Was Atlantis a Bronze Age Metropolis in Northafrica?

U. HofmannIndependant Researcher, Germany

ABSTRACT

No doubt about Plato's intension: The Atlantistale was placed to illustrate the 'Ideal State'developed in his earlier work 'Politea'.Repeatedly that fact was used to claim theAtlantis tale was pure fiction. But neitherPrehistoric Athens nor the description ofAtlantis show sufficient correspondence withthe Ideal State. Plato admits the Atlantis talewas unperfect but emphasizes the tale's merit:its authenticity. When Plato wrote down'Critias' almost any greek might have visitedegypt to prove the story either to be right orwrong. None of the critics seems to have takenthat point into account. An often neglecteddetail is, Plato describes Atlantis to have ruledLibya as far as egypt. That means Atlantismust have been located west of Libya andLibya itself must already have been part of theAtlantean territory. That coincides with astatement of Herodotus who tells about alibyan people called 'Atlantioi' living far in thewest of Northafrica. Also greek mythologyplaces Atlas near Lake Tritonis in westernNorthafrica. Plato describes Atlantis as a hugeisland with a large central plain everywheresurrounded by high mountains. Thatdescription matches very well the Maghreb.The Algerian plateau is everywheresurrounded by the chains of the Atlasmountains. The shallow Atlantean sea thatshould have finally vanished resembles thedescription of Lake Tritonis which today isidentified with Chott el-Djerid. Further detailslike the elefants of Atlantis correspond verywell to the west of Northafrica. Infraredsatellite images show a huge sunken

geological structure consisting of severalconcentric circles recently discovered in theeastern part of the Algerian plateau. For 100years already egyptology has knowledge of amighty ancient power that in fact ruledNorthafrica from west as far as egypt: Tehenu.

1. THE QUESTION OF CREDIBILITY

Frequently the argumentation by critics can befound Plato's only intension was to illustratethe 'Ideal State' that he had developed earlierin his previous work 'Politea' and therefore theAtlantis account simply was not more butfiction. — Indeed, the critics are right: In hisdialogues 'Timaeus' and 'Critias' Plato leavesno doubt about that his intension to place theAtlantis tale exactly was to illustrate the IdealState of Politea. In 'Timaeus' Plato letsSocrates say:'I should like, before proceeding further, to tellyou how I feel about the State which we havedescribed [in 'Politea']. I might comparemyself to a person who, on beholding beautifulanimals either created by the painter's art, or,better still, alive but at rest, is seized with adesire of seeing them in motion or engaged insome struggle or conflict to which their formsappear suited;' (Plato,Timaeus).Since that passage is part of the introduction toCritias' summary of the Atlantis tale it is quiteobvious that the Atlantis story in fact was usedfor illustration of the Ideal State. But theimportant question is: Does that really implyor prove the Atlantis account needs to be purefiction? Of course not. There is a number ofgood reasons to believe that Plato's Atlantis

account refers to a true story: First of all, Platoclaims and emphasizes several times theAtlantis account was true. Also Solon's travelto Sais in egypt is verified by Herodotus andothers. We also know the chain oftransmission can have taken place in that wayPlato described (Welliver, 1977). Furthermorethe enormous diversity of details the storypresents, for example the Atlantean flora andfauna, can be valued as indication of a truetransmission. Even the unbelievably highcount of 9,000 years that after Plato shouldhave passed since the time of the Atlanteanwar is not a fact that reveals Plato as a liar.Quite the reverse, that agrees very well to thefact that also Herodotus tells about an age ofegypt of more than 11,000 years (Herodotus,History). Since also the egyptian priestManetho reported such an unbelievably highage of egypt it must be assumed there was asystematic error in the egyptian chronicles atthat time. That is why Plato's '9000 years'should be regarded as indicating authenticityof the Atlantis story. But in addition to thateven the critics should recognize that theAtlantis tale is a hopelessly unperfectillustration of the Ideal State that hardly showscoincidence with the state Plato developed in'Politea'. After reading the presentations of thetwo states Ur-Athens and Atlantis in 'Critias'it does not immediately come clear which ofthese two states was to represent the IdealState. This only is assigned by a single explicitstatement of Critias:'The city and citizens, which you yesterdaydescribed to us in fiction, we will now transferto the world of reality. It shall be the ancientcity of Athens, and we will suppose that thecitizens whom you imagined [in 'Politea'],were our veritable ancestors, of whom thepriest spoke;' (Plato, Timaeus).So, if Ur-Athens was the Ideal State and notAtlantis, we need to ask why then thedescription of Atlantis is about twice as largeas the description of Ur-Athens? And it is alsorather mysterious that in the dialogue 'Critias'the only coincidence of Ur-Athens and theIdeal State which Plato developed in 'Politea'seems to be the seperated class of warriors.Does it make sense to assume the Atlantis talewas a fictive story that was created to illustratethe Ideal State and Plato only forgot to include

and describe the most important characteristic:the philosophers who were to be the head ofPlato's Ideal State? Of course, the answer is'No'. Plato himself was aware of the fact thatthe Atlantis story was unperfect but theimportant merit of the story was itsauthenticity. That is why Plato repeatedlystresses the story was true. That merit was tobe put value on in an appropriate way. Butthere is still another very important reason forwhy Plato's Atlantis account should not beregarded as a fictive story: At that time whenPlato wrote down the dialogues 'Timaeus' and'Critias' almost any greek might have visitedegypt to prove the story either to be right orwrong. So the question is: Why Plato shouldhave taken the risk to be revealed as liar?None of the critics seems to have taken thatpoint into account. And does it make sense tobelieve that Plato wrote down a false andfictive story and then just to continue andworsen his lie he claimed he had got the storyfrom Solon, the probably most famousstatesman of Athens? — What a sinner thenPlato must have been.As a result of these arguments it seems to bemore likely that Plato transmitted a true storyand so scientists and academics are probablyfaced with a real challenge to at long lastreveal the Atlantis mystery.

2. ATTEMPT OF AN OBJECTIVE SEARCH

As a consequence of the preceding discussionand argumentation this chapter now tries tofocus on an unbiased search for Atlantis byregarding Plato's Atlantis account as apredominantly unfalsified historictransmission. The base for that latterassumption however can only be Critias'statement of an existing passed on manuscript.

2.1 Geographical references

Plato describes Atlantis as a huge island,larger as Libya and Asia put together, situatedin front of the 'Pillars of Heracles'. Commonlythis latter expression is regarded as an ancientsynonym for the Straits of Gibraltar. Atlantiswas either surrounded by or at least borderingon an ocean called 'Atlantic'. Obviously thestraight-forward solution then is supposed tobe a location of Atlantis on the other side of

Gibraltar in that ocean which still today iscalled 'Atlantic Ocean'. Most of the publishedtheories about Atlantis therefore locateAtlantis west of Gibraltar. But is thatapparently obvious solution really inaccordance also with the further details ofPlato's description? Geologists have shownthat none of the Atlantic islands situated on theother side of Gibraltar can be remnants of alarge sunken island as what Plato describesAtlantis. Quite the reverse, most of theAtlantic islands are of volcanic origin andshow a continuous uplift (Kukal, 1984). Butthe most important detail that absolutelycontradicts an assumed location of Atlantiswest of Gibraltar is found in the description ofthe ocean that Plato calls 'Atlantic'. In'Timaeus' we read at first:'This power came forth out of the AtlanticOcean, for in those days the Atlantic wasnavigable'and later on the text says:'because there is a shoal of mud in the way;and this was caused by the subsidence of theisland' (Plato, Timaeus).In no way that description matches the oceanthat we call 'Atlantic' today. Nobody wouldcharacterize the Atlantic Ocean as notnavigable or muddy. The 'Atlantic' that Platodescribes, however, must be thought of as anocean that does not even exist any more. Butthere are still further characteristics in Plato'saccount that do not match with the Atlanticthat we know. In 'Timaeus' we find these lines:'for this sea which is within the Straits ofHeracles is only a harbour, having a narrowentrance, but that other is a real sea, and thesurrounding land may be most truly called aboundless continent' (Plato, Timaeus).What sea ever the word 'harbour' might havebeen assigned to, either the Mediterrainean orthe Atlantic Ocean, none of these can beregarded as harbour and neither do the Straitsof Gibraltar represent a narrow entrance. Andwhat does it mean when Plato speaks of 'a realsea' and of a 'surrounding land'? Thatdefinitely describes Plato's Atlantic as aninland sea, surrounded by land. So obviously itseems as if the 'Atlantic' Plato mentions mostlikely was not the sea that we today call'Atlantic Ocean'. What we probably have tolook for is an inland sea of ancient times that

today does not exist anymore. But to continueour search it is necessary first to follow animportant geographical clue that Plato presentsin his dialogues: In 'Timaeus' Critias tells:'furthermore, the men of Atlantis hadsubjected the parts of Libya within thecolumns of Heracles as far as Egypt, and ofEurope as far as Tyrrhenia' (Plato, Timaeus).If Atlantis ruled Europe as far as Tyrrhenia(Southern Italy) and Lybia (the land west ofegypt) as far as Egypt then Atlantis must havebeen situated west of Libya and Libya itselfmust already have been part of the Atlanteanterritory. So, without exaggeration, the landwest of Egypt can be understood as "entrance"to Atlantis and of course that important traceneeds to be followed.

Figure 1: Plato's text not only describes the extension ofthe Atlantean territory but also the direction of ruling:Atlantis ruled Libya as far as Egypt, that means fromWest to East. Atlantis must have been situated west ofLibya.

It is hence of particular interest that also theGreek historian Herodotus, who like Solonhad visited the old priests at Sais in Egypt,reports about a libyan people living far in thewest of Northafrica that was called 'Atlantioi'(Atlantes). He gives a detailed description ofhow to get in a 50-days' journey westwards tothese Atlantes, starting from Thebes in Egypt.He names all the different libyan peoplesalong that way, the first of which was theAmmonians, inhabitants of the oasis of Siwa inthe libyan desert. The next people was theNasamones who after Herodotus were livingsouth of the Syrtis Major in the oasis ofAugila, that still exists today. Further to thewest, after Herodotus, were living theGaramantians. Their capital Garama was

� � � � � � � � �

� � � � � � �

�

�

situated in the Fezzan mountains and stilltoday the village of Germa exists at exactly thesame place (Ruprechtsberger, 1997). Afteranother ten days' journey, so Herodotusreports, there was a people called'Ataranteans'. And finally Herodotus tellsabout a libyan people called 'Atlantes'. Theywere situated close to a huge mountain called'Atlas', regarded by the Atlantes as the 'Pillarof Heaven' (Herodotus, History). As far as tothe Garamantians there is evidence of theexistence and localization of these differentlibyan peoples. As for the last two peoplesHerodotus mentions in his account, however,archeology so far has neither successfully triedto prove their localisation nor their existance.But if we extrapolate the so far verfied journeyby assuming a further stage distance of about500km and a continuing course in parallel tothe coastline then the Ataranteans were to belocalized in about that region where the threecountries of Tunisia, Algeria and Libyaborder. Thus, proceeding another 500km theAtlantes then were supposed to have inhabitedthe Algerian Atlas. Of course thatgeographical localisation would make sensealso because of the coincidence of that name'Atlas' Herodotus mentions and that name'Atlas' what we call these mountains today.

Figure 2: Herodotus describes a journey to the Atlantesthat leads to the farthest west of Northafrica.

2.2 Parallels in Greek Mythology

In addition to that geographical hint whichHerodotus presents further details of interestcan be found in Greek mythology. Also thesemyths assign the farthest west of Northafricaan extraordinary role and show someinteresting parallels to Plato's description ofAtlantis. In one of the most famous mythsHeracles searching for the Golden Apples of

the Hesperides reaches Libya. On his way tothe west he first beats in a battle the libyanking Antaios, a son of Poseidon. Then afterproceeding further to the west he reaches theregion where Atlas was taking the heaven onhis shoulders, close to the gardens with thegolden apples.In another famous myth Perseus reaches Libyaand beheads the Gorgon Medusa. Some strongwinds then carry him westwards into the realmof king Atlas. Atlas there was tending a groveof golden fruits. When Perseus asked Atlas forshelter he was refused, so Perseus took thehead of Medusa and Atlas was instantaneouslyturned to stone and changed into a highmountain.So also in these Greek myths Atlas waslocated in the farthest west of Northafrica(Libya). From these myths we are alsoinformed that these western parts of Libyabelonged to the realm of Poseidoncorresponding to what Plato says aboutAtlantis. And further coincidence exists alsowith respect to the fertile but alsomountaineous landscape. So the assumption isobvious that Plato, Herodotus and the Greekmyths all might refer to the samemountaineous northafrican region that eventoday is called Atlas. The question is: WasPlato's island Atlantis almost identical withwhat we today call 'Maghreb'? Even if theword 'Maghreb' deduces from the Arabianexpression 'Maghreb al-Jazair' which means'Island of the West' it is not an island in thenormal sense of the word. But a few thousandyears ago that situation might have beendifferent. South of the Maghreb once existed alarge forland sea the remains of which are stillvisible. South of the Sahara-Atlas mountainsthere is a chain of huge salt-lakes that are inlarge parts laying below sea-level: Chott el-Fedjadj, Chott el-Djerid and Chott el-Gharsain southern Tunisia and Chott Melrhir inAlgeria. These lakes still become dryer anddryer but even today within the winter monthsextensive rainfall lets the water level rise and alarge shallow inland sea becomes existentagain. In summer these huge lakes are in largeparts covered by a salt-sand-crust. Amongthese salt-lakes Chott el-Djerid with a size of7700km2 is the largest. That is about eighttimes the size of the so-called 'Dead Sea' at

� � � � � �

� � � � � � � � �

� � � � � � � � � � �

� � � � � � � � � �

� � � � � � � � � �

� � � � � � �

Israel's border. The whole chain of salt-lakesextends about 400km eastward from Algeria tothe Mediterrainean sea.

Figure 3: The chain of salt-lakes in southern Tunesiaand Algeria once must have built a huge inland sea.

It should now have become visible for almosteveryone that Plato's description of theAtlantic sea that finally disappeared does notneed to be fictive at all.More than 2,400 years ago the region of thechotts south of the Maghreb still must havebeen a large navigable inland sea that wasaccessible from the Mediterrainean through anarrow entrance. This is verified by an oldnavigation manual, the Periplus of Skylax ofCaryanda, that dates back to the 5th centuaryB.C.. That inland sea was called 'LakeTriton(is)' in ancient times. The Periplus ofSkylax says: 'In this Syrtis there are an island(and a lake), called Triton, and a river Triton,and there is a temple of Athene Tritonis. Thelake has a narrow mouth, and an island isclose to the mouth, and at times, when thewater is low, the lake seems not to have anynavigable inlet. But this lake is large, having acircumference of about one thousand stadia.'(Nordenskiöld, 1897)Also Herodotus reports several times aboutthat Lake Tritonis. Especially interestinginformation is linked with a story thatHerodotus tells about Jason and theArgonauts: 'The voyage was prosperous as faras Malea; but at that point a gale of wind fromthe north came on suddenly, and carried himout of his course to the coast of Libya; where,before he discovered the land, he got amongthe shallows of Lake Tritonis. As he wasturning it in his mind how he should find his

way out, Triton (they say) appeared to him,and offered to show him the channel, andsecure him a safe retreat, if he would give himthe tripod. Jason complying, was shown byTriton the passage through the shallows'(Herodotus, Book IV)The Periplus of Skylax and the story of theArgonauts that Herodotus transmits describelake Tritonis as a shallow but navigable inlandlake with a narrow channel to theMediterrainean sea. I think, this is reallyexactly, what also Plato's Atlantis accountoriginally reported but there seems to havehappened at least one misinterpretation orwrong translation. The relevant lines are thoseof the Timaeus where Critias says: 'for this seawhich is within the Straits of Heracles is onlya harbour, having a narrow entrance, but thatother is a real sea, and the surrounding landmay be most truly called a boundlesscontinent' (Plato, Timaeus). The error is mostlikely as follows: One can be sure the originalexcerpt did not compare two different oceans,a huge Atlantic and a small Mediterrainean, asit is nearly always understood. Instead it onlyexplained as what the entrance of the Atlanticappeared: from within the Straits of Heraclesthe Atlantic simply looked like a harbour witha narrow entrance. But in reality on the otherside of the Straits of Heracles there was not aharbour but a real sea and the surrounding landbuilt a boundless continent. Only then this textmakes sense. One should recognize that thereports transmitted by Herodotus and Skylaxin the same way put special emphasize on thedescription of the narrow channel. That is notonly pure chance. The Atlantic that Platorefered to and Lake Tritonis that wasmentioned by Herodotus, Skylax and otherswere most likely one and the same sea. Thiscan be further supported by informationtransmitted by Diodorus Siculus. He tells usthat Lake Tritonis disappeared in the course ofan earthquake, which again shows a closecorrespondence to what Plato says about thedisappearence of the Atlantic. And Diodorusalso mentions that Heracles set up his pillars inLibya after a successfull battle against theAmazons and Gorgons which were situated

� � � � � � � � � � �

� � � � � � � � � � �

� � � � � � � � � � � �� � � � � � � � � � � � �

� � � � �

���� ����� ����

� � !

� � � � � � � � �

� � � � � � � � � �

close to Lake Tritonis. So the original Pillarsof Heracles were once most likely situated inthat region that today is called Golf of Gabesin southern Tunesia. But if an earthquake letthis inland sea disappear it let of course alsodisappear these original Pillars of Heracles.The recollection of such a straits however,maybe was maintained and so it might havehappened that the Pillars of Heracles lateron'moved' to another entrance to another sea thatwas situated farther to the west.

Figure 4: The supposition: Plato's Atlantis was identicalwith what today is called Maghreb. The Atlantic was aninland sea the remnants of which are the Chotts. Theoriginal Pillars of Heracles were situated at the entranceto Lake Tritonis (the Atlantic).

As figure 4 tries to vizualise the inland sea inancient times must have been much larger thanthe chotts extend today. Due to that inland sealarge parts of the Maghreb must have got thecharacter of an island (pensinsula). So it doesnot seem unlikely that Plato's island Atlantiswas identical with what we today callMaghreb.

2.3 Is there further support of this hypothesis?

As a very special consequence of the justdescribed geography of the Atlantic and theisland of Atlantis it becomes clear that thestraits that was called Pillars of Heracles toone half must have belonged to Atlantis and tothe other half was part of Libya. We must askwhether this is consistent with Plato's Atlantistale? There is a part of the Critias that shows itis consistent: 'To his twin brother, who wasborn after him, and obtained as his lot theextremity of the island towards the Pillars ofHeracles,' (Plato, Critias). The Pillars of

Heracles were a part of Atlantis and not to bethought of as a straits far apart.What the dialogue Critias tells about the floraand fauna of Atlantis is also verified to becorrect even in roman times. The elephants forexample are an important hint in the Atlantistale. That animal was still existent in theMaghreb in the second centuary B.C. asancient numidian coins show.

Figure 5: Numidian coins from the 2nd centuary B.C.indicate that elephants were still existent in the Maghreb

Plato describes Atlantis as a huge lofty islandwith a large central nearly rectangular plainthat had a length of 3000 stadions, which isabout 560km, and a width of 2000 stadions,which is about 370km. That plain he reportswas to all sides surrounded by high mountains.That is a description which largely fits to thecentral plateau of the Maghreb. In the norththere are the chains of the Tell-Atlasmountains with a heigth of up to 1800m. In theWest there are the foothills of the MorrocanHigh-Atlas rising up to 2600m. South of theAlgerian plateau there are the chains of theSahara-Atlas mountains with a height up to2200m and to the east this plateau is borderedby the Aures mountains. The lenght of theAlgerian plateau is actually about 560km. Thewidth however is only about 120km which isless than half the width that Plato reports.

Figure 6: The Algerian high-plain is to all sidessurrounded by chains of the Atlas mountains

" � � � � �

� � � � � � � �� � � � � � � � �� � � � � � � � � �

� � � � � � � � � � � � � � �

� � � � �

2.4 Where to search the Atlantean Metropolis?

Plato tells the Metropolis of Atlantis waslocated at such part of the central plain whereall the water currents and artificial channelsmet, naturally that part of the plateau with thelowest elevation:'It [the ditch] received the streams which camedown from the mountains, and winding roundthe plain and meeting at the city, was there letoff into the sea' (Plato, Critias).The Algerian plateau has an elevation of 400min its eastern parts and rises up to a height of1000m in the far west. All the precipitationthat comes down from the slopes of the hills iscollected in seasonal large rivers at the fringesof the plateau. These rivers bring the waterseastwards and empty into the huge Hodna-basin, the central part of which constitutes anearly 100km long saltlake, the chott el-Hodna. In case the already describedidentification of Atlantis and its central plainwas correct then the Metropolis could onlyhave been located in the Hodna-basin in theeastern part of the Algerian plateau. In the firstmoment that encircled localisation appears asa contradiction to the fact that Plato states theditch emptied into the sea. The Hodna-basin atleast is about 400m above sea-level. Howcould that match? Plato however alsodescribes the whole land as lofty, and hespeaks of the plain and mountains thatdescended towards the sea. That indicates itcould have been an elevated plainconsiderably above sea-level. When theMetropolis was located at close distance to thesea then it might have been a large inland seaon the plateau, within the Hodna-basin.

Figure 7: The circular concentric geometry of themetropolis as Plato reports

As Plato tells the Metropolis was anarrangement of several concentric circularzones of land. These were separated bycircular zones of water (see fig. 7). In thecentre of these zones was an island with adiameter of about 1km.Since Plato tells it was the god Poseidon whocreated that ringstructure it seems likely that itwas originated from natural geologicalprocesses, probably tectonic elevationfollowed by erosion. An example for such acircular concentric zone structure is the so-called 'bull's eye' at Richat in Mauritania.As was explained above in detail Plato'sdescription of Atlantis matches very well whatwe today call Maghreb. Corresponding withPlato's account the central part of the Maghrebconstitutes a large plateau that is to all sidessurrounded by mountains. Now followingPlato's further topographic description theMetropolis has to be searched in the easternpart of the Algerian plateau. There the plateaudescends towards the region of the chotts,which were supposed to be the Atlantic inancient times.

Figure 8: A huge circular concentric structurediscovered on a AVHRR-satellite image (courtesy ofDLR)

Actually, a blurred concentric circulargeological structure was recently discoveredwithin the targeted area (the Hodna-basin) ona low resolution NOAA-AVHRR-Satelliteimage (fig.8). Because of the poor spatialresolution of that AVHRR-sensor of not betterthan 1km no further interpretation of thatstructure was possible. That's why amultispectral Landsat 5TM scene was used fora more detailed investigation (Hofmann,2004). The sensor of the Landsat5TM satellitehas a spatial resolution of 28.5m and provides

image data in 7 different spectral channels,ranging from the visible spectrum via the nearinfrared to thermal infrared.Figure 9 shows the relevant part of the Landsatscene in a nearly true-colour image (spectralchannels 1,2,3). The circular structure is onlypoorly visible in that spectral range.

Figure 9: Part of the Landsat 5TM scene (channels1,2,3). The picture shows a nearly 60km wide aspect.Large water currents coming from west and north emptyinto the chott el-Hodna (the white area). The river Ksobfollows in two parallel paths a circular course into theChott. At the upper part of this image from left to rightthere can be seen some broad white, black and redcoloured bands representing different geological layers(limestone, marl and sandstone). Plato mentiones thatquarries of such coloured stones were in the immediatevicinity to the city.

Having a closer look onto the surroundingmountains one can see that in the past largeand numerous avalanches of mud must havecome down the slopes (fig. 10).Using the near infrared spectral channels(4,5,7) of the Landsat 5TM data set it ispossible to have a look through the salt crustof chott el-Hodna. The circular course of thecovered geological structure seems to becontinued under water. It is supposed there is a

huge concentric ring structure laying underdeep alluvial deposits. In order to get moreinformation on it the thermal infrared band canalso be used. However, the spatial resolutionof the thermal infrared channel is only 60m. Incombination with the near infrared bands animage classification lead to an interestingresult (see fig.12). Several more or lessconcentric circles become clearly visible andsupport the supposition of a huge ringstructure laying under thick alluvial deposits.

Figure 10: Avalanches of mud of enormous dimensionhave come down the slopes of the Hodna mountains.The fine black horizontal line represents a road.

Figure 11: Enormous alluvial deposits seem to cover ahuge circular concentric geological structure. The nearinfrared bands (4,5,7) of the Landsat scene enable alook through the salt crust of chott el-Hodna (part atright bottom). It becomes visible that also under the saltcrust the circular course of the geological structurecontinues.

M'Sila

Chott el-Hodna

Ksob

Chott el-Hodna

Figure 12: A concentric ringstructure becomes visiblewhen including the thermal channel of the Landsatdataset into the classification.

Figure 13: The relevant structures of several differentdata set evaluations were all included and intensified(hand-drawn) in this image.

This ringstructure consists of circles withdiameters of 4km, 14km, 22km and 34km.Of course these results do not prove theexistence of Atlantis and its metropolis butthey can be used as starting point of a moreintensive and detailed study based on remotesensing, geological and archeologicalinvestigations. The Hodna-basin belongs to an

area of active seismicity (Harbi, 1999). On 21August 1856 an earthquake destroyed Jijelitown and its surroundings. Severalearthquakes in the past also had its epicentreswithin the Hodna basin. When Plato reportsthe whole island of Atlantis sunk within oneday and one night into the sea it is impossibleto think of an island of continental size. Butthe Atlantis tale mentions two islands, the oneof continental size and the Metropolis ofAtlantis. It is possible that an earthquake and aflood distroyed the Metropolis and let it sinkinto the sea and maybe that was what the textoriginally recorded. Thus it is to be verified byexperts if the huge concentric ringstructure ofChott el-Hodna that is so extensively coveredby alluvial deposits really covers the sunkenMetropolis of Atlantis or not.

2.5 Who were the Atlantians ?

As far as we know today the Maghreb entershistory not earlier than the 11th centuary BCwith the first Phenician settlements. Thiswestern part of Northafrica becomeshistorically important with the rise of Carthagobut not until the 5th centuary BC. Hence, thereseems to be a huge discrepancy concerningwhat we already know about the Maghreb andthat what is supposed in this paper, i.e.identification of the Maghreb with Atlantis.Despite the fact that early written sources aremissing one should nevertheless try to see ifthere are still any other sources or facts thatcan provide information about the earlyinhabitants of the Maghreb. In 1847 numerousrock engravings showing elephants,rhinoceroses, lions, gazelles, giraffes,ostriches, etc. were discovered in the easternparts of the Sahara-Atlas mountains. Sincethere was also engravings showing alreadyextinct animals like the classic buffalo(bubalus antiquus), one could conclude thesepictures had to be at least eight thousand yearsold (Kuper). Other engravings of similar agerepeatedly represented a ram that wasdecorated with a disk on his head and a collararound his neck. The collar can be valued asindication of early domestication of animals inthe Atlas mountains. But even moreinteresting was the striking resemblance of theram decorated by a disk with the well-known

Amon-ram of ancient egypt that wears thesundisk. Scholars therefore tried to explain thesundisk of the Atlas-ram as being copied fromEgypt. But that was impossible since the rockengravings representing the sundisk-ram wereseveral thousand years older than the EgyptianAmon-ram. So it must have been quite thereverse situation. The worship of the sundisk-ram originated in the Maghreb and expandedeastwards and finally reached the Nile valley(Resch, 1967).

Figure 14: Striking resemblance of the Egyptian Amon-ram (left) and the sundisk-ram of the Atlas (right).

Numerous coloured rock paintings discoveredalso in the southern Algerian mountains, theTassili and the Hoggar, showed that theSahara some thousand years ago was a greenand fertile country. People inhabiting westernNorthafrica at that time were of light skincolour and had European features. These werethe ancestors of that people of Northafricawhich we today call 'Berbers'. The Berbers arethe original inhabitants of the Maghreb.Especially in the northern parts of Algeria andMorocco these Berbers frequently have blondor red hair, light skin and European features.The northafricain rock paintings give reportabout many different epochs: The first one is aperiod when people were hunters. Then in thenext period first domestication became visible.In the following epoch of rock paintingspeople were depicted together with large herdsof goats and cattle. Again one period later rockpaintings report of the time when the horsewas domesticated. Paintings of this period areof extraordinary importance for the focus ofthis paper. More than 300 paintings andengravings of chariots have been discouveredin western Northafrica. Most of them show thecar pulled by two horses in flying galopp. Tilltoday no scholar is able to explain thesenumerous chariot paintings. And nobody

knows which people was depicted there(Camps,1982). The regions for whichexistence of the chariot is verfied do not at allcorrespond to the far distribution of paintingsrepresenting the chariot. Fact is, thesepaintings od the chariot cover an area of morethan 2000km in north-south extension andmore than 2500km in its west-to-eastextension (fig. 16). There must have beennumerous chariots and there also must haveexisted a high degree of organisation to enablethe use of an animal that can not adapt todryness. One should imagine, they actuallyreached south as far as the Niger.

Figure 15: A car with two horses in flying galopp,armed with spears (after (Hugot, 1984))

Figure 16: Historically verified existence of horse car(hatched area) versus unexplained distribution of horsecar rock paintings (dots). (After (Camps, 1982))

The geographic distribution of rock paintingsdepicting the chariot shows clearly that thechariot was a western achievement.Traditional thought however, tries to explainthese representations of chariots with thearrival of horse and car in Egypt at thebeginning of the 13th centuary BC. But at thattime the Sahara should already have becomealmost as dry as it is today and thus it does not

# � � � �

� � � � � � �

� � � � � �

" � � � � � � �

� $ � � � � �" � �

� � � � � � �� � � � � �

� � � � � �

seem reasonable to think of such a lateintroduction of an animal that is not adapted todryness. The broad and extensive use of thehorse and chariot as is reflected by the rockpaintings can only have been possible by acontinueous development process that musthave lasted at least many hundred years.Furtheron, the distribution of chariot paintingsitself clearly indicates that horse and chariotdid not arrive from Egypt since such paintingsdepicting the chariot are not found in theeastern parts of the Sahara. But the mostimportant point is, these paintings of thechariots are in perfect correspondence to whatPlato reports in his description of Atlantis:'The leader was required to furnish for the warthe sixth portion of a war-chariot, so as tomake up a total of ten thousand chariots; alsotwo horses and riders for them, and a pair ofchariot-horses without a seat, accompanied bya horseman who could fight on foot carrying asmall shield, and having a charioteer whostood behind the man-at-arms to guide the twohorses;' (Plato, Critias). It is therefore ofparticular interest that the Algerian rockpaintings show chariots with a seat and thosewithout, those with only one person and thosewith up to three, exactly as Plato describes.

2.6 Information from ancient Egyptian sources

Plato describes Atlantis with attributes, that letus classify it as an empire of the Bronze Age.So it does not make sense to stick to the 9000years he mentions. That unbelievable high ageof Egypt appears several times in ancientsources but can only refer to a different way tocount years. Plato also transmits that it was thehigh priest of Sais who used an additionalsource, obviously not the Atlantis-text, toindirectly calculate the time that had to havepassed since then. The 9000 years most likelydid not belong to the original Atlantis account.So one should therefore concentrate the furthersearch mainly onto historic times. Since Platoreports Atlantis had ruled Libya as far asEgypt we must conclude that both peoplesmust have been in contact and this contactshould in some way have become visible inancient egyptian sources. So, if Plato's talewas true then there must have been a mightypower that ruled from west as far as egypt. We

therefore now have to check if the ancientegyptian sources verify this or not.More than 1000 years before the beginnig ofdynastic Egypt the early predynastic culturesin the Nile valley show distinct westerncharacteristics. The deads are buried incontracted position laying on the side. Paintedpottery of the important Naqada-culturerepeatedly shows depictions of people in acharacteristic adorant posture that is identicalto early rock engravings and paintingsdiscovered in Algeria. Several times thesedrawings also show that these people werewearing a phallus sheath which is a typical'libyan' characteristic. The oldest images ofmen wearing the phallus sheath are supposedto be the rock engravings in the Sahara-Atlas(Algeria). Another typical 'libyan'characteristic is also seen on Naqada-pottery:Several times the decorated pottery depictsmen wearing feathers in their hear. The laterEgyptians of dynastic times did neither wearthe phallus sheath nor the feather in their hairnor did they burie their dead in an contractedposition. So these were clearly westerncharacteristics. The people of Naqada were oflight skin colour and many of them seem tohave had red or blond hair. Best example is thenatural mummy of a man of Naqada, called'Ginger' that is exhibited in the BritishMuseum. 'Ginger' has light yellow skin colourand red weavy hair which still today is typicalfor Berbers of the Maghreb. Interestingly, it isnot unlikely that these early predynasticsettlers belonged to an already highlydeveloped culture that may have arrived theNile valley by ships with numerous oars. Rockengravings in the Wadi Hamamat in theeastern Egyptian desert show numerous largeships that can be associated to the Naqada-culture because of the same portrayals ofadorant posture and of men wearing feathersand phallus sheaths. Archeology andegyptology both can not offer any explainationfor these large ships with up to fifty oars. Andunfortunately it does not even seem to be aresearch topic.In very close connection with the begin ofdynastic Egypt an important but unknownpeople appears in the earliest egyptian sources:Tehenu. The so-called Tehenu-palettedescribes Tehenu as an fertile island with large

heards of cattle, donkeys and sheep and inpossession of a characteristic plant: the olivetree. Another palette of the same time, the so-called 'Narmer-palette' shows how the upperegyptian king Narmer defeats the Tehenu andthus gains control over all over egypt. Thatoccurence was of such fundamentalimportance for egypt's existence that itappeared several times again under kingSahure, king Niuserre and king Pepi II.The Tehenu therefore must have been the mostimportant forreign power from the LatePredynastic (ca. 3200 B.C.) on at least till theend of the Old Kingdom (ca. 2134 B.C.). Fromthe Sahure relief it becomes quite clear thatTehenu was a power that was situated in thewest since it was the goddess of the west whohanded over the captured Tehenu king. Thatalso can be concluded from the fact that thedepicted Tehenu people were all wearing thephallus sheath. Tehenu later became asynonym for the whole country laying to thewest of egypt (Hölscher, 1934). So it alreadybecomes more and more clear that Tehenuactually was that important power that ruledLibya as far as egypt. Several hundred yearsthen the Tehenu probably did not play such apredominant role. However, that situationchanged in the 13th centuary B.C.. Anotherlibyan people called the 'Lebu' that probablywas a subgroup of the Tehenu built a coalitiontogether with the Sherden, Shekelesh, Ekwesh,Luka, Teresh and others. The Sherden havebeen associated with Sardinia, the Shekeleshwere associated with Sicilia and the Tureshwere supposed as to be the Tyrrhenians. TheEkwesh were also called the 'people of thesea'. During the reign of the egyptian kingMerenptah (1213 B.C.) that coalition startsinvading egypt from the west but Merenptahfinally beats them. Even the critics should seethe close correspondence to Plato's account ofAtlantis and the war. Plato mentions Libya aswell as Tyrrhenia, and both people wereactually involved in the campaign againstegypt during the reign of king Merenptah. Butthis probably was not more than the begin ofthe war that Plato transmitted. The ancientegyptian sources tell of another campaignagainst egypt during the reign of Ramses III(1187 B.C.). Another coalition of differentlibyan peoples (Temehu, Tehenu, Seped, Lebu,

Meshwesh) and a coalition of the peoples ofthe sea (Peleset, Tjeker, Shekelesh, Danuna,Weshesh) seems to have made the attempt tosubdue all the countries of the easternmediterrainean in a single attack. As a resultof that occurence many bronze agecivilizations were extinct. Among these werethe Myceneans, the Hittite Empire, Troy,Cyprus and Ugarit. As Plato correctlytransmitted the Greeks lost their writing andthe surviving Mediterrainean cultures fell intothe so-called 'dark ages'.

3. CONCLUSIONS

Plato's Atlantis tale is supposed to be true. Thewar he mentions most probably was a wellorganised coalition campaign of the so-calledsea peoples and the peoples of Tehenu at about1200 B.C.. That war lead to the collapse ofnumerous bronze age civilizations. It wasshown that Tehenu was a mighty power thatruled Libya (Northafrica) as far as Egypt.That's also what Plato told about Atlantis.Based on Plato's detailed description it can beconcluded that Atlantis was most likelyidentical with the Maghreb. Satellite imagesshow a huge concentric ringstructure in theAlgerian chott el-Hodna covered under thickalluvial deposits. It will be the task of expertsto verify if this ringstructure covers theMetropolis of Atlantis or not.

REFERENCES

Plato, transl. by Jowett. Timaeus. Oxford.Plato, transl. by Jowett. Critias. Oxford.Welliver, W. (1977). Character, Plot and Thought in

Plato's Timaeus-Critias.Philosophia Antiqua 32:1-6.Herodotus. The HistoryKukal, Z. (1984). Atlantis in the light of modern

research. Prag.Ruprechtsberger, E. (1997). Die Garamanten. Mainz.Nordenskiöld, A. (1897). Periplus:an essay on the early

history of charts and sailing-directions, NorstedtHofmann, U. (2004). Platons Insel Atlantis. NorderstedtHarbi, A. et al. (1999). Neotectonics and associate

seismicity in the Eastern Tellian Atlas. Journal ofSeismology 3: 95 -104

Kuper, R. Sieben Fragen zur Felsbildkunst, in: Sahara-10000Jahre zwischen Weide und Wüste, Köln

Camps,G.(1982), Le cheval et le char dans la prehistorienord-africaine et saharienne, in: Acte du colloque deSenanque,21-22mars 1981, Aix-en-Provence

Hugot, H. (1984), Zehntausend Jahre Sahara, Darmstadt