the Mediterranean Area (Dordrecht): 25-43.

Rackham, O. 1982. "Land-use and the Native Vegetation of Greece." In M. Bell and S. Limbrey eds., Archaeological Aspects of Woodland Ecology (Oxford, BAR International Series 146): 177-198.

Shotten, F.W. 1972. "An Example of Hard-water Error in Radiocarbon Dating of Vegetable Matter." Nature 240: 460-61.

Turrill, W.G. 1929. The Plant Life of the Balkan Peninsula: A Phytogeographical Study (Oxford).

Van Zeist, W., H. Woldring, and D. Stapert. 1975. "Late Quaternary Vegetation and Climate in Southwestern Turkey." Palaeohistoria 17: 53-143.

Wright Jr., H.E. "Vegetation History." In W.A. McDonald and G.R. Rapp Jr., The Minnesota Messenia Expedition: Reconstucting a Bronze Age Regional Environment (Minneapolis): 188-99.

Yazvenko, S.B. 1991. "Modern Pollen-Vegetation Relationships in the Southeastern Caucasus." Grana (Stockholm) 30: 350-56.

Yazvenko, S.B. 1993. "A Palaeoecological Study of Holocene Peat Bog Dynamics Based on the Analysis of Pollen, Algae, Rhizopod and Fungal Remains (the West Caucasian Mountains)." Bulletin of the Moscow Society for Natural Research, Biology series. In press.

Zohary, M. 1973. Geobotanical Foundations of the Middle East (Stuttgart).