Early hominin presence during the Middle pleistocene at Marathousa 1 in the Megalopolis basin, Greece

(1)

Luke Oomen

S1791117, Leiden University

EARLY HOMININ PRESENCE DURING THE

MIDDLE PLEISTOCENE AT MARATHOUSA

1 IN THE MEGALOPOLIS BASIN, GREECE

ARCHAEOBOTANICAL RESEARCH OF AN ELEPHANT KILL SITE LOCATED

IN AN ACTIVE LIGNITE MINE

(2)

Photo on the front page: Fig. 1. Overview of the Megalopolis basin with the powerplant in the background.

(3)

EARLY HOMININ PRESENCE

DURING THE MIDDLE PLEISTOCENE

AT MARATHOUSA 1 IN THE

MEGALOPOLIS BASIN, GREECE

ARCHAEOBOTANICAL RESEARCH OF AN ELEPHANT

KILL SITE LOCATED IN AN ACTIVE LIGNITE MINE

Luke Oomen S1791117 BA Thesis Supervisor: Dr. M.H. Field Specialization: Archaeobotany Leiden University, Faculty of Archaeology

Final version 01-07-2020

(4)

Chapter 1: Introduction

When talking about archaeological research in Greece, it is safe to assume that most people think of the rich history of the ancient Greeks which has been studied intensively over several centuries. Greek ruins scatter acrossthe landscapes, with the Acropolis of Athens as a prime example. However, more stories, older stories, are hidden belowthat Classical landscape. Palaeolithic archaeological research is not a common area of

interestamong the archaeological studies in Greece. The Bronze age and Classical period are the main focus of most academic studies. Another reason for this fixed focus is because of the geological activity in Greece. Erosion, uplift, subsidence and deposition of geological material cause the Palaeolithic archaeological record to be obscured and difficult to find (Harvati et al. 2009, 131). Furthermore, the change in water level throughout the years between the inter-glacial and glacial periods have caused sites to disappear or to become unreachable (Wenban-Smith 2002, 3). Nevertheless, this lack of research leaves a gap in the archaeological information surrounding Palaeolithic Greece. For example, Greece can hold important information on the migration of early hominins and animals during the Pleistocene from Africa to Europe. As it is today, Greece could have been and probably was one of the most likely routes for migration from Africa to Europe. Moreover, research suggests that Greece contained refugia during multiple ice ages which were essential for flora and fauna, which also includes hominins to survive these cold periods. This further provides evidence to why Greece is a potential route for migration during the Pleistocene (Harvati et al. 2009, 131).

Numerous sites are known from the Palaeolithic period in Greece. For example, Kokkinopilus in Epirus (Western Greece); Petralona cave in Macedonia (Northern Greece); Rodia in Thessaly (Eastern Greece); and Apidima Cave A in Mani (Southern Greece) (Harvati et al. 2009, 132). More than 250 sites and counting have been discovered and analysed. The type of sites which are being studied are caves, rock shelters and open-air sites. One of the most recent open-air sites which are being studied, is Marathousa 1 (See fig. 2 for location). This site is dated to the Middle Pleistocene around 400-500 ka, which puts it roughly in MIS 12 (starts 478 ka ends 424 ka) corresponding to the Anglian or Elsterian (Jacobs et al. 2018, 171). Only a few sites have been dated to the Lower Palaeolithic, most of them having disappeared due to geological activity. Because Marathousa 1 is located in a lignite mine in the Megalopolis basin in Greece MAP?, it was dug out and therefore accessible.

(7)

Fig. 2. A map of the Megalopolis basin and its location in Greece. On the right is a stratigraphic diagram of the area (Panagopoulou et al. 2015, 1).

The site was first surveyed in the 1960’s, when a human tooth was found. This raised the expectation that Marathousa 1 could potentially hold evidence for early hominin

activity. The importance of the site was rediscovered during a survey in 2013 withing the “Palaeoanthropology at the Gates of Europe” project, also known as the PaGE project. Several lithic artifacts, and the cranial of an elephant from the species Palaeoloxodon antiquus were found during the survey and excavations started shortly after. After two seasons of excavation (2014 and 2015) lithic artifacts had been uncovered

stratigraphically in context with the elephant skeleton and other faunal remains (Tourloukis et al. 2017, 171). Some of the bones show striations most likely caused by human activity. Furthermore, botanical samples have been taken from the same context of the elephant skeleton, above it and below it. This was to reconstruct the ecology of the site in the time of human activity and to observe the change of the environment throughout the periods.

(8)

- Research goals

For my thesis I have been working with a sequence of six botanical samples which were taken in the context of the elephant skeleton. These six samples represent a change of the local vegetation and environment from the period before the elephant died, until the skeleton was completely buried by geological processes, this all taking place somewhere between 500 ka and 400 ka in the Middle Pleistocene. The main research goal is to reconstruct the environment of the periods before, during and after the elephant lived and analyse the changes that occurred between these periods. This is an interesting thing to study because the period of human activity at this site coincides with a cold or dry period (Panagopulou et al. 2015, 2).Furthermore, this site is one of the oldest sites in Greece and gathering information on past environments with data from Marathousa 1 will aid in the research towards Palaeolithic environments in Greece.

Another reason to study the botanical remains is to answer the research question of this thesis: Why was there human activity happening during the Middle Pleistocene at the site of Marathousa 1 in Greece?

Several sub questions with which I will be exploring the possible answers for the research question are:

- What form of archaeological evidence do we see of the early hominins that were present at Marathousa 1?

- In what kind of environment were these early hominins living?

- Can this site be connected to the migration of early hominins from Africa to Europe in the Palaeolithic, if yes how?

With these questions I am going to delve deeper into the main points of attention that exist within the study of the Marathousa 1 site. First, I will explain the research

methods, how the data was generated and what the results are, which will be depicted in several graphs. Second, I am going to paint a picture of the environment and of the hominins themselves and their influence on this environment which we can see in the archaeological evidence. Next, I will try to answer the question about what Palaeolithic environment looked like during the period of hominin activity and also why there was hominin activity at this site during the given period, and I will connect the answer to the last question with several theories on migration. And last, I will briefly summarize the

(9)

evidence and discuss possible explanations before concluding and given my answer on my research question.

(10)

Chapter 2: Research

Before the questions of this thesis will be addressed and a satisfactory answer will be formulated,the data will be portrayed by explaining the methodology with which the data was gathered and how the data was shaped into processable lists and tables.

- Methodology

To make it clear in advance, the author did only work on two of the samples (20-30cm and 50-60cm) in the later stages of the analysis in the lab and was not part of sampling in the field or subsampling in the lab. Therefore, the general procedure will be explained of how bulk samples were processed in the archaeobotanical lab at the faculty of archaeology of Leiden University. The bulk samples were taken in 2017 at the site of Marathousa 1 in area A (938/672 in the grid). The elephant skeleton is located in area A and the 6 samples were taken from the segment next to the elephant for every 10 centimetres with a volume of 2 litres. The sampling was done from bottom to top to minimize the risk of contamination. Afterwards the bulk samples were gathered in plastic bags and sealed before they were shipped off to the Netherlands (Barbier 2019, 26). When they arrived at the archaeobotanical lab in Leiden the samples were

subsampled by taking 200 cm3 out of the sample. This was measured through carefully adding the dirt to 200 ml water in a 500ml beaker until the water level reached 400 ml. This mixture was poured into a plastic bucket with the specifications of the sampling written on the lid. Regular soap (C₁₇H₃₅COONa WASHING UP LIQUID!) was added to the mixture, which helps with the segregation of the soil particles making it easier to pick out the macro botanical remains.

The rest of the bulk samples were kept in a fridge to have a backup for when the subsamples are contaminated or to use for different studies. The subsamples were placed in the fridge as well and left for about a week to make sure segregation could take place. The material is left in the fridge for longer if it is more difficult to be segregated. The next step is to sieve the material through a series of four meshes. The sizes are, from the largest mesh size to the smallest and placed in this order from top to bottom 1000 μm, 500 μm, 250 μm, and 150 μm. The content of the bucket was

deposited in the top mesh and washed down with a lukewarm weak stream of water. The meshes are gently shaken during the sieving process, because rubbing the residue with fingers or other tools will rather damage the already fragile macro botanical

(11)

remains instead of getting them through the mesh. This was done until the water coming out at the bottom mesh had no or almost no black particles in it. If one is not satisfied with the results of the sieving, the material is put back in the bucket to soak for a few more days. Most of the time, the bigger sizes (1000μm and 500μm) will be put back in the bucket to soak, because these two sized are likely to have bigger denser clumps of material which will not fall apart easily. Otherwise, the now sorted to size residue is divided into four beakers filled with water to prevent dehydration. The following step is to pick out any recognizable and/or interesting bits of botanical material (this is the part from which the author helped with processing the samples). The four different sizes of residue were gone through separately. This makes it easier to pick out the macro botanical remains. In this process were used: a stainless steel

tweezer (for the larger, sturdier specimens); a modified small soft brush with a few hairs (for the smaller, more delicate specimens); Petri dishes, filter paper, preservative (one part water, one part ethanol, one part glycerine) to keep the remains in while they were picked out, identified and counted; and a microscope which can magnify from 6.3 to 40 times. Residue and some water from one of the beakers was put in a Petri dish

underneath the microscope and with the tweezers or a brush it was examined for botanical materials which were recognizable or interesting looking.

These specimens were identified up to species level if possible, otherwise to genus or even family level. To identify the botanical remains the Digital seed atlas of the Netherlands by R. Cappers et al. (2006), a large collection of reference material and other literature, and the vast knowledge of the botanical specialists walking in and out the lab were called upon. The nomenclature used in Cappers book is also used in this thesis. If a specimen was unrecognizable it was labelled as “unidentified”. All the remains were sorted on their identification and put in formations to count. After one beaker was done, the remains were counted and put into small vials filled with

(12)

- Generating the data

In 2017 the samples were partially processed by students from Leiden University under supervision of Dr. Field. In 2018 and 2019 the samples were further finished by master student S. Barbier and by the author.

During the picking stage, the data on the number of fossils and types of taxa was recorded in a notebook. After the picking, counting and identifying was completed a table was generated in Word Excel. Dr. Field kept a separate notebook with the data for himself as a backup. All the plants were put in groups according to the environmental zones in which they thrived. The zones are woodland and shade tolerant; grassland, disturbed and open land; waterside and damp ground; aquatic; and unclassified. And put in alphabetical order within these groups. This procedure makes the data more transparent for further analysis. With the data, bar charts have been made by using an analytic program called ‘Tilia’ (see appendix 1 to 4). With this program it is possible to create graphs portraying the number of fossils found per specimen and grouped in their appropriate zone. Furthermore, it is possible to show additional data like sample depth and lithology. With these graphs one can compare the number of fossils found between the different taxa, but also through time between the same taxa. The results of this comparison give an idea of what the environment looked like in the appropriate period and how this environment changed between the periods to which the different samples belong.

A total of four graphs, instead of one, have been made each representing an environmental zone. Except that ‘woodland and shade tolerant’ and ‘grassland, disturbed and open land’ have been merged because of the relatively small number of taxa found in these groups. The decision to split the data into multiple graphs was made because of the number of taxa making the graph too obscure to be utilised in any research. Also, a list of all the taxa will be provided with the amount of fossils found in a table, instead of a graph, along with a list of the environmental data of the taxa (see appendix 7 and 6). These tables along with the graphs made with Tilia will be used for the environmental reconstruction of Marathousa in the Middle Pleistocene.

(13)

Chapter 3: General information on the site

- P.a.G.E.

The Marathousa 1 site has been a part of the PaGE project since 2013 when the first survey after a long pause of archaeological activity took place in the Megalopolis basin. The project Palaeoanthropology at the Gates of Europe was initiated as a Research Council Starting Grant awarded to Katerina Harvati of the University of Tübingen in 2011 with the goal to fill in the gap of information that exists in the research towards

palaeoanthropology in south-eastern Europe. This research is mainly focussed on the fact that this area is located on the crossroads of Europe, Asia and Africa (Harvati et al. 2013, 43). Therefore, it can be expected that the idea this area was used by early hominins to cross over to Europe from Africa has been widely hypothesized about (Dennel and Roebroeks 2005; Derricourt 2005; Muttoni et al. 2010; Parés et al. 2013). Of course, this is only one of the many theories existing in the field of early hominin

migration and many variations have been proposed. The excavations at MAR 1 within this project are being conducted by a team of researchers from the Palaeoanthropology department of the University of Tübingen in Germany and from the Ephorate of

Palaeoanthropology-Speleology of the Hellenic Ministry of Culture in Greece and are still ongoing.

- Geological/geographic background

The site is located in the Pleistocene deposits of the Megalopolis basin surrounded by the mountainous region of the Peloponnese located in a depression which was created during the Late Miocene/ Pliocene because of tectonic movement. The find bearing layer was buried underneath 4 meters of sediments and is located in between two layers of lignite, at the edge of an active lignite mine. It is highly likely that part of the site has been destroyed by the mining activities. The lignite, also known as brown coal, was used as fuel for the powerplant located close to the site. These lignite layers are part of the Marathousa Member of the Choremi formation. This formation consists of fluvial sand, lacustrine sand and lignite layers which accumulate to 250 meters of sediment and is thought to be the result of the cycle of glacial and interglacial periods (Giusti et al. 2018, 138). The Marathousa 1 sequence can be divided into an upper (UA1-UA3 and UB1-UB5) and a lower (UA4-UA7 and UB6-UB10). The sequence is described by units or UA for units from area A and UB for units from area B. Each unit is linked to a

(14)

layer in the geological stratigraphy of the site. UA1 and UB1 are part of the same

geological unit which is lignite seam 3 and UA7 and UB10 are part of the same geological unit which is lignite seam 2. The archaeological horizons are located between the units UA3 and UA4 in area A, and between UB4 and UB5 in area B. This find layer is located in dark grey, massive silty sand, rich in organics and clay intra-clasts with fine pebbles. The sediment has been interpreted as mudflows that flowed into the paleolake and

therefore the location of the finds are secondary depositions and not their original discard position (Tourloukis 2018, 49).

In the Megalopolis basin two areas have been excavated. In area A, the partial skeleton of the Palaeoloxodon antiquus was found along with some lithic artifacts and in Area B, 60 meters to the South, a more elaborate lithic collection was found together with several bones belonging to the elephant, cervids and carnivores. The excavation started as a salvage excavation in 2013 and transformed into a full-fledged research project going on for several years considering a stretch of 100 meters along the exposed parts of the lignite mine (Panagopoulou et al. 2018, 36).

(15)

(16)

- Dating of the site

In the article of Jacobs et al. (2018) it is described how the age of the site was estimated using luminescence dating techniques. These techniques can determine when sediments and associated artifacts and fossils are buried by measuring the amount of electrons which were obtained during explosion to external radiation, left in mineral grains after natural decay occurs from the moment the grains were buried by natural forces. This is called the equivalent dose. To calculate a result, the environmental radiation has to be taken into account. By dividing the equivalent dose by the environmental dose, the outcome will be the burial date in calendar years ago. Quartz optically stimulated luminescence (OSL) and potassium-rich feldspar infrared stimulated luminescence (IRSL) both failed to give an accurate outcome. Only the minimum age could be determined. Therefore, the multiple aliquot regenerative dose (MAR) method was introduced together with the post-infrared IRSL (pIRIR) method to be able to use multiple samples for the procedure, one from Marathousa and one from the Choremi mines.

Furthermore, it was needed to measure the palaeo-water volume of the basin at the moment the sediment of the find bearing layer was deposited. As water weakens the external radiation, meaning that the estimated age of the site would be younger than the actual age (Jacobs et al. 2018, 171).

The following dates have been estimated from the samples which were taken. The layers above and containing the fossils and artifacts have been dated to between 380 ± 48 ka and 480 ± 39 ka. And the layers underneath the fossils and artifacts have been dated to between 438 ± 38 ka and 508 ± 42 ka. Which results in an average of 460 ± 29 ka (Jacobs et al. 2018, 173). This dates the site roughly to MIS 12 which corresponds to the Anglian or Elsterian glacial in the Lower Palaeolithic.

(17)

- Background on bone material

o Avian

The group of avian fauna remains found at this site has the most diversity in taxonomy among the vertebrate groups and because of the difference in preference between the different avian species, studying the avian remains will add to the reconstruction of the palaeoenvironment (Michailidis et al. 2018, 85). In total 121 skeletal elements have been uncovered, 86 from area A and 35 from area B. The remains have been gathered through excavation or water-sieving with a 1 by 1 mm mesh. 23 different species have been identified by using comparative material from the Malcolm H. Wiener Laboratory (Michailidis et al. 2018, 87). Most of the species found are part of bird families

associated to fresh water environments (e.g. Anas crecca; Fulica atra; Spatula clypeatra; Anhinga sp.) and from this can be concluded that water was present in this area in the form of a pond, lake or marsh. The second largest group was associated with steppe-grassland environment (Michailidis et al. 2018, 90).

Most of the bones do not show signs of weathering or only slight weathering because of surface weathering or digestion. The reasons for this phenomenon are on the one hand the taphonomic processes quickly burying the remains before weathering could occur, and on the other hand avian bones weather rather fast and disappear completely. This, together with how the lithics and other faunal remains have been deposited (see below), suggests that the remains were not transported over a long distance and are locally deposited (Michailidis et al. 2018, 87).

Although some of the avian remains have been found in association with lithic tools, there is no evidence that these birds were exploited as a source of food by the early hominins. The accumulations of bones bear no marks of butchering or gnawing by hominins. A different explanation is the influence of animal predators like foxes, wolves or even birds of prey. However, no clear indications have been found on the bones or in the ratio in which different types of bones were found to single out one dominant influencer. It probably is a combination of different species and depositional processes (Michailidis et al. 2018, 89).

o Large mammals

Several different species of large mammals have been identified from the remains which have been uncovered at the side. The most important and largest animal unearthed is

(18)

the Palaeoloxodon antiquus, which was represented in the assemblage by its cranium, vertebrae, ribs, a humerus, an ulna, the pelvis, a femur, a tibia and multiple bones belonging to the feet. All the bones which were found in area A are associated with each other, but the bones are not in an anatomical position. By measuring the available limb bones, the shoulder height of the elephant was calculated to be approximately 3.7 m. The body mass was estimated to be 9 tonnes based on recent species of elephants. And the age was estimated by measuring the weathering on the teeth which were found, to be 64 to 71 years old. This animal had likely a longer life span than the modern elephant considering its larger size. Therefore, it should be in its late adulthood. Further, it was established that the elephant was male based on the progress of the epiphyseal fusion sequence and the size of the animal (Konidaris et al. 2018, 68). Additional bones from the P. antiquus (tusk, vertebra, rib and tibia fragments) have been found in area B and in between the two area. While there is no clear evidence that states the bones in area A and area B are associated to each other, the possibility remains that there is.

There is a variety of other large mammals which were identified in the assemblage. Castor fiber, or the beaver was identified by the find incisors, cheek teeth and postcranial bones. Lutra simplicidens, or the otter was represented by a mandible fragment. Mustela sp. (large-sized) is not identified to species because only a tibia was found. A lower carnassial was found belonging to the Vulpes sp., or the fox. The axis and a foot bone were found from Canis sp., or the wolf. And from the Felis sp., or the wildcat was present in the form of an upper carnassial. A tooth and a foot bone were found from the Hippopotamus antiquus, or the hippo. A foot bone was found from the genus Bison. The Dama species, or fallow deer was recognized in a mandible with several teeth. Cervus elaphus is represented by teeth and postcranial elements (Konidaris et al. 2018, 71-74).

The presence of these animals all fit in the given time frame within Marathousa 1 has been dated. Also, in a paleoecologic perspective the assemblage of large mammals is cohesive and fitting within the known ecological frame of Marathousa in the Middle Pleistocene. The region was a refugia for animals during the glacial MIS12. The elephant species normally occur in interglacial environments, just like the fallow deer which is not flexible in living environments. This in contrast with its cousin the red deer, which is adapted to live in all kinds of environments, glacial and interglacial. These species also point towards an open vegetation with possible shrubs and tree vegetation. The (semi-) aquatic hippo, beaver, and otter indicate the presence of a large slowly flowing body of

(19)

water like a lake with aquatic vegetation. The cat, dog and fox are flexible and fit in the picture of a lake shore with a lot of vegetation in with small mammals can hide

(Konidaris et al. 2018, 74-76).

The area was likely a refugia for a variety of flora and fauna, therefore it is not strange to see marks made by early humans on the remains of these large mammals. First of all, the lack of rounding on the bone surfaces both from area A and area B indicates little to no transport before the bones were buried. And the well-preserved surfaces had almost no cracking or flaking. This is important because it means that the remains with

evidence of butchering were buried rather quickly after disposal and not carried with by the hominins for further use or that they were brought from a different area (Konidaris et al. 2018, 77-81).

o Small mammals

Compared to the other faunal assemblages, the small mammal assemblage is by far the smallest group of taxa but also has the smallest amount of remains found. Some of the taxa are represented by just a few teeth. However, there is a lot of difference between the teeth when it comes to small mammals and therefore it is possible to identify those animals. The areas have overlapping taxa. In area A five different taxa have been found: Crocidura sp.; Arvicola mosbachensis; Microtus cf. arvalis; Apodemus sp.; and Alactaga indet. In area B three different taxa have been found: Crocidura sp.; Arvicola

mosbachensis; and Microtus cf. arvalis (Doukas et al. 2018, 98).

Although, only a small number of small mammals have been found, these taxa can still be important for palaeoecological reconstruction and biostratigraphical dating of the site. Unfortunately, because of the small amount not much can be said about the ecology. The most informative is the high concentration of Arvicola mosbachensis, which is a species of vole that preferably lives in marshes close to ponds and lakes. For

biostratigraphical dating the difference in molars is important. Rapid evolutionary processes cause the molars of some rodent species to change and these can be used to date a site, along with the presence and absence of certain species. The only species found at MAR 1 with a clear evolutionary path in the changing of the teeth and therefore can be used for dating is the Arvicola mosbachensis or the water vole. By comparing assemblages from different sites dated to different periods, a timeline can be drawn for the evolutionary stages of the water vole. The water vole found at MAR 1

(20)

could be roughly dated to the Early Middle Pleistocene or MIS 11 (Holsteinian) (Doukas et al. 2018, 101-102).

- Background on botanical material

The already available data on plant macro- and microfossils comes from research done by Field (Field et al. 2018) on the samples gathered during the excavation season in 2014. Samples were recovered of carpological remains, wood, wood charcoal, phytoliths and diatoms. In 2014 carpological remains and wood charcoal were extracted via a floatation method and in later years via wet sieving. The remains are preserved very well because of waterlogged and anaerobic conditions (Field et al. 2018, 110). In this thesis I will only discus the data generated by researching the carpological remains from the samples taken in 2014 by Field to be able to compare those results with the results of the macro botanical research done on the samples from 2017 portrayed in this thesis. A total of forty-four samples have been taken from the excavated layers in area A and B. Each sample had an average volume of 7.3 litres. The samples were processed via a flotation device on the site. The device divided the light floating fossils and the heavier sinking residue. The floating fossils was sieved in through a 0.3mm mesh and the heavier residue through a 1mm mesh. The residue was then dried, and nine samples were used to pick out mainly carpological remains. After identifying these remains, they were used to reconstruct the environment during the period of hominin activity (Field et al. 2018, 110).

In their article Field et al. (2018) describe the carpological assemblages from area A and area B as diverse and dominated by aquatic, waterside and damp ground taxa. Examples of aquatic taxa identified are Zannichellia palustris, Brasenia schreberi, Euryale ferox, Nuphar lutea, Nymphea alba, Aldrovanda vesiculosa, Azolla filiculoides, and Salvinia natans. This is a collection of submergent, emergent and floating plants. A plant present in large amounts in all the samples is Potamogeton or pondweed. Other plant taxa found in the samples consist of a lot reeds like Botumus umbellatus, Cladium mariscus, Oenanthe aquatica, Scirpus lacustris, Sparganium erectum and Typha sp. Examples of woodland and shade tolerant species which have been found in the assemblages are Acer sp., Rubus fruticosus and Sambucus cf. nigra. These taxa probably have grown in the near vicinity outside of the wetter areas in which the aquatic taxa grew (Field et al. 2018, 113).

(21)

The data gathered was put in a table in the form of a present and absence list in which is indicated which taxa have been found in the samples per area per stratigraphic unit. This table was used to create a picture of the Palaeolithic environment. The findings were that the environment did not change a lot from the period before the elephant died, to the period in which the elephant died, to the moment of after the death of the elephant. The article argues that this indicates a short period being portrayed in the sediments researched (Field et al. 2018, 113). The general picture that was sketched with the carpological assemblage shown in the article is of a water body with slow moving or stagnant water up to two meters deep with a possible reed swamp within the vicinity and a dispersed amount of trees and shrubs close to the swampy area in a temperate environment. The other proxies (wood (charcoal), diatoms and phytoliths) used in the article by Field et al. (2018) elaborate on and confirm the environmental reconstruction.

(22)

Chapter 4: Anthropogenic presence

What form of archaeological evidence do we see of the early hominins that were present at Marathousa 1?

Before claims can be made about the connection of this site with the possibility of it being a part of a larger migration process of the early hominins from Africa to Europe in the Middle Pleistocene, it has to be established what kind of archaeology we find at the site of Marathousa. And what this archaeology tells us about the type of activity and the reasons for why there is hominin activity. By looking at the lithic and bone artifacts and their association with the faunal remains and butcher marks found on these remains, the question why early hominins came to this place can be addressed.

- Lithic artifacts

A total 4 flakes and 63 chips and debris was found in area A close to the elephant skeleton. This amount is a low number compared to the number of lithic artifacts found in area B. The context of the lithic assemblage includes the elephant bones with

anthropogenic cut marks, which are still in approximately anatomically correct position although in a tighter grouping; it also includes remains of other fauna, which will be discussed below (Tourloukis et al. 2018, 49). Area B contained the largest lithic assemblage with an amount of 1105 pieces. In area B the artifacts are in context with some elephant bones and other faunal remains and are in the same stratigraphic layer as the elephant skeleton in area A. The number of flakes found in both area A and B points towards a flake-based industry (Tourloukis et al. 2018, 49). The raw materials used for the lithics are red radiolarite, grey flint, limestone, quartz, brown radiolarite and black flint (see fig. 5). Most of the materials could have been found locally in the form of water transported cobbles. The flint could have been transported from somewhere else, because it is uncommon in this area. The same goes for the red radiolarite which has a source 4 to 5 km from the site.

The condition of the lithic artifacts in area A as well as in area B are pristine. The edges are still sharp, and the artifacts do not show any sign of rounding through abrasion or erosion by water transport. This coincides with the theory that the artifacts and other materials were transported by a fine-grained mudflow or a hyperconcentrated flow (a waterflow filled with a fine-grained matrix of debris). The artifacts could therefore only have been moved locally and not have come from outside the site (Tourloukis et al.

(23)

2018, 50). The composition of the lithic assemblage consists of a high percentage of flakes (63,3%) and chips (15,1%) which are believed to belong to a flake-based industry. The percentage of the debris is 10.1% with mainly small debitage. There are only 7 cores present (0,6%) and also an amount of 42 tools (3,6%) (See fig. 6). This composition matches the assemblages of simple knapping events such as producing, using and repairing tools.

Most of the tools have been made out of flakes. The preferred flakes to use as blanks for the tools were the thicker and longer flakes. The average sizes of the debitage flakes found at the site were between 15 and 25 mm in length and width, while the tools have an average length of 28 mm. The thickness of the tools has an average twice as big as that of the flakes. This preference shows that the hominins made an economic decision, they choose for tool lifespan and reusability instead of light and efficient transport. Lips on the cores show that hard hammer percussion was used on these cores which is common in order to knap longer and thicker flakes, which coincides with the economic decision (Tourloukis et al. 2018, 55).

Fig. 5. Flakes from the MAR 1 site: 1-4, 7, 8, 11) red radiolarite; 5) grey flint; 6) limestone; 9) quartz; 10) brown radiolarite; 12) black flint. Source: Tourloukis et al. 2018, 54.

(24)

Fig. 6. Examples of retouched tools from MAR 1: 1) a scraper on a pebble; 2, 5, 7, 9) a composite tool; 3) a denticulate; 4) a notch; 8) a scraper fragment. Source: Tourloukis et al. 2018, 57.

(25)

- Bone artifacts

A far smaller amount of bone artifacts has been found than lithic artifacts at the MAR 1 site. All the artifacts were found in the find horizon in area B (UB4). These bone artifacts are important, because no other bone artifacts have been found at any site in Greece predating the Upper Palaeolithic. However, different processes could have created pseudo artifacts: artifacts which look like they are anthropogenic but are produced by for example animal trampling, gnawing by rodents and

carnivores, or natural causes (Tourloukis et al. 2018, 59).

Therefore, the possibility that these artifacts were not created by anthropogenic forces had to be excluded. This was done by closely examining the bone artifacts and comparing them to bones with marks found at the site which clearly were pseudo artifacts. It is possible that flaking occurs when carnivores gnaw on bones. this flaking would be accompanied by damage done to the bone surface because of the gnawing. No such marks were found on the bone artifacts believed to be anthropogenic. Clear gnawing damage was identified on other bones to be able to make the comparison.

Artifacts found were a bone flake (1), a denticulated bone flake (2), and a bone percussion tool (3) (See fig. 7: 1, 2, 3). The bone flake is the smallest of the artifacts. It is believed to be anthropogenic because of all the flake traits combined and the large width of the flake (35 mm) is typical for a flake knapped from a long bone shaft. This flake could have been produced during the processing of bone marrow (Tourloukis et al. 2018, 60).

The denticulated bone flake is identified as anthropogenic because of the direction in which the flake was produced. The force which produced the flake was directed from the medullary surface (inside of the bone) to the cortical surface (outside of the bone). The force that was used by most carnivores is on the opposite direction. Another factor is the dimensions of the bone flake, which is close to the dimensions of the lithic flake which were used to make tools out of. Although, some species of carnivores (lions and hyenas, which did occur in Megalopolis) create flakes with closely resembling patterns. This denticulated flake is still believed to be

anthropogenic because these species have not been found in the bone assemblage (Tourloukis et al. 2018, 60).

(26)

The final bone artifact of the assemblage is a diaphysis fragment from most likely an elephant long bone. This fragment has been used as a percussion tool shown by the cluster of percussion marks on the one end of the bone fragment and further following the fragmented used edge. Also visible on the one side are a flake scar and cut marks. All these elements combined indicate that this bone fragment was modified by hominins and probably used as a soft hammer. Although, there is a lack of soft percussion marks on the lithic artifacts to prove the use of this tool, it is shown in a knapping experiment that it is often difficult to identify the difference between hard and soft percussion (Tourloukis et al. 2018, 60).

Fig. 7. Bone artifacts from MAR 1: 1) Bone flake; 2) Denticulated bone flake; 3) Bone percussor. Source: Tourloukis et al. 2018, 60.

(27)

- Evidence of butchering

In area A cut marks were found on some of the elephant bones (tibia and astragalus). Cut marks produced by stone tools can be identified by the striations which are v-shaped in cross-section. Striations left by carnivores would be u-v-shaped in cross-section, which were not found on bones in area A. Trampling by animals is also unlikely because of the lack of a stony underground and this would leave more broad grooves.

In area B cut marks were found on a broken rib fragment of an elephant along with scrape marks. These marks were all clustered close together on the bone. These can be contributed to the removal of the periosteum during butchering or before breaking the bone for marrow extraction. This indicates an early accessibility of the hominins to the carcass. Other taxa which present cut marks are a fallow deer and a small/medium mammal. In area B carnivore gnawing has been found on bones from the elephant, fallow deer and another cervid. Because these are the same taxa as the ones with cut marks, it could indicate a race for food between carnivores and hominins. The combination of stone tools, cut marks and broken bone fragments point towards the butchering of certain taxa by early hominins and the knapping of bones to make tools (Konidaris et al. 2018, 77-81).

Notable to say is that only remains of some of the large mammals show signs of butchery. The avian fauna and the small mammals show no sign of exploitation. Of course, there is the possibility that these signs are not visible because of a variety of reasons. As mentioned above avian remains can be very delicate and weather rather quickly resulting in disappearing altogether. Another possibility is that the captured birds were taken with to be processed somewhere else. If you look at the practicality of taking with an elephant against taking with a duck, it becomes clear that there exists a change. Not many species belonging to the small mammals were found (see above) and the amount of remains representing these rodent-like animals was not abundant. Therefore, not much can be said about the exploitation of these animals.

(28)

Chapter 5: Environmental reconstruction

In what kind of environment were these early hominins living?

To make the environmental reconstruction the data will be used that was generated by researching the samples taken at the Marathousa 1 site, in Greece. Two tables and four graphs have been created with the data and these will be used as a base for the

reconstruction.

- Environmental reconstruction based on the botanical data

By interpreting the present taxa per sample and describing the change in taxa appearing or disappearing throughout the samples, a reconstruction of the environment and how it changed through time will be given. Chronologically, the samples are ordered from bottom to top. I will start with sample 6, which is the oldest and end with sample 1.

o Sample 6 (60-50cm)

The remains found in this sample describe a marshy area in which different types of reeds (Typha sp.; Schoenoplectus lacustris) and some grass-like plants (Cyprus longus; Cyprus fuscus; Juncus sp.) grow in or close to the water together with ferns (Dryopteris sp.) and possibly trees. Possibly, because no carpological remains have been found in sample 6, only the remains of Cenococcum geophilum which is a fungus that is being associated with tree roots. Further, we see taxa which grow in or close to the water in marshes (Lycopus europaeus; Damasonium alisma) or waterbodies like a pond or lake (Zannichellia palustris; Myriophyllum sp.; Salvinia natans; Azolla filliculoides; Ranunculus subg. Batrachium; Potamogeton crispus). An indicator of deep water is Nymphea alba which is a floating plant and needs a depth of 30 to 150cm for it to be able to grow. Also, remains of the Characeae family have been found which indicates the presence of algae in the water. No remains have been found from the ecological zones: Grassland,

disturbed and open ground; and Woodland and shade tolerant.

o Sample 5 (50-40cm)

The first species (Ranunculus cf. sardous) in the category of ‘Grassland, disturbed and open ground’ has been identified in this sample. This could indicate a grassland in the area; however, this cannot be said for certain because of the low number of fossils found and the fact it is the only species in this group. In the aquatic group most of the taxa have increased in amount and others appeared in the picture, like the floating

(29)

Nuphar lutea which commonly seen along with the already present Nymphea alba. These taxa could possibly compete with S. natans and A. filliculoides and the emergence of N. lutea could explain the decrease in the amount of remains found from S. natans and A. filliculoides. More represented is the Potamogeton genus (P. trichoides; P. natans; P. crispus) or known as the pondweed, which is an indicator of clear water. The first appearance of Ceratophyllum cf. demersum could potentially be the cause of the slight loss in other submerged plants (E. hydropiper; E. cf. hexandra) because of its invasive nature. Another new appearance is the Epilobium hirsutum, which is a plant that needs a lot of sunlight to be able to grow. Some of the herbaceous and flowering plant species from the ‘Waterside and damp ground´ group disappear (Ranunculus flammula; Solanum dulcamara; Eupatorium cannabinum; Lycopus europaeus), while Lythrum salicaria is well represented. It is possible that because of a higher water level these species disappeared making place for the more resistant L. salicaria. This would explain the only appearance in the assemblage of Rumex maritimus, which frequently grows in flooded areas. Also does this explain the emergence of Butomus umbellatus, a plant that grows along the waterside and up to 3m deep in the water.

o Sample 4 (40-30cm)

A steep decline in the number of fossils found and the number of taxa identified in the samples is observed. In all the zones most of the taxa disappear or have their amounts reduced. The complete disappearance of the Potamogeton genus might indicate the disturbance of the water by a local event of a mudslide, which would have an effect on all plant life. Interesting is the appearance of two species: Cladium mariscus and

Chenopodium rubrum/glaucum. C. mariscus grows along lakesides and in boggy areas on base-rich soil and is resistant to soil pollution. Ch. Rubrum/glaucum grows in damp, nitrogen rich soils in dried-up areas. The connection between the two taxa is the phenomenon of the mudslide which created a habitat for Ch. rubrum/glaucum because the mud dried up and C. mariscus could have sprouted because its advantage of resistance over the other taxa.

o Sample 3 (30-20cm)

In this stage the number of the fossils and taxa is restored and the effects of the phenomenon happening in sample 5 and 6 are not visible anymore. Except for the taxa which stayed (C. mariscus; Ch. rubrum/glaucum). Some species thrived (S. lacustris) while others disappeared (Dryopteris sp.). Newcomers in this assemblage are Urtica

(30)

dioica and Rubus fructicosus. The return of the Potamegeton genus (P. natans; P. cf. distinctus; P. trichoides; Potamogeton sp.) in the assemblage indicates the recovery of the water quality or that the lake is shallow along the banks. The environment is in general comparable to sample 6 with a lake or pond surrounded by a reed swamp where sedges (C. fuscus; C. longus; Carex sp.), grasses (Juncus sp.) and reeds (Typha sp.; S. lacustris) grow together with flowering and herbaceous plants (S. dulcarama; M. cf. aquatica; L. salicaria; E. cannabinum). From the reed plants S. lacustris has the largest growth in amount, it being a pioneering species. There is also a presence of trees indicated by a relatively high amount of C. geophilum.

o Sample 2 (20-10cm)

Not much changes from sample 3 to sample 2 with only some increases or decreases in the number of fossils found for the taxa. New taxa appearing are Potamogeton

coloratus, which needs a specific habitat with nutrient poor shallow water up to 1m and deep peat; Carduus/ Cirsium sp., which thrives in overgrazed or trampled areas; and Linum perenne, which grows in grasslands. In this sample C. geophilum did not appear, but R. fructicosus is still present indicating some sort of woodland vegetation. However, not much can be said about the development of this zone. Aquatic floating plants decrease (N. alba; S. natans) in amount and some even disappear (N. lutea). This could indicate the water getting deeper because these plants have to put their roots in the soil underwater and therefore need shallow water.

o Sample 1 (10-0cm)

There is a steep increase in the number of fossils found throughout all the ecological zones, except for ‘Woodland and shade tolerant’ were there is no presence of R.

fructicosus anymore. Sample 1 is the sample with the highest number of fossils found. In the ‘Grassland, disturbed and open ground’ group L. perenne is still present and

Ranunculus cf. sardous increases in amount, this could be interpreted as a wet sunny open ground or grassland with calcareous soil in the vicinity; Verbena officinalis is a newcomer, and also needs a calcareous soil. In the group ‘Waterside and damp ground’ there is an interesting newcomer. Alnus glutinosa is a tree species which is a pioneering species and it grows at the edges of rivers and swamp in wet grounds. It needs a lot of sunlight to grow and it will disappear once other tree species will make their debut. This tree species is accompanied by the presences of C. geophilum. In the category of reeds, grasses and sedges, the presence of S. lacustris is declining while other species start to

(31)

thrive (C. longus; C. fuscus; Juncus sp.; Typha sp.). Cladium mariscus completely disappears from the picture. Taxa which do well together with sedges, grass and reed plants like M. cf. aquatica and E. cannabinum experienced a high increase in

appearance. Two other newcomers in this group are Urtica kioviensis, which is a plant occurring in wetlands and marshes; and Polygonum lapathifolium, this plant is present in disturbed grounds and in damp soils along rivers and lakes. The latter plant could be an indication for soil slaking. In the ‘Aquatic´ group, floating plants (N. alba; S. natans; A. filliculoides) are thriving. Some of the emergent taxa disappear like Groenlandia densa, Myriophyllum sp., Potamogeton natans, and Potamogeton coloratus, while others re-emerge like the Potamogeton distinctus and Potamogeton trichoides. The disappearance of these taxa could be contributed to the increase of floating plants, especially S. natans and A. filliculoides, which are fast growing floating ferns and block the sunlight from getting into the water and to the plants growing there. A new species found is the Oenanthe aquatica, a plant growing in bodies of water with eutrophic soils and a variable water level. It cannot tolerate salt water.

-

Short Overview

It becomes clear after looking at sample 6 that the environment was dominated by aquatic plants. A fresh water lake was probably present, surrounded by a swampy area. Species which grow in grassland, disturbed, and open ground appear in sample 5, which indicates that some sort of meadow is starting to form in the area. Sample 4 shows a large decrease in fossils and the appearance of species thriving on disturbed ground. This might indicate the occurrence of a mudslide in the area disturbing the local plant life. In sample 3 the amount of fossils has increased with again aquatic plants

dominating. However, the area is less swampy. Not much changes from sample 3 to sample 2. Sample 1 has an increase of fossils and shows signs of pioneering trees.

(32)

Fig. 8. An artist’s interpretation of the flora that could be found in the environment based on the data gathered. The species which are portrayed in the reconstruction are: Mentha cf. aquatica; Potamogeton crispus; Hippuris vulgaris; Urtica dioica; Cladium mariscus; Alnus glutinosa; Rubus fructicosus; Butomus umbelatus; Typha latifolia; Nuphar lutea; Nymphea alba; Azolla filiculoides; Cyperus fuscus; Salvinia natans; Sagittaria sagittifolia.

(33)

Chapter 6: Early hominin migration

Can this site be connected to the migration of early hominins from Africa to Europe in the Palaeolithic, if yes how?

Before the question of whether the existence of this site can be contributed to the migration of early humans from Africa into Europe, it is needed to research the timeline of the appearance of hominins in Palaeolithic Greece. Also important is to understand the different theories behind the migration to Europe. Only then a connection between the presence of hominins and migration can be drawn. The last step is to not only map the possible routes of the early hominins that were present at MAR 1, it is also

important to figure out where the straight-tusked elephant came from.

- Migration patterns

“Out of Africa” is the general name for the phenomena of hominins leaving Africa to go to Europe or Asia via different routes. This is believed to have happened twice: once around 800kya in the Middle Pleistocene, a dispersal of Homo erectus and again around ca. 120-100kya with the movement of Homo sapiens (Derricourt 2005, 121).

Derricourt (2005) clearly summarized the process that went on before and during the “Out of Africa” phenomena and this summary will be based on his with some additions. It is believed that around 1.8Ma the first hominin dispersal took place by Homo ergaster and/or Homo erectus in the direction of North Africa, the Middle East and possibly Asia (Dennell and Roebroeks 2005, 1099). The oldest known evidence of early human occupation was found in Dmanisi in Georgia and dates to 1.85-1.78Ma (Ferring et al. 2011, 10432). This first migration coincides with vegetational changes and migrations by large mammals. A further spread was hypothesized to be around 1.4Ma by the Homo erectus. The oldest hominin remains found in Europe (Sima del Elefante, Atapuerca) belong to the species Homo antecessor and is dated to 1.2Ma (Carbonell et al. 2008, 465). The next major migration by Homo erectus, as I mentioned above, is around 800kya and is called “Out of Africa 1”, with a second movement around 120-100kya by Homo sapiens.

Four different routes have been suggested for any of the migrations from Africa to Eurasia. The first one is a land route via the Sinai Peninsula. The other three involve

(34)

crossing water and are via the Sicilian Channel; the Strait of Gibraltar; or the Bab el-Mandab at the southern Red Sea. The map (Fig. 9) below shows the possible routes.

Fig. 9. Map with the possible routes taken by early hominins from North Africa to Eurasia. Source: Derricourt 2005, 122.

- Earliest appearance of hominins in Greece

Marathousa 1 (Peloponnese) is at the moment the oldest known Palaeolithic open-air site in Greece and even in the Balkan and South-eastern Europe (Tourloukis and Harvati 2018, 48). As mentioned above the site has been dated to 0,48Ma - 0,42Ma, which is corresponding to MIS 12. Other sites in Greece roughly dating to the same time period or close to but a little later, are Kokkinopilus in Epirus (Western Greece) dated to 0,35Ma – 0,25Ma; Petralona cave in Macedonia (Northern Greece) dated to an approximate age of 0,25Ma; Rodia in Thessaly (Eastern Greece) dated to 0,4Ma – 0,35Ma; and Apidima Cave A in Mani (Southern Greece) dated to 0,21Ma – 0,17Ma (Harvati et al. 2009, 132). (See fig. 10 for the locations.)

As shown on the map below these sites are located all over Greece. Most of them (Marathousa 1; Petralona Cave; Apidima Cave A) have remains from early hominins. The presence of these remains provide evidence for the fact that hominins have been in these areas. However, no hominin remains have been recovered dated to the early Middle Pleistocene and therefore there are no remains old enough to be from early hominins taking part in the migration movement from Africa into Europe. Only the molar found at Marathousa in 1960 could possibly be old enough, but there is no certainty about its identification further than Homo sp. (Harvati et al. 2009, 135).

(35)

Late Middle Pleistocene sites bear plenty hominin remains. The oldest two sites with their findings will be listed below. At Petralona Cave a cranium was found believed to belong to the H. heidelbergensis with facial features starting to resemble the

Neanderthal (Harvati et al. 2009, 136). The Apidima caves is a network of four caves. From Apidima Cave A two hominin crania (Apidima 1 and 2) have been recovered dated to 210kya (Apidima 1) and 170kya (Apidima 2). In the other three caves only lithics have been found. With both crania a comparative analysis was conducted. The results of this analysis were that Apidima 2 belongs to H. heidelbergensis with the beginning

resemblances of the Neanderthal. And in contrast, Apidima 1 resembles H. sapiens (Harvati et al. 2019, 500-503). These remains of H. sapiens in Greece could be the oldest known evidence of the modern human in Europe, indicating that early modern humans could have migrated out of Africa earlier than thought (Harvati et al. 2019, 504).

Fig. 10. Important Pleistocene sites in Greece. Map A is Middle Pleistocene/ Lower Palaeolithic sites. Map B is Late Pleistocene/ Middle Palaeolithic sites. Source: Harvati et al. 2009, 133.

(36)

According to Tourloukis and Karkanas (2012) most of the Aegean sea was partially dried up during the MIS 12 resulting in the connection of the Greek peninsula with the many island, which are surrounded by water in modern day (See fig. 11). This would create unique ecological zones with lakes and ponds attracting a variety of flora and fauna, also hominins, which could reach these optimal ecologies via the land bridges that appeared when the water level dropped. These zones would act as refugia in periods of climatic stress. This would mean that a large part of the possible Palaeolithic sites in Greece would be eroded away by the water rising when the ice started to melt and fill up the Aegean Sea again. This is a good explanation for why there are so little Palaeolithic sites found in Greece (Tourloukis and Karkanas 2012, 13). This of course is complemented by the fact that a lot of lithics that have been found throughout Greece have not been studied (Harvati et al. 2009).

Fig. 11. Reconstructions of the Aegean Sea in different ice ages. Source: Tourloukis and Karkanas 2012, 10.

(37)

Tourloukis and Karkanas (2012) further discuss the possibility of a different migration route with Western Eurasia as a central point of dispersal from South-west Asia (See fig. 12, redline) by species like Home heidelbergensis, as a specimen was found from in Petralona. Another new possible route could have been via the coast of modern Turkey over the land bridges through the Aegean Sea (See fig. 12, blackline). Although

Tourloukis and Karkanas do stress that these new routes are assumptions based on scarce evidence which have to be researched further.

Fig. 12. Map of Eurasia with the possible new migration route of early hominins to Europe. Source: Tourloukis and Karkanas 2012, 12.

- Palaeoloxodon antiquus distribution

The Palaeoloxodon antiquus belongs in the Biharian mammal age during the early and middle Pleistocene. It was commonly found during the Holsteinian (MIS 11) and Eemian (MIS 5). In the later Eemian and early Weichselian (MIS 4) it became extinct (Pushinka 2007, 226). The distribution of this species is mainly centred in Europe and dispersed in a lower density towards West Asia and the Middle East (more specifically the Levant) (Pushinka 2007, 228). It can be observed that the elephant does is not come up in Greece. However, the research done at Marathousa adds the presence of P. antiquus in Greece during the Middle Pleistocene.

(38)

Chapter 7: Discussion

- Traces of human activity

From the evidence for butchering at the site, in the form of cut marks on some of the bones, and the possible evidence of knapping of local materials resulting in lithic and bone tools, it can safely be said that there was indeed hominin activity. This would include the exploitation of several large mammal species present locally during the Middle Pleistocene (470-420kya) and according to research done by Barbier (2019) also possible exploitation of the vegetation. This could be the most important reason for the hominin activity at this site, along with the presence of water which is an important resource for the hominins as well as for the fauna and flora.

With the evidence for carnivores at the site in the form of faunal remains and gnawing marks in some of the bones the question arises whether there is a competition to get to the food between the hominins and the carnivore species. This also brings up the question whether the hominins are the hunters in this situation or the scavengers. According to the cutmarks and processing indications on the bones of the elephant and the absence of gnawing marks, it is probably the latter.

- Environment

Throughout the samples not much change is observed regarding the local environment before, at the moment of and after the death of the elephant. Except of course for the event happening in sample four because of which a large part of the taxa disappears. This could be caused by environmental influences before or after the deposition of this layer. A mudslide would be the most logic with indication of disturbance of the clarity of the water; second would be trampling or over grazing by the large mammals present at the site seen in the bone assemblage, which could disturb the ground heavily and make place for other plants found in sample 4. If I would have to discuss a change it would be the fact that the aquatic taxa, as well as in fossils as in types, slowly increases

throughout the assemblages, and waterside and damp ground taxa is second on the list. This would indicate that the area gets wetter, which could be the cause of the rise of temperature. This would be a logical cause seeing through this assemblage we are going from the Esterian glacial into the Holsteinian interglacial. As we already see a lot of taxa which prefer temperate climates. The appearance of Alnus in sample 1 could be a forecast for the increase in tree species, turning this area in a temperate forest.

(39)

When comparing the data from this research to the data in the article from Field et al. (2018), it can be said that the findings are rather similar. However, the event from sample 4 is not seen in the data portrayed by Field et al., meaning that this event was local or it might have been error in the data gathering causing some of the taxa not appearing in the sample.

- Migration

The possibility of the two new found migration routes (Tourloukis and Karkanas 2012) from Africa via the coast of Turkey, through the Aegean sea into Europe; and via South-west Asia into Western Eurasia make it more likely that sites like Marathousa, the Petralona Cave, and the Apidima Caves have been visited by different species of early hominins migrating through this area. When applying the map of the distribution of the P. antiquus over the map made by Tourloukis and Karkanas (2012), it is possible to see a connection between the two patterns. From the levant early hominins could have followed the elephants all the way to Europe. It has been established by Dennell and Roebroeks (2005) that during the earlier migrations (at 1.8Ma or 800kya) Homo species went to Asia, and therefore the same can be said about the early hominins migrating from South-west Asia into the direction of Europe. Of course, these are just sketches of possible scenarios.

- Limitations of the research

The P.aG.e project has little limitations in general because of its multidisciplinary nature. All the different aspects of archaeology can be combined and used to gather

information. For botanical research this means that these other aspects can be used to create a more detailed picture of the ancient environment. Using pollen data can

broaden the range of the reconstruction. Where macro botanical remains can be utilized to recreate the local environment, micro botanical remains, like pollen, can be used to sketch a picture of the wider environment or ecology.

However, improvements can still be made to the botanical research with the goal of gathering more data and with that create a more detailed reconstruction of the environment. These improvements can be found in the sampling for example. By changing the sample intervals (the distance between the spots where the samples are being taken) it is possible to get more data points through time. As in the change of the ancient environment will be portrayed more gradual in the data. By changing the

(40)

sampling resolution (the amount of samples taken per interval) a more complete and detailed picture can be extracted from the gathered data. Which will enhance the reconstruction at one point in time.

(41)

Chapter 8: Conclusions

This research had the main goal to reconstruct the change in the local environment of the Marathousa 1 site in Greece during the Middle Pleistocene (470kya-420kya) from before the Paleoloxodon antiquus died, at the time of the deposition of the remains, to after death. This was paired with the main research question: Why was there human activity happening during the Middle Pleistocene at the site of Marathousa 1 in Greece? To be able to answer this question I explored multiple aspects from this site. The macro botanical (carpological) data that was gathered through collecting the 6 samples from Marathousa 1, Megalopolis, Greece in 2017 (315 area A 938/672) was partially

processed, researched and interpreted by the author himself. The rest of the data was researched through literature.

The archaeological evidence found at the site was in the form of lithic assemblages which could have been made from local sources consisting of debitage and tools, bone tools and fragments, and bone with indication of butchering in the form of cut and scraping marks and evidence of bone processing. All these things could be interpreted together as this site being a kill and/or butchering site. Although, it is not entirely clear whether the hominins hunted the fauna, focussing on the elephant, or scavenged the carcasses after predators hunted the fauna.

The area would look like a lake shore environment with a reed swamp close by and dispersed trees in the vicinity. The taxa point towards a wet marshy environment around a lake with clear water. Floating plants would cover the surface of the lake while concentrations of reeds would grow along its shore, together with low creeping and tall standing flowering and herbaceous plants. The water is likely nutritious and non-acidic. It represents the change into a warmer period. The presence of water is the main reason for the diversity and amount of flora and fauna in the area. Also, the hominins have probably been attracted by this important resource.

The early hominins which were at this site could have been part of the migration from either Africa or Asia. This migration likely took place over many generations and this site gives us only one point in time. There is a possibility that these hominins were following a herd of Paleoloxodon antiquus from either the Levant after they went via Sinai

peninsula over the land route towards Europe via the land bridges through the Aegean sea, or from South-west Asia through Western Eurasia in the direction of Europe.

(42)

However, it is almost impossible to say whether they went further into Europe or went back to where they came from after the hominins finished their activities at

Marathousa.

The answers to these sub questions formulate the answer to the main research quiet well, in my opinion. The hominins lived in a temperate climate in an area with a wide variety of vegetation and animals to exploit. They probably came here hunting a following an elephant either from South-western Asia or via Aegean Sea, where other sites like Marathousa might have been if the water level was not that high in modern days. Whether the early hominins went back home taking with their kill or pushed on further into the unknown looking for a new place to live. A question for further research, or rather a different angle on this research could be: What if these people came from Europe and were going in the other direction? More research and excavations in the Megalopolis basin could provide a better insight into this matter and maybe more sites like Marathousa 1 could be discovered in the area.

(43)

Bibliography

- Internet pages

www.iucnredlist.org/species/167816/6388238, last visited on 20/02/2020. www.wikipedia.nl, last visited on 20/2/2020.

- Literature

Barbier, S., 2019. The influence of climate and environment on early hominins at the Middle Pleistocene site of ‘Marathousa 1’. Leiden (unpublished MSc thesis, University Leiden).

Blamey, M. and Ch. Grey-Wilson, 1989. The Illustrated Flora of Britain and North Europe. London: Hodder and Stoughton.

Cappers, R.T.J., R.M. Bekker and J.E.A. Jans, 2006. Digital Seed Atlas of the Netherlands. Eelde: Barkhuis publishing (Groningen Archaeological Studies 4).

Carbonell, E., J.M. Bermúdez de Castro, J.M. Parés, A. Pérez-González, G. Cuenca-Bescós, A. Ollé, M. Mosquera, R. Huguet, J. van der Made, A. Rosas, R. Sala, J. Vallverdú, N. García, D.E. Granger, M. Martinón-Torres, X.P. Rodríguez, G.M. Stock, J.M. Vergès, E. Allué, F. Burjachs, I. Cáceres, A. Canals, A. Benito, C. Díez, M. Lozano, A. Mateos, M. Navazo, J. Rodríguez, J. Rosell and J.L. Arsuaga, 2008. The first hominin of Europe. Nature 452, 465-470.

Dennell R. and W. Roebroeks, 2005. An Asian perspective on early human dispersal from Africa. Nature 438, 1099-1104.

Derricourt, R., 2005. Getting “Out of Africa”: Sea Crossings, Land Crossings and Culture in the Hominin Migrations. Journal of World Prehistory 19, 119-132.

Doukas, C., T. van Kolfschoten, K. Papayianni, E. Panagopoulou and K. Harvati, 2018. The small mammal fauna from the palaeolithic site Marathousa 1 (Greece). Quarternary International 497, 95-107.

Ferring, R., O. Oms, J. Agusti, F. Berna, M. Nioradze, T. Shelia, M. Tappen, A. Vekua, D. Zhvania and D. Lordkipanidze, 2011. Earliest human occupations at Dmanisi (Georgian Caucasus) dated to 1.85-1.78Ma. National Academy of Sciences 108, 10432-10436.

(44)

Field, M.H., M. Ntinou, G. Tsartsidou, D. van Berge Henegouwen, J. Risberg, V. Tourloukis, N. Thompson, P. Karkanas, E. Panagopoulou and K. Harvati, 2018. A palaeoenvironmental reconstruction (based on palaeobotanical data and diatoms) of the Middle Pleistocene elephant (Palaeoloxodon antiquus) butchery site at Marathousa, Megalopolis, Greece. Quarternary International 497, 108-122.

Giusti, D., V. Tourloukis, G.E. Konidaris, N. Thompson, P. Karkanas, E. Panagopoulou and K. Harvati, 2018. Beyond maps: Patterns of formation processes at the Middle

Pleistocene open-air site of Marathousa 1, Megalopolis basin, Greece. Quarternary International 497, 137-153.

Harvati, K., E. Panagopoulou and C. Runnels, 2009. The Paleoanthropology of Greece. Evolutionary Anthropology 18, 131-143.

Harvati, K., C. Röding, A.M. Bosman, F.A. Karakostis, R. Grün, C. Stringer, P. Karkanas, N.C. Thompson, V. Koutoulidis, L.A. Mouloupos, V.G. Gorgoulis and M. Kouloukoussa, 2019. Apidima Cave fossils provide earliest evidence of Homo sapiens in Eurasia. Nature 571, 500-504.

Harvati, K. and V. Tourloukis, 2013. Human evolution in the Southern Balkan. Evolutionary Anthropology 22, 43-45.

Jacobs, Z., B. Li, P. Karkanas, V. Tourloukis, N. Thompson, E. Panagopoulou and K. Harvati, 2018. Optical dating of K-feldspar grains from Middle Pleistocene lacustrine sediment at Marathousa 1 (Greece). Quaternary International 497, 170-177.

Karkanas, P., V. Tourloukis, N. Thompson, D. Giusti, E. Panagopoulou and K. Harvati, 2018. Sedimentology and micromorphology of the Lower Palaeolithic lakeshore site Marathousa 1, Megalopolis basin Greece. Quarternary International 497, 123-136. Keble Martin, W., 1972. The Concise British Flora in Colour. London: Book Club Associates.

Konidaris, G.E., A. Athanassiou, V. Tourloukis, N. Thompson, D. Giusti, E. Panagopolou and K. Harvati, 2018. The skeleton of a straight-tusked elephant (Plalaeoloxodon antiquus) and other large mammals from the Middle Pleistocene butchering locality Marathousa 1 (Megalopolis Basin, Greece): preliminary results. Quarternary