Boizer_Late glacial and postglacial climatic variation in the Near East.pdf

8/12/2019 Boizer_Late glacial and postglacial climatic variation in the Near East.pdf

http://slidepdf.com/reader/full/boizerlate-glacial-and-postglacial-climatic-variation-in-the-near-eastpdf 1/15

K. W. Butzer: Late glacial an d postglacial climatic variation 21

L A T E G L A C I A L A N D P O S T G L A C I A LC L I M A T I C V A R I A T I O N I N T H E N E A R E A S T

Karl W. Butzer

Mit 5 Bildern

Spät- und nacheisêitlicbe Klimaschwankungen

im Vorderen Orient

Zusammenfassung V er f . u n t e r n i m m t d en V e r s u ch , d i eKl im asch wan kun gen se i t der l e t z t en Ei szei t f ür e inen T ei ld e r s u p t r o p i s ch en B r e i t en , d en V o r d e r en O r i en t , zu r e -k o n s t r u i e r en . F ü r d i e s en R au m l i eg en v e r h ä l t n i s m äß i gbrauchbare Ergebni s se vor , wenn diese auch n icht d i e

Genauigkei t e r r e i chen wie in höheren Brei t en , da d i ePol l enanalyse h ier n i cht anw end bar i s t . Auf Gr un d s t r a t i -g r ap h i s ch e r , m o r p h o l o g ' s ch e r , a r ch äo l o g i s ch e r u n d h i s t o -r i sch e r B e fu n d e k ö n n en f o l g en d e P e r i o d en au s g eg l i ed e r tw e r d en :

1 . Pos tp luv ia l 1 ( anschl i eßen d a n d i e l e t z t e P luvia l ze i t ) .Trockener und e in wenig kühler a l s heute mi t ausge-p r äg t e r W i n d ab t r ag u n g . H ö h ep u n k t w ah r s ch e i n l i ch i m11. Jhr t sd . v . Chr .

2 . Subpluvia l I (wahr scheinl i ch im 9 . Jhr t sd . ) . Z ieml ichfeucht mi t Nieder schlä gen, d i e si ch e twa auf halber H öh ezwischen denen e ines Pluvia l s und denen der Neuzei t be-w eg en . N i ed r i g e r e T em p er a t u r en m i t e i n em l e t z t en V o r -s t o ß d e r k l e i n en W ü r m - K ar g l e t s ch e r d e s H o ch l an d es .

3 . Pos tp luvia l I I a ( ca . 8000— 6500 v . Chr . ) . Wied ert r o ck en e r a l s h eu t e u n d an s ch e i n en d w ä r m er a l s w äh r en d

d e r v o r h e r g eh en d en P h as e .4 . Pos tp lu via l I Ib ( ca . 6500— 5500 b zw. 5000 v . Chr . ) .

Etw as f euchter , ähnl i ch dem j e t z igen K l ima.

5 . Subpluvia l I I ( ca . 5000—2400 v . Chr . ) . Diese zwei t epos tp luvia l e Feuchtzei t i s t i hr er In t ens i t ä t nach e twa demS u b p l u v i a l I zu v e r g le i ch en , d au e r t e ab e r län g e r u n d w a r -zu s am m en m it d e r f o l g en d en P e r i o d e - d u r ch e t w asg r ö ß e r e W ä r m e ( K l i m a o p t i m u m ) g e k e n n z e i c h n et . K u r z -f ri st ig e N i ed e r s ch l ag s v e r m i n d e r u n g en k u r z v o r u n d k u r znach 4000, sowie um 3000 v . Chr . ze i chnen s i ch ab .

6 . Pos tp lu via l I I I ( 2400—850 v . C hr . ) . N ach e inemp l ö t z l i ch en K l i m au m s ch w u n g t r e t en e r n eu t a r i d e V e r -h ä l t n i s s e — m i t äh n l i ch en T em p er a t u r en w i e v o r h e r —auf. Mind es t ens e inmal ( im 12. Jhrhdt . ) l äß t s ich e in feuch-teres Interval l fes ts tel len.

7 . Pos tp luvia l IVa (850 v . Chr .—700 n . Chr . ) . Im großen

und ganzen g le i ch dem heut igen Kl ima, nur e in wenigt r o ck en e r m i t e t w as w ä r m er en W i n t e r n . M er k w ü r d i g i s te ine ausgeprägte Dür re ca . 590—645 n . Chr .

8 . Pos tp luvia l IVb (Sei t 700 n . Chr . ) . Von e twa 650 n .Chr . macht s i ch e ine zunehmende Zahl käl t er er Winterb em e r k b a r , b e s o n d e r s w äh r en d d e r fr ü h i s lam i s ch en Z e i t .Das In t erval l 700—1000 n . Chr . , sowie das späte 13 . ,f rühe 15. und f rühe 17. J ah rhd t . w aren r e l a t iv f eucht .Das 12 . J ah rhd t . e twas t rock ene r . Sei t 1900 i s t e ine Nie-d e r s ch l ag s ab n ah m e v o n v i e l l e i ch t 1 0 — 1 5 % s p ü r b a r , d i eansch einend mi t der " r ec ent c l imat i c f lue tuation" zu-s a m m e n h ä n g t . G e r i n g e ö r t l i c h e T e m p e r a t u r z u n a h m e ns i n d v o r h an d en , d ü r f en ab e r n i ch t fü r d en V o r d e r e n O r i en tvera l lgemeiner t werden. Sei t e twa 1938 deute t s i ch j edoche i n e r n eu t e r U m s ch w u n g zu e tw as f eu ch te r en V e r h ä l t -nissen an.

General Problems associated with the Chronology

Clima tic v ariation as a f ield of serious investi-gation is rapidly assuming a role of fundame ntalimpor tance, and the prodigous am ount of l i tera-ture devoted to the subject in recent years, indi-cates the universal interest shown both by geo-

graphers, meteorologists and geologists. Frequentreferences to supposed climatic changes in theeastern half of the classical world can alread y befound in the Greek and Arabic au thors , and simi-lar speculations have continued into the twentiethcentury. Climatic variation was only consideredas a means to an end, either to substantiate orrefute the deterministic theories abounding at thetime. Conseq uently this fretting speculation bytravellers and other non-specialized observerswas disastrously affected by the introductio n ofnew and less fallible method s of inve stigation innorthern Europe, leading to the tr iump h of theh isto ric al p o s s i b i l i s t e s a nd tho se v eh em en t

adherents of a theory of s tatic climate "since gla-cial t imes". However, l i t t le observed by the cog-nate f ields, a new a nd less circumstan tial techni-que, the intensive géo morphologie s tudy of selecttype areas, was beginning to accumulate a greatnumber of unintegrated and often obscure con-tributions to this theme of climatic variatio n,often only focused upon the plu vials of thePleistocene epoch. Largely neglecting this newerliterature a nd often based only upon that highlydisputable criterion of nomad ic migrations, C. E.P. Brooks (1926) presented a general survey ofNe ar Eastern climatic f luctuations. O nly the moreprofitable althou gh still quite scetchy syntheses of

D. M. A. Bate (1940) and S. A. Huzayyin (1941)mark ed the firs t great s tep to a new inaugurationof palaeoclimatological s tudy in the desert m ar-gins of the Old W orld. Subsequen tly, G. W. Mur -ray (1951) and H. Bobek (1954) have attemptedto give a regional outline of climatic evolution inEgy pt and Persia respectively, but both of theseoutstanding explorers were hampered by the im-possibility of compa ring overall data from thewider base area essential to to such a study, andwere almost obliged to consider the Holocene asan appendage to their valuable t reatment of thelast pluvial-glacial phase.



22 Erdkunde Band XI

Those familiar with the extensive specializedliterature on the postglacial chronology of Europewill probably find the m ethodology employed inthis treatise rather different and probab ly evenless definite and not sufficiently qua ntitativ e, butthis is inavoidable through the difference of thegeographical factors , and thereby of the climaticindicators involved. Although the penetration ofcold air m asses effects cyclogenesis to no incon-siderable degree, and temperature conditions re-gulate the rate of evaporation , the direct morp ho-logical evidence of temperature variatio n in theNe ar East, in cont rast to Euro pe, is very se-condary to that of rainfall f luctuations. Yet theeffects of increased precipitation and decreasedevapo ration are largely identical, and often re-flect an ide ntical p rim ary cause, so tha t one candraw no sharp distinction between these two pro-cesses in the Near East. Undoubtedly the lack ofbotanical evidence is of the greatest hindranceand inconvenience since pollen-analysis, the foun-dation of the Euro pean clim atic succession, isalmost inapplicable under the existing conditionsof oxidation. The preservation of the pollen re-quires constant immersion and therefore a con-stant watertable, something that is very rarelyachieved in the Near East except in the vicinityof the great exotic r ivers (H.Gross). Fur thermore,neutral or alkaline water is fatal for their preser-vation. Since the s tratigraphical profile can neversupp ly the precise chronological datin g of the

pollen profile, resort must be ma de to the newtechnique of radiocarbon dating, an d such datesas are assigned in the climatic succession establish-ed below are largely provided by C14. The lastgreat adv antag e of pollen diagram s comes fromthe relatively good quan titative estimates thatcan be made of existing climatic conditions ac-cording to the movements and relative percentageof well-known , botanical species. Morphologicalprocesses in themselves are continually the sub-ject of s trong controversy, and too lit t le unam ityhas been acieved in regard to the climatic en-vironment associated with them.

In this paper a f irs t attem pt w ill be made toestablish a parallel to the European chronologywithin a section of the northern margins of thegreat Afrasian Steppe, the necessity of m aintain -ing unity of variation confining us to the ancienthistorical lands of the Near East: Egy pt, Syria,Mesopotamia and Persia. This geographical areacomposes the southeastern qua dran t of the Me-diterranean climatic province, with its westerlycirculation and rainy season in winter, its trade-wind, subtropical desert climate during summer.Southwards and eastwards the climate gradesinto desert and eventually into the peripheralsphere of the great monsoo nal provinces, north -

ward s into regions with rainfall m axima duringthe summer or during the transitional seasons. I tis hoped that the detaile d clim atic successionestablished here indep enden tly of, b ut so curious-ly alike, tha t of Europe will s timulate furtherwide and non-ded uctive regional s tudies of H o-locene climatic variation in the horselatitudes, foronly then can we establish that transitional linkvital for a satisfactory reconstruction of the ge-neral atmospheric circulation pa tterns of thePleistocene.

Temperature and Rain fa ll during the Late

Glacial Period

The existence of periods of considerably grea-ter rainfall d uring the Ne ar Eastern Pleis tocene

has been kno wn for m any years , and it was toaccount for these fossil traces of former climatest h a t t he t er m p l u v i a l w a s i nt ro d uc e d as e ar lyas 1889. Th e contem porane ity of these lowermiddle latitude pluvials with the world -wideglaciations has been repeatedly confirmed andaccepted in the contemporary literature by P.Woldstedt, C. Troll , R. F. Fl in t, M. Pfannenst ie l,S. A. Huzayyin, J. Büdel, H. Mensching andma ny others . Consequen tly we are free to referto the last phase of glaciation as the 'Last Pluv ial 'in the Nea r East, and we may speak of a 'post-pluvial ' as well as a 'postglacial ' period. Howeveris the significance of the last two te rms identical?

Chron ologically there appears to be need of adistinc tion. The postglacia l period is usuallyregarded as beginning with the drainage of theBaltic ice-lake abo ut 8000 B. C., wh ile the lastW ürm pluvial subphase had prob ably ceasedseveral thousand years earlier , apparently withthe end of the last glacial max imu m. In fact, i tlooks as if the general characteris tic of the LateGlacial period in the Ne ar East was an aridclimate, with a rainfall c onsiderably less than to-day in view of the prevailing temperatures s til la li t t le below those of the present.

In Egypt, the Nile undercut and degraded itsbed, to 46m below the modern alluvium at Asvutand to at least 30m below sea level in the Delta(Fourtau 1915, Sandford Arkell 1939, p. 75-6),as a response to the glacio-eustatic lowering ofworld sea level during the Würm glacial maxi-mum when large masses of the oceans ' waterswere held in the great continental ice sheets. Asa result the free com munica tion of the river andthe last Nile- controlled Fayum pluvial lake wasinterrup ted in Uppe r Palaeolithic times after theNile level dropped below a certain threshholdvalue, somewhere between 18m and sea levelaccording to Miss Caton-Thompson and MissGardner (1934, p. 14) on the grou nds of pec u-



K. W. utzer: Late glacial an d postglacial climatic variation 23

l iarities of the molluscan fauna and great preci-pitations of lime in the shoals of the Id wa bank.The lake, left to itself , shrank rapidly and almostor completely dried up permitting aeolian agentsto attack the exposed lacustrine sediments sever-ely. This shrinkag e implies a low rainfall sincethe subsequent Neolithic lake built up impressivebeaches during protracted halts under identicalcircumstances of isolation and with somewh athigher temperatures. Further, s ince wind denu-dation of these Neolith ic an d D ynastic beach for-mations of the F ayum has been negligible, i tappears that aeolian deflation was p articula rlypronounced in Upper Palaeolithic times. Simil-arly in Libya, where G.W. Murray (1951) pointsto the s triking absence of surface finds of Up perPalaeolithic and microlithic f lint im plements in

Cyrenaica, C. Rathjens (1928) was able to ascribethe deposition of the Tr ipolitan ian loess to anarid period with more frequent and strongersoutherly winds than are known to-day, some-where between pluvial and historic times.

After the last, weak 'Pluvia l C ' of the JordanValley (pro bab ly to be identified with the lastWürm Pluvial subphase) L. Picard (1929) notesthat the Jordan and its tr ibutaries cut deeply intothe soft Pleistocene sedim ents, leading to the iso-lation of the 'upper terrac e' and aggrada tiongravels of Pluvial 'C ' . The vertical incision pro-ceeded to below the present alluvial plain by atleast 9m, and after the subsequent a ggradation ofthe ' lower terrace' or modern floodplain, degra-dation has only been able to excavate a shallowchannel since Neolithic times. Similarly in theChemchemal P lain of Kurdis tan , H. E. Wright, Jr.(1952) points out that the s ite of Jarm o w asoccupied about 5000 B. C. (earliest occupationallayers radiocarbon dated 4756 B, C. ± 320 byArnold and Libby, 1951) when the postpluvialerosion had already incised the small river valleyto four-fifths its present depth.

The extensive loess deposits on the southernshore of the Ca spian Sea, reaching a depth of10m nor th of As tarabad, offer fur ther subs tan-tiation of early postpluvial a ridity. H. Bobek(1937) believes this deposition of loess would nothave been possible had the present moist Caspianforest then existed here, and he concludes tha t thelower treeline against the steppe must have lainseveral hundred meters higher upon the moun-tain slopes. The origin of th e Aste raba d loess isto be found, at least in part, in the exposure ofthe old sea floor sed iments of the shrinkin gpluvial (Chvalyn) Caspian Sea to wind-attackduring early postpluvial t imes. From the KaraKam ar Cave near Haibak, A fghanis tan , we mayconclude th at the bulk of the loess depositionhad stopp ed before 9000 B. C. but had been

proceeding for at least 25,000 years before thatdate (Coon Ralph 1955). The deflation of the'younger sediments' (or Seelöß = lacustrine loess)of the southern Lut Basin of Iran necessarily im-plies a great fall of the wa ter tab le or w aterlevel of the kavir , as the basis of aerial denud-ation is only then achieved when the newly ex-posed sediments lie close above this water-satur-ated horizon (G. Stratil-Sauer 1952). As theseyounger sediments were deposited subaeriallvand to some extent 'subaqueously' to some 150m(,Stratil-Sauer 1952, A. Gabriel 1942) and thenexcavated to well below the level of the presentNamakzar Kavir , we must suppose a great de-crease in moisture since the last local pluvial anda water table somewhat below that of the pre-sent, prior to the subsequent recreation of the

present kavir or salt pan.Th e persistance of lower temp eratures even

in latitu de 33 ° N until c. 8000 B. C. can bededuced f rom the Ha ua Fteah C ave, Cy renaica ,where a ngular limestone fragments , due to frostweathering of the roof and wa lls of the cave,are most plentiful in radiocarbon dated horizonsof o lder date (C. B. McBurney and others 1953,H. E. Suess 1954). Similarly a last horizon ofangular limestone fragments due to therm oclasticweathering occurs at —1.5m in the cave depositsat Ksar Akil, Lebanon, compared with Würm Iand Würm II archaeologically dated horizonsindicating greater cold and moisture at —15 and—10m respectively (/ . F. Ewing 1951). The lastsuggests the last world-wide glacial readvanceleading to the formation of the Salpausselkä andMankato end-moraines, (dated at 8800—8100B. C. by H. Gross (1954) taking a mean of thegeochronological and radiocarbon dates), andthere is ample indirect testimony of a s imilarphenome non in the mo ntane regions of the Nea rEast. A. v. Reinhard (1925 ) identified three stagesin the retreat of the Wü rm glaciers of the Cau -casus, which he even ventured to designate asBühl (Schiern), Gschnitz and Daun in referenceto the Alpine chronology. More important is thatthese s tadia were characterized by snowline de-pressions of 8— 900, 550—600 and 3—400m .R. Leutelt (1935) also recognized stadial mor-aines in Lazistan which he considered of Dau n,and some possibly of Gschn itz age on accoun t ofsimilar snowline depressions. In Hakari, Bobek(1940) identified at least one distinct early post— Würm maximum s tage with a snowline de-pression of 350m, i . e. half t hat of the W ürm,which he suggested may represent Gschnitz. Theretreat stadia of the Elb urz a re even less distinctand general so that Bobek (1937) can only ten-tatively suggest stages with snowline depressionsof 700j 450—500 , 300, 250, 170 and 100m, com-



24 Erdkunde Band XI

pared w ith a Wü rm depression of 800m. La stly,A. Desio (1934) recognized a stadial in the retreatof the Zarde h Kuh glaciers in Southwest Iranwith a snowline depression of 300m comparedwith 600— 650m for the last glacial max imum .We cann ot conclude too much from these details ,other than that i t appears most probable that theworld w ide temperatu re relapse of the 9th m il-lenium was also characterized by a temporaryreadvance of the already much smaller W ürmglaciers in the highlands of Anato lia and Iran,with a minimum snowline depression of some-where between 100 and 300—400m.

However this last glacial relapse does not seemto have passed unnoticed in the precipitationcurve of the N ea r E ast, just as the terraces at9—10, 7—8 and 3—5m of the eastern Kh argaScarp (Caton -T hom ps on 1952, p. 3—11) seem toindicate a series of minor, tempo rary improve-ments of local moisture conditions subsequent tothe last, pluvial subphase with tufa formation(early Upper Palaeolithic). Firs tly in the Hagfetet Tera Cave, Cyrenaica, a s talagmite horizonoverlying a considerable depth of Up per Palaeo-lithic material indicates a quite humid phaseantedat ing a fa ir ly evolved Cap s ian- type b lade-industry associated with remains of C ER VU S S P.

(C . Petrocchi 1940, p. 8—34). Similarly G.Knetsch (1950) has identified a short phase of'pluviation' during Capsian times in the Djebeles Soda of Libya, where erosion and depositionof angular gravels took place, possibly accom-panied by the local formation of red earths in theCentral Sahara. Thanks to the radiocarbon da-ting of events in Northw est Africa and tele-connection with equatorial varves in East Africa(c. f. H. Alimen 1955, Table XI) , the Capsians e n s u l a t o seems da ted at abou t 7000 to13,000 years ago, so that a 9 t h millenium datewould not be improbable as a f irst estimate.Further, at Heliopolis in Egypt, S. A. Huzayyin(1952b) notes a pre-microlithic, Upper Palaeo-lithic industry with noticeable Neolithic tenden-cies found in a 30cm thick bed of gravels brou ghtdown during some 'semipluvial' phase by localtorrents.

In Palestine a similar interruption of thecharacteris tic early postpluvial a ridity of theNear East can be noted. Miss Bate (1940) pointsto a faunal change in Palestine amo ng specieswith widely differing biological requirements inNatu f ian ' ) t imes , where the d isappearance of ahalf dozen species of gazelles, a hedgehog and aspecies of hyae na suggests the oncom ing of a m ore

' ) The Na tuf ian appe ars to have begun some 12,000years ago according to the lates t excavat ions by / . Perrot,see „Der For tsch r i t t" , Düsseldorf , 14. 6. 1956.

humid phase. In the Jordan Valley, degradationceased at the beginning of the Holocen e (to beconsidered in a broade r sense of the word), andat firs t some 2m of sand were deposited by aswifter r iver able to carry a greater load than atpresent, remark able s ince only the finest mate-rials can be transported to this particular pointto-day (Picard 1929). Perhaps the 80cm horizonof well rounded gravels at the base of the ' lowerterrace' exposed at Eddesiye at the Jarmukestuary (Picard 1932) was also contemporary.Picard mentions a 2m strandline of Lake T ibe-rias , and the author has noted numerous com-paratively recent terraces on the northwesternend of the Dead Sea. Aneroid determ inationsgave approx ima te elevations of 10, 12—15, 20,25, 35, 45 and 50m above the present Dead Sea

level (392m below m.s.l. in 1948) for these mino rabrasional terraces c u t i n t o the seasonal ly-banded, varve-like Lisan sediments possibly con-nected with the older (constructional?) terracesdescribed by G. S. Blake (1928, p. 24—5). It isnot possible to date the younger terraces by cul-tura l associations as it is next to hopeless to findprehistoric sites close to this sterile and forbid-ding body of w ater, yet on comp arison, we sug-gest that the terraces from 10 to 50m are m orpho -logically as fresh and as well-pre served as theNeo lithic terraces of the Fay um , so tha t a Ho lo-cene date does not seem improbable.

Lastly the Ghar i-Khamarband (Belt) andHo tu C aves near Ashraf on the Caspian coastof Iran offer dating possibilities for the tempo -rary humid interval noted earlier for Libya, Egyp tand Palestine. In the Belt Cave (Coon 1951) anearly M esolithic layer of cave earths is separatedfrom the overlying horizons by a 30cm layer ofloess, which we believe represents one of theoscillations to an arid climate recorded in theupper 3— 4m of the Mazand eran loess by 3—4alternatin g layers of light-coloured loess anddar k,cl aye y transformed loess beds (Bobek 1937).The early ( 'Seal ' ) Mesolithic has been radiocar-bon dated at 9525 B. C. ±550 by Ralph (1955)and at 6150 B. C. ± 900 by Arnold Libby

(1951). Further the overlying loess beds gave adate of 10,320 B. C. ± 825 (Ralph 1955) so thatthere appears to have been some contamination,alteration or redeposition here. The nature of theSeal Mesolithic deposits suggests a moister cli-mate which i« substantiated by the occurrence ofa small species of seal, likewise found in a horizondated at 9905 B. C. ± 840 in the Hotu Cave,which suggests the close pro xim ity of the Casp ianshore. Th e base of the entrance to the Belt Cavebeing 15.4m above the 1951 level of the Cas pian ,it seems quite likely that the post-Chvalyn10—12m 'Dagestan Stadium' shoreline identified




Abb. 1: Postpluvial Dead Sea terraces (with embankmento f Jeri cho road in l ef t f oreground)

Abb. 2: Postpluvial incision of the Jordan: 'upper terrace'scarp (55 m) wi th ' lower terrace' alluvium and river

meander in foregroundAbb. 3: Upper Pleistocene varve-like annual layers ofgypsum or calcium carbonate alternating with silt or clay

Photos by

(1,5 m vertical section: the seasonally-banded layersaverage 2—3 mm in thickness)

Abb. 4: 2 m raised beach of the Medi terranean probablydating f rom 1—2000 years B .C. (between S id i abd e l

Rahman and El Alamein)Abb. 5: Shorel ine of the Neol ithic 'B' Fayurn lake at 4 m

dating f rom c . 4000 B .C. (North o f Kom Aushim)

: au thor .



26 Erdkunde Band XI

by Leontyev Fedorov (1953) was contempor-ary. Similarly a brown cave earth layer in the

Kara K am ar cave, dated at 8625 B. C. ± 300(Coon Ralph 1955) suggests a moister periodin comparison to the previous loess deposition.

From the above evidence one may, for theNear East, deduce a temporary moist phase,which indirectly seems to be associated with thelast glacial relapse of the 9 t h millenium. It mustbe emphasized however that all postpluvialhumid fluctuations cann ot be comp ared w ith thepluvials proper, being relatively insignificantgeologically. Their importance lies rather in theecological environment they provided during therapid dev elopme nt of the earlier Ne ar Easterncultures at this time. From all indications webelieve an identification with the M akalian phaseof East Africa may one day be possible, but p end-ing more conclusive evidence to the contempo-raneity of ' tropical ' and 'subtrop ical ' pluvials weprefer to adopt a quite indepen dent termino-logy here, not limited by regional or culturaldesignations. Consequently we shall simply andarbitrarily call those periods with a rainfallsimilar to or less than tha t of to-day by the term'Postplu vial ' phase, with a suffixed number forpurposes of identif ication. Similarly the moisterintervals shall be designated 'SubpluviaP phases,so that the original arid period from perhaps16,000 or at latest 13,000 B.C. 'Subpluvial I ' ,and the succeeding drier phase, 'Postpluvial II '

Climatic Variation during the Early Holocene

(until 2400 B . C.)

T h e P o s t p l u v i a l I I p h a s e . T hemoist phase described above and designated as'Subpluvial I ' only represented a temporaryamelioration of moisture conditions, a prolongedinterruption of the general trend of earlier post-pluvial aridity which from all indications seemsto have continued at least a millenium or twoafter the close of the Subpluvial I .

Throu ghout the Near East a lack of evidence

to the contrary suggests that temperatures roseto their present values during the 8 t h and 7 t h

millenia, while the waters of the world oceansrose rapidly in the so-called 'Flandrian Trans-gression' as the continental ice masses rapidlydisintegrated during the Preboreal and earlyBoreal. In response to this rising sea level theNile will necessarily have begun to aggrade itsbed, raising its level till it overflowed across theHawara Canal into the Fayum a second time,giving rise to a new lake at 18m of local pre-Neolithic date (Caton-Thompson Gardner1929, 1934 p. 15— 7). The lack of an y associatedcultural s ites or in fact of any 'M esolithic ' imp le-

ments per se suggests to us that the climate andvegetation of the unsubmerged parts of the oasis

cannot have been very inviting to man. TheSurvey of Egypt (O. H. Little 1936) contendedthat the Fayum lake had a shoreline at 22—24msomewhere between the 12 t h and 18 t h Dynasties ,basing themselves on the typolo gy of a s ingle potfound in s i t u for the date . Such archaeologicalevidence is all but convincing, and Caton-Thompson, Gardner Huzayyin (1937) haverefuted these contentions, suggesting that the22—24m beach appears to be the s torm-beach ofthe 18m Neolithic lake. However N. M. Shukr i

N. Azer (1952) have been able to show in-disputably on mineralogical grounds that the22—24m 'Gisr ' beach formations are of Palaeo-lithic date, as the mineral content does not com-pare favourably with that of more recent N ilo t icsediments . Consequently a free connection withthe Nile at more or less its present level at BeniSuef will already have pro vided the necessarygradient for the establishment of the 18m lake,which howe ver was not of long duratio n. G reatclimatic or hydrographic changes seem to havebeen taking place in Abyssinia at this time as asecond great pha se of silt deposition began inLower and Middle Egypt during the Mesolithic,blocking the mouths of the wadis and app arentl ythe Hawara Canal as wel l ( H u z a y y i n 1952a), sothat the Fayum was isolated a second time andthe 18m lake began to fall . The isolation of thelake is obvious as the Nile aggraded its bed to20 m above its present level at Turah duringNeolithic times ( H u z a y y i n 1952a) while theFayum lake dropped continuously in successivestages, becoming increasingly brackish as is wit-nessed by the firs t appea rance of the brac k-watermollusc H Y DR O BIA P E RA U DIER I after the fallfrom the 18m beach (Caton-Thompson C Gardner1934).

A similar return to more arid conditions inPalestine after the Subpluvia l I is possibly indi-cated by a lower level of Lake T iberias , whereR. Koppel (1932b) found innum erable s ilex

implements of pre-Ceram ic date to a depth of1—2m below the present water level and to adistance of 20— 30m out from the present shore.Three primitive 'quais' built at such a lower levelsuggest a post-Palaeolithic date. Further, moreconclusive evidence comes from the Belt Cave inPersia where the 20 cm Seal Mesolithic horizondated in the 10 t h millenium or later by radiocar-bon is overlain by 30 cm of loess which ap pa r-ently represents one of the final alternatinglayers of the southern Caspia n loess which Bobek(1937, 1953) has shown to be an evidence ofsizeable climatic oscillations. The overlying Ga-zelle Meso lithic horiz on date d at 6615 B. C.



27

± 380 by radiocarbon ( R a l p h 1955) is domin-ated by a Steppe form GAZELLA SUBGUTTEROSA

but also contains BOS PR1MIGEN1US

and CERVUS

ELAPHUS which with the absence of loess suggestan un mistakable trend to m oister conditions, prob-ably with a climate not much different than thatcharacterizing the present.

Briefly, following the Subpluvial I there wasapparently a relapse to more arid conditions forpossibly some 1500 years which were graduallyreplaced by a climate closer to tha t of the presentduring the first half of the 7 t h millenium, some-thing further substantiated by the relatively s lowfall of the isolated 18 m Fayum lake which seemsto have taken place some 7—8 000 years agojudging by the Egyptian cultural sequence.

T h e S u b p l u v i a l I I P h a s e o r N e o -l i t h i c M o i s t I n t e r v a l . T he c lim aticimprovement already noticeable during the laterstages of the Pos tplu vial II stage took on an im-por tant character in the following centuries anddeserves recognition as a second 'SubpluvialPhase' which was characteristic of the Ne arE aste rn N eo lit hi c p a r e x e l l e n c e , a nd pr ob -ably this relaxa tion of the 'climatic crisis ' wasintimately associated with the rapid spread of theearliest food-producing economy an d the estab-lishment, or rather the resumation of w idespreadcultural contacts and intercommunications duringthe Neolithic ( H u z a y y i n 1941 p. 319). The ex-

istence of such a prehistoric moist phase has beensuspected for m any years now (Caton-T hompson

Gardner 1929, Bate 1940) but no systematicstudy of the overall evidence has been made todate so that i ts duration and character were quitedifferently assessed by the indiv idua l a uthors. Sodoes S. A. Huzayyin (1941), for exam ple, give adate of c . 5500—2500 B.C . , G. W. Murray(1951) c. 8000—4000 B. C. while H.Bobek (1953)even speaks of a prono unced arid phase fromc. 9000—4000 B. C.

In Northern Egypt, the s ilts deposited to some20 m durin g high an d unusu al floods reaching theouter borders of the Nile Valley were e ventuallyinterbedded with angular wadi gravels along thesides of the valley in dicat ing a retu rn to localfluviatile activity and a moister climate ( H uz a-yyin 1952a). In all eve ntu ality th e increase inlocal rainfall leading to the protrac ted halt of thefalling 18m Neo lithic lak e at the 10m shoreline{Caton-Thompson Gardner 1929, 1934) wascontem porary. After the establishment of thisbeach the Fayum was colonized by Neo lithicpeoples. Radioca rbon dates of this Neo lithic 'A'industry have yielded figures of 4440 B. C. ±180 and 4144 B. C. ± 250 (Arnold L ibby 1951,Libby 1951), and since the earliest Ne olith ic of

the Haua Fteah Cave, Cyrenaica, which shouldbe a li t t le later than the Egyptian Neolithic in

general, has been da ted a t 4850 B. C. ± 350 bySuess (1954), a date of about 5000 B. C. for thebeginning of the Subpluvial II with its greaterrainfall should not be too far from the truth. Atthe close Neolith ic 'A' , the prehistoric Fayumlake began to fall again, temp orarily regained anequilibrium at 4m and then droppe d further to -2m(m. s. 1.) at which level the formation of an im-pressive salient beach, as well as the archaeologicalevidence, suggest a very constan t level of perh aps1700 years , remaining static throughout predy-nastic times, and extending at least into the periodof the 4 t h Dy nas ty (c. 2400 B. C.) whose sett-lements were found along the shoreline ( Caton-

Thompson 8c Gardner 1934). A greater rainfallthan today must again have counterbalancedevapo ration both du ring the brief halt at 4m andduring the long functioning of the -2m beach,something that is nicely shown by the lagoons,isolated by successive drops of the eventua llyseveral kilometer distant lake, which were ableto persis t as rain-fed lakes into the EarlyDyn astic pe riod. The absence of landshells in thebeach deposits, how ever, indicates tha t eitherthere was insufficient rain to give rise t o acontinuous ma t of vegetation or that the drainagewas only short and intermittent ( Caton-Thompson c Gardner 1934, p. 16-17). M. Am erC Huzayyin (1952) have identified several

'cycles ' of gravel disposition by lateral w adis in-terrupted by limited local erosion at the site ofMa 'adi. Possibly there may be a justification toattem pt a correlation of these periods of f luviatileactivity to the fluctuations of the Fayum lake,and we suggest tha t the aggradation of the Wa diDiglah characterized b y the deposition of coarsegravels before the earliest Midd le Predy nasticsettlement at Ma'ad i and again during Earl yDyn astic times coincided with the -2m Fayumlake.

Even in the more elevated areas of the S aharathere was apparently a relative increase in preci-pitation at the time. The artifacts of the Khargan'Beduin Microlithic ' (H au a Fteah 'Microlithic 'date d 5350 B. C. ± 300 by Suess (1954) are notfound near the mou nd springs b ut are lightlyembedded in the silt pans collected by rain washand run-off on the Kha rga S carp, suggesting thatthe intervals between rain were less prolongedat the time (Caton-Thompson 1952, p. 34—5).Th e oldest class of rock pictographs incised upo nthe exposed boulders and scarps throughout thenow utterly barren Libyan Desert and the Fezzan,as far south as latitude 22 ° N in Egyp t, are mostprobably Mesolithic to early Neolithic in date



28 Erdkunde- Band XI

(.H. Breuil 1928). These rock drawings variouslydepict giraffes, ibex, oryx , gazelles, lions, ostriches

and even people in the appa ren t p osture ofswimming (Djebel Uweina t, Gilf K ebir) , and tothis list must be added the elephants, rhinoceros,crocodiles and buffaloes appearing on the Fezzandrawings (Wadis Issaghen and In Habeter) (L.Frobenius, 1937). In later Neolithic times theelephant, rhinoceros, crocodile, buffalo and someof the antelope species have becom e locally ex-tinct as they are absent from the pictogra phs, or,presented in such a conventionalized and primi-tive fashion as if the artists were m erely carryingon a tradition drawing animals long disappearedfrom their environm ent as P. Graziosi (1951) hasdescribed it. To-day all these steppe animals are

gone and even jackals are a rarity, so that oneneed not hesitate to suggest a m ore abu nd antvegetation, something possible here only througha somewh at greater precipitatio n. Ho we ver theexact chronological position of the Bushmen andLibyan series, as Breuil has called the earlier andlater groups of draw ings, is not certain. Frobenius(1937, p. 72— 3) associates the older draw ingswith a Cap sian-ty pe industry, while Graziosiidentifies the earliest 'Berguig hun ters' group as'Neolithic'. It is, all in all, well possible that apart of the oldest group dep icting the characteris-tic tropical fauna are actually older and belongto the Subpluvial I, but at the present state ofour know ledge it is advisable to defer judgeme nt.Elsewhere in Egy pt there is further substantiationof greater hum idity until c. 2400 B. C. Th eauthor noted wall-reliefs of 5 t h Dynas ty datein tombs at Sakkara depicting A D EN O TA K O B,

C ER VU S E L AP HU S, O N OT RA GU S M E GA CE RO S,

O R YX A L GA Z EL , IBE X an d F E L IS L E O, animalswhich are f irst encountered in the Ce ntral Sudanto-day. These species are largely steppe animalsand would not have favoured the riverine zone,rather they would have lived in adjacent patchesof grassland, How eve r Egy ptian wall-reliefs oflater date seldom depict these animals again, whileto-day the desert gravels encroach right upon theirrigated fields of the N ile f loodplain, practicallywithout a single shrub existing on the west bankabove the highest flood levels.

Th e main period of aggradation of the ' lowerterrace' of the Jord an Valley is connected with agrey-black marshy loam, well seen overlying thegravel ho rizon and a 70 cm light grey loam to adepth of 1 m in the Eddesiye profile (Picard1932). The further occurrence of this grey-blacksoil to a depth of 1 m betwe en Jenin and Affulein Samaria, reaching to the bottom of the wad isand containing mollusca which cannot live in anyof the dry w adis of the area to-day , suggest a

meso-neolithic period of greater rainfall andgreater warmth to Picard. Similar black earth

layers have also been found in the Wad i Musraw aand at Tel Aviv, as well as at Ksar Akil andTripol i (Ewing 1951), where one contained Me-solithic tools at its base.

From a study of the zoological biotypes foundin the post-Natufian or early Ne olithic level ofthe Abu Usba Cave, Mount Carmel , Stekelis andHaas (1952) have suggested a period of somew hatgreater rainfall that the present, possibly am ount-ing to 70—80 cm or over (compared with 55—60today). Typical s teppe species are absent, andreptilian types such as O PH IS AU RU S A PU S, A GA -

MA STELLIO, an d C H AM A E LE O C H AM A EL E ON ,

numerous thrushes, and the mollusc CYCLOSTOMA

OLIVIERI all favour a quite dense vegetationalcover and usually exist under somewhat morebount ifu l ra infall condi tions . In Arabia , H.Philby

(1933) found gravel spreads and other lacustrinedeposits containing fresh-water shells . These de-posits in the northern fringes of the Rub al-Khalicontained innumerable stone implements classifiedas 'Neolithic ' . In the Hasa, P.B.Cornwall (1946)found a chalcolithic site situated b eside lacustrinesediments likewise abounding in fresh-watershells near Uqu air . T his former lake bed remaineddry even after a 100 mm rainfall in Decem ber1940, the influx of drainag e ba rely able to formsmall channels in the ove rlying sedim ents. In

general there is much evidence in favour ofa numerous Neolithic population in the SyrianDesert, l iving with an a bunda nce of wate r,pasture and game (5. Passarge 1951, p. 473—4).In Baluchistan and the M akra n, Sir A. Stein

(1934) found numerous chalcolithic s ites markedby great stone barrages intended to serve as re-servoirs and secure irrigatio n for v alleys whichare now u tterly desolate, which fact he can onlyexplain by a substantially moister climate somefour millenia ago.

Lastl y, the biblical references to the Deluge areconfirmed in ancient Babylonian trad ition, and,the Gilgamesh Epic found in Assurban ipal 'slibrary as well as the legends from Bamb yce re-corded by Pseudolucian, seem to suggest a ca-tastrophic s torm both in Mesopotamia and inSyria accompan ied by strong winds, unusuallygreat electricity, and much rainfall extendingover a period of several days and causing largeareas adjacent to the Euphrates to be temporarilyinund ated.S tratigrap hical profiles have confirmedthis controversial legend, in particular since L.Woolley (1954, p. 31— 35) was able to find an8—11 f t. thick uniform layer of water - la in mud,material bro ught dow n from the middle reachesof the Eup hrates an d deposited in water at least




25 ft . deep upon the Neo lithic s ite at Ur. If weassign an a verage of 35 years per reign, as is the

case of the more credible king-lis t of the Is t

Dy nasty of Ur, we can calculate a date ofsomewhere between c. 3500 and 4500 B. C. fromthe Babylonian royal chronology, depending uponwhether we consider the I s t Dyna s ty of Erechand the I s t D ynas ty of Kish contemporary ornot. These traces of a unique, g reat f lood are notisolated an d can be encountered at Erech (Wark a),as well as at Dar-i-Khazineh on the Karun River.One may at least speculate on the meteorologicalimplications of the phenom enon. A t the presenttime major f loods occur on the Euphra tes whenexceptionally great quantities of winter snow inAnatolia are suddenly melted by strong, warm

and moist southwesterly winds which add theirown precipitation to the swollen rivers . Howeverpermanent snow was at a lean minimum duringthese millenia, as S. Erinq (1952) was able toshow that the present cirque glaciers of Anatoliaare generally recreations following an almosttotal disappearance, most probably during the'Cl imat ic Optim um ' . Consequent ly an extra-ordina ry rainfall ap pears to have been main lyresponsible.

In retrospect, the return to arid conditionsafter the close of the Subpluvial I of the 9 t h

millenium lasted until perhaps 5500 or 5000 B. C.but as was seen, the last m illenium or so of its

duration was more like the present and a li t t lemoister than the earlier climatic subphase. Con-sequently a Postpluv ial I l a from p erhaps roug hly8000—6500 B. C. and a Postpluv ial l i b fromc. 6500—5000 B. C. can be tentatively identified,of which the former was somewh at more arid.Subsequently there was a definite climatic im-provement characterized by an appreciableincrease in moisture in the southeastern quadrantof the Me diterranean climatic province, andapparently accompanied by a local increase intempe rature as well. This period which lastedfrom c. 5000 to 2400 B. C. is designated as Sub-pluvial II and represents the Neolithic humid

phase as already recognized by Huzayyin (1941)and others . Geologically speaking, the SubpluvialII was again fairly insignificant, b ut the increasein precipitation was sufficiently great to pe rmitthe flourishing of a more exubriant fauna andflora in areas climatically unsuitable to-d ay, andwas thereby of great s ignificance to the humanhabitat. From the fluctuations of the Fayu m lakeas well as from the phy siogr aphi c cycles atMa'adi (A m e r 8c Huzayyin 1952) there were atleast three tempora ry decreases in rainfall duringthe Subpluvial II , two of them a lit t le before anda little after 4000 B. C., and the th ird at the closeof the predynastic period ab out 3000 B. C.

Chronologically the Postpluvial II a and bwould coincide approximately with the Preboreal

and Boreal periods respectively in Europe, whilethe Subplu vial II clearly was almost synchronouswith the Atlantic phase. From a meteorologicalstandpoint this chronology and climatic succes-sion pro ve to be of considerable interest, as it haslong been a popular theory that the postglacialClimatic Optimum corresponded to an extensiveperiod of maximu m aridity in lower middlelatitudes. Further below it shall be noted that thePostpluvial III phase, which coincides more orless with the European Subboreal, was more aridthan the present day, so that there appears to be acurious parallelism between moisture trends inEurope and in the Near East. However the indi-

vidual meteorological problems associated withthis climatic succession deserve a more detailedattentio n, and m ust therefore be reserved to asubsequent paper.

A recent arid phase c. 2400—850 B. C.

Th e climatic f luctuations of the last 4000 yearshave not been of sufficient duratio n or magn i-tude to leave much physiographical evidence, andit becomes necessary to resort more and more toarchaeological evidence and literary sources. Fol-lowing upon the Subpluvial II the climate, orrather, the rainfall of the Nea r East deterioratedvery severely and during the maxim um of the

short but arid period, the Postpluvial III as weshall call it, the average annual precipitation ofthe Nea r East was less than tha t of any other timesince c. 9000 B. C.

It appears that the ancient Fayum lake beganto fall after c. 2400 B. C. s top ping briefly at alevel of —11m and this fall con tinued to at least—13m, at the time when Ptolemy Philadelphus(285—247 B. C.) began the systematic p roject ofreducing the lake to its present low level at—45m (Caton-Thompson and Gardner, 1934).The re ap pears to ha ve been a general exodus fromthe Libyan Desert in VI t h Dynas ty t imes , where-af ter the rock pain t ings of the abandoned Neo-

lithic sites ceased, and the archaeological evidenceof the immigrat ion of the Nubian C-group f romthe Sahara in VI t h Dynasty times is substantiatedby the historical rec ords referring to the firstappearance of the Tehennu Libyans in the N ileValley during the reign of Pepi I I (O. H. Myres1939). In other words, a pronounced reduction ofrainfall a nd pasturage in Egy pt and the LibyanDesert quite certainly took place after 2400 B. C.

As a f irst l i terary source, the Old Testam entma y be cited, since the biblical references tosevere droughts and cata strophic famines ha ve adefinite be aring upon rainfall conditions in an-cient Palestine. Particular attention should be



30 Erdkunde Band XI

dra wn to the fact th at of over eleven specificreferences to severe and prolonged drou ghts or

famines found in the Bible2

), only two belong tothe period after 850 B. C. W e find reference tosuch drough ts in the days of Abra ham (c. 1800B. C.) , Isaac and Jacob, during the rule of theJudges, and the reigns of Da vid (c. 1010— 970B. C.) and Ahab (876—853 B. C.) . Furtherm oreR. Koppel (1932a) describes a Dead Sea level of—8m on the grounds of a submerged m arine ero-sional bench for which he suggests a da te ofc. 2500— 1200 B. C. on accou nt of a specificBronze Age type implement nearby.

M.E.L.Mallowan (c . f. Bobek 1954) points o utthat the number of known N eolithic and B ronzeAge sites from the s teppe of M esopotamia are in

the ra tio of 5 : 1 and he concludes that man ysettlements had to be aband oned after the middleo f the 3 r d millenium B. C. due to a declining rain-fall. In the Aste rabad loess area, H. Bobek (1937)notes that the numerous Bronze Age tepehs ofthe area were obviously occupied at a periodwhen the luxurian t forest was replaced by steppeand loess deposition, probably recorded in thefinal oscillating layers of the uppe r loess. And inthe Karakum Deser t , A. Schultz (1927) identifiesa period of recent desert conditions d uring whichthe mobile dunes, found especially in the vicinityof the Am uda rya, came into existence at a timecorresponding to the European Subboreal.

At Enkomi-Alas ia , Cyprus , C. F. A . Schaejfer(1952, p. 358—9) identified six gravel beds alte r-nating with sand, which he assigns to six conse-cut ive years of par t icular ly abunda nt precip i ta-tion. An alluvial layer of almost lm depthaccumulated after this , which he s tratigrap hicallydated at 1150— 1100 B. C.

All in all the existence of a dry p eriod betweenc. 2400— 850 B. C. seems rela tively certain ,although the indications of a tempora ry fluctua-tion in the late 12 t h century B. C. from Cyp russuggest that the maximum aridity had been reachedbefore tha t date, possibly midwa y in the 2 n d

millenium. But, that the Postpluvial II phase wasby no means over until at least 850 B. C. is borneout by the repeated biblical references to verysevere droughts in the early 1 s t millenium B. C.Chronologically this Postpluvial III periodcorresponds well with the European Subboreal,just as the Subpluvial II was contemporary tothe Atlantic of Europe . The climatic relationshipas regards a pattern of general atm ospheric cir-culation is not quite so obvious, and presents aninviting problem for further research.

2 ) I Gen. 12; I Gen. 26; I Gen. 43; I Gen. 47; Ruth 1;I I Sam. 21; 1 Kings 17 , 18; II Kings 4; II Kings 6; Joel 1;Jeremiah 14.

The climate of the historical period since 850 B. C.

Perhaps our best evidence to show that aper iod of mois ter condi tions of greater perm a-nency had set in comes from H. E. Wright Jr. 's(1952) geological study of the archaeological sitesof the Chemchemal Plain in Kur distan, wherealluvial f il l deposited to a depth of almost 10min a deep erosional gully of the Qu adh ai Chaicontained po ttery te ntativ ely identif ied as lateAssyrian (750—600 B. C.) . How eve r none of theother r ivers nearby w ere sufficiently advan ced intheir cyle of erosion to go over from degradationto aggradation, a process absent in the generalarea since the pluvials. Probably this isolated andtem porary exception records a f luctuation tomoister conditions. Confirmation is availablef rom the Karakum where a renewed phase ofdune-fixation took place conte mp orary to theonset of the European Subat lant ic (Schultz 1927).

From this time onwards climatic conditions inthe Near East seem to have approached a normresembling t ha t of the present in all respects. Onema y deduce this largely from a lack of evidenceto the contrary. From the scanty historical infor-mation available from literary sources we can dolittle more th an suggest that the last five centuriesB. C. were very similar to the firs t half of the1 s t m illenium A. D. In later Rom an times andespecially in Byzantine and Islamic times weknow considerably more. However it is notnecessary to go into detail at this point, and thereader is referred to the readily accessible wo rkof C . E. P. Brooks (1926, 1931) on the Nile levels,E. Brückners (1890) invaluable s tudy of theCaspia n Sea fluctuations in historical t imes, andlastly to the detailed weather chronologyassembled by R. Hennig (1904) with furtherreferences in A. Wagner (1941). How eve r it willnot be superfluous to discuss these wea thercriteria themselves briefly, and to outline an ddiscuss the short-term climatic f luctuations of thelast millenia on t he basis of the ab ove sources.

Although Brückner believed the Caspian Sea

level fluctuates in direct propor tion to the high-wate r level of its chief tr ibu tary the Volga, L. S.Berg (1934) has pointed out that whereas therainfall of the Volga drainage basin increasedafter 1881, and especially after 1900, the C aspianhas fallen steadily, alone by 3.5m durin g 1900—1925. Evidently the evaporation over the Caspianor over the whole watershed increased con-siderably. A. Wagner believes th at since only thew i n t e r ra infall of Russia had increased, theevaporation over the northern watershed re-mained the same, and he concludes that evapo-ration over the Casp ian itself, in all prob abilityover the southern half of its great expanse, must




have increased. A good proof th at the C aspianSea fluctuations reflect Ne ar Eastern moisture

trends can be obtained by a comparison of therecorded var ia t ions of the ann ual maxima of theDead Sea (M. R. Bloch) and the Caspian (W.Koppen 1936, G. A. Taskin 1954) for the o nlycomparable, 19-year period 1928—1946. Thelinear correlation of the relative movem ents ofboth non-outlet lakes is very good, yielding acoefficient of correlation of r = + 0.7 0, where-from, considering th e large distances and localfeatures invo lved , as well as the poor correlatio ncoefficients obtained between tree-ring curves invarious localities of the Am erican Southwest, i twill be seen that the we may consider the histori-cal fluctuations of the Ca spian Sea as an in-valuable criterion of Ne ar Eastern climate.

The curious pattern of Nile levels has longproved to be a perplexing problem , which is by nomeans simplified by the doubtful reliability of somuch of the da ta. H ow eve r as a f irst point aclear dis tinction must be drawn between totalv o l u m e , a nd h igh a nd l ow -w a te r l e v e l s .Very interestingly the total Nile waterflow1870—1900 was 10 > above average whilethat of 1900—1920 was 14.5°/» below norm al,which is in full accord w ith the m ean an nua l pr e-cipitation of Jerusalem am ounting to 736 mm in1880—1900 and 660 mm in 1900—1915, not tomention the 3.5m fall of the Caspian during

1900—1925. This again lends support to thetheory tha t rainfall trends north and south of theSahara are very similar. However the problemof high and low-w ater level patterns is a lit t lemore complicated, and other than that we knowtha t the flood-level is solely determin ed by theBlue Nile and its tributaries in Abyssinia, whilethe equatorial drainage entering through theWh ite Nile above the mo uth of the Sobat makesup less than 5 °/o of the N ile volum e (Pietsch1910), we do not as yet understand this patternof maxima and minima. ]. Hovermann (1954)has shown that seasonal distribution and absoluteamounts of precipitation in Ethiop ia, as well as

the historical f luctuations of the lower limit ofsnowfall all play their part to ma ke the problemrather complex.

The Nile f loods of the 1 s t century A. D. werehigh and there is evidence of a great droughtlasting many years in Central Asia about A.D. 80.How ever the Nile floods became considerablylower during the 2 n d century and the travellerCosmas refers to perennial snowfields in Ethiop iaat the time (H over m ann 1954). The year 145 issignalled out by catastrophic f lood through outthe Mediterranean. I t is no mere coincidence thatC . E. P. Brooks with Miss L. D. Sawyer (1931)

af ter inves tigat ing the wea ther record of Clau-dius Ptolemy (observabat A. D. 127—151) taken

at Alexandria, should arrive at the conclusionthat occasional rare depressions interrupted thesteady fine summ er weathe r of the Eastern M edi-terranean at this time. All in all the early 2 n d

century appears to have been quite moist, in con-trast to the 1 s t century A. D. which was appa-rently drier .

There are records of great heat and drough taround A. D. 260, and about the year 310 Cyprussuffered f rom 36 years of e xtraordinary drought .In A. D. 333 the level of the Dea d Sea, whichJ. Enge (1931) has shown to be in proportion tothe local rainfall , w as the same or probab ly alitt le lower than tha t of the present. Between 360

and 363 there are further indications of droughtsin Asia and Africa, b ut by the close of the cen-tury, a period of moister and cooler co nditionsset in. The re was an abnorm ally great rainfall inA. D. 395, and the winters 400/01 and 418 or421 were extremely cold. In 400/01 the Black Seawas supposedly covered in its entirety (surelyexaggerated ) by a sheet of ice for twe nty days.This more humid phase was cut off about A . D.450, and the years 454 and 484 were very dry.During the reign of the Sassanid Firuz (A. D .459—484), and during the 6 l h and 7 l h centuriesthe Caspian Sea seems to have hovered aroundthe —4m 'Derbent S tadium' of Leontyev and

Fedorov (1953). Following another severedrought f rom A .D . 512—517 in Palest ine ,moister conditions prevailed until the late 6 t h

century when we can point four phenomenal lycritical years 591, 593 or 594, 598 and 605 or 606when a per iod of a lmos t uninter rupted droughtand heat lasted nine consecutive months in theMediter ranean wor ld Arab legends fur ther re-cord a seven-year droug ht ' in the days of M u-hammad' , a severe drought is recorded in A. D.630, and during the nine month long droughtof the 'Year of Destruc tion' in 640, the heat'burn t the soil of Arabia to ashes ' as the Arabicauthors have put it . All indications show that

there was an extremely dry snap between about590 and 647 which latter date marks the gradualbeginning of a long series of seldom interruptedrecords of very cold winters or w et years in AsiaMinor such as 673, 717/18, 763/64, 800/01, 829,859/60 and 1010/11, which are curiously paral-leled by excep tionally low Nil e f loods betweenA. D. 700—1000, which reached an absoluteminimum A. D. 750—800.

When Istakhri vis ited Derbent between 915and 921, the Caspian seems to have stood at 10.5mabove the present level, but during the 12 t h cen-tury seems to have fallen to —2.5m Pe rhaps this



32 Erdkunde Band XI

tem porar y low level occurred at the time of theaccentuated dry years 1158—60 in the Mediter-

ranean world, since the years 1099/1100 and 1129were still cooler and humid in the Near East.After this drier interval, the Casp ian rose veryrapidly at the end of the 13 , h century and attaineda maximum of 13m before it began to fall againin A. D. 1307, althoug L. S. Berg (1934) woulddiscount this and the 915/21 maximum on thegrounds that the indicating mollusc, CARD1UM

EDULE, of the modern C aspian does not occurc o n t i n u o u s l y above 6m over th e presentlevel, However such very temporary high levelscannot be expected to produce wholesale move-ments of the m ollusca, while the excuse of 'sheeraccidents ' cannot explain the numerous excep-

tions to this 6m 'continuous' occurrence ofC. EDULE. Ne ither can the mediaeval inflow ofthe Amud arya be called upon to account for thishigh level, since newer Russian on the SarykamishDepression (Tolstov, Kess and Shdanko, 1954)has revealed that there has been no recent note-wo rthy or sustained flow of of w ater in theUzb oi Ch ann el, as the Bronze Age sites of thelowest Uzboi terrace have not been disturbed.

During the early 15 t h century the Caspian roseonce again to a level of 6m and simultan eouslythe N ile flood-levels were so low as to causesevere famines in Egypt. T he n ext informationcomes from the f ir st half of the 17 t h century,

when we note a new C aspian level of 7m, somevery cold winters in Asia Minor such as 1608 and1621 when the Bosporus was icebound, as wellas a lowering of the climatic snowline with areadvance of the Near E astern glaciers . H. Bobek(1940, 1937) identifies nu mero us m oraines ofFernau age in Kurdistan, implying a considerableadvance of the Cilo and Sat D ag glaciers , whilesimilar moraines occur in the Tacht e Suleimangroup of the Elburz . S. Erinç (1952) has assigneda major advance to the Anatolian glaciers duringthe 17 t h century, confirming the observations ofR. Leutelt (1935) in Lazistan. The snowline de-pression appears to have been between 50 and

100m in both Iran and Turkey. Whereas theEthiopian highlands enjoyed lit t le or no snowfallin the 16 t h century, the snowline apparentlyhovered at about 4200—4400m in the early 17 t h

century according to ma ny detailed traveldescriptions of contem porary Europeans (Hover-mann 1954).

After dropp ing to + 2.3 m by 1715/20, the levelof the Caspian rose rapid ly c. 1740 to reach amaxim um 1742/43, falling to a new low level1765/66; correspondingly the winte r of 1709 wasvery mild in Is tanbul while tha t of 1742 wasquite severe in Syria. In the 1780's the Caspian

began to rise once more, this time achieving a newmaxim um of about 5m in 1809/15, but falling

back to +2 .4 m by 1830, about which level itf luctuated for the remainder of the century .Anothe r general advan ce of the Anatolian andElburz glaciers before 1850 was paralleled byadvances in the Caucasus where the Azau orBaksan glacier advanced rapidly in 1849, only toretreat by 1.5 km during 1873—1911. Againfrom a number of travel descriptions fromEthiopia (H ôver m ann 1954, Werdecker 1955) itappears that w hile the lower limit of snowfallwas at about 3300m during the years 1830—50,little or no snow fell in Abyssinia durin g the lasthal f o f the 18 l h century, as has been the casesince about 1880.

With regards to the well-known 'recent cli-matic f luctuation' experienced in higher middlelatitudes during the last century or so, someremarks can be made from a s tudy of some ofthe few longer meteorological records of the Ne arEast (H. H . C layton 1927, 1934, 1947). In thecase of temp erature there appea rs to be no out-spoken trend, although a s light tendency towardshigher temperatures can be noticed. The period1911—1940 in Alexandria had a mean annualtemperature of 0 .4 ° C higher than th at of 1870—1910. Similar results can be cited for B eirut w ithan increase of 0.8 ° between 1884— 1913 and1911—1914, but there was no change whatever

in Bushire, while in Krasnovodsk a bare increaseof 0.1 ° C occurred between 1883—1910 and1911 —194 0. A n the w hole the trend betweenmean annual temperatures and mean wintertempe ratures w as identical. Different is the pic-ture presented by the precipitation records. Alex-andria received an average annual precipitationof 220 mm for the period 1887—1910, 173 mmfor 1911— 1940; Beirut 932 mm for 1876—1910,858 mm for 1911 — 1940; Q uetta 246 m m for1878—1910, 218 mm for 1911 — 1940. In otherwords there was a general and considerabledecrease in precipita tion, and if we average thedecrease from the four long-term records of

Alexa ndria, Beirut, Bushire and Q uett a for theyears 1881 — 1910 and 1911 — 1940 we ob tain anapproxim ation of 10 — 15 % decrease on the60 year mean.

Th e testimony of the weath er records is borneout in many othe r ways. The w ater-tab le fell bymany meters in Syria, Egypt and the Saharaduring the present century , and from the descrip-tion of many travellers one can note a rapidextinction of organic life and increasing deso-lation in the Liby an Dese rt and in T ibesti(W. Thes iger 1939). A sharp decrease in rainfallin Syria between 1920—1934 resulted in a series




of bad crop failures. H ow eve r as the we ll-knowntemp erature trend of higher latitudes was defini-tely interrupted since 1938 by the repeated oc-

currence of very cold winters , so too has thedecline in Near Eastern precipitation apparentlybeen reversed, the cold winters of higher latitudesaccompanied by bountiful rainy seasons in theeastern Mediterranean Basin. I t does appearlikely th at the present climatic fluctuation, andthereby the menacing desiccation of the N earEast has been halted or at least interrupted.

Reviewing the climate of the historical period,it is clear that there has been no long-term desic-cation 's ince Roman times ' , rather we may dividethe associated clima tic phase, for which the nam ePostpluvial IV is suggested, into two subphasesof which the firs t or IVa p eriod c. 85 0— B.C.—A.D. 700 was generally characterized by lakelevels equal to or lower tha n those of the prese nt,by numerous records of warm sum mers andsevere droughts and very few references to coldwinters or wet years; the subsequent PostpluvialIVa phase, except for tw o intervals in the 12 t h

and 20 t h centuries , was characterized by CaspianSea levels well above those of the present, byman y cold winters and other indications of agreater rainfall or lower evaporatio n. In short,the earlier IVa subphase was a li t t le warm er thanand also drier than the later IV b subp hase.Within this latter period w e can identify distinctbut temporary m axima of humidi ty f rom A. D.

700—1000, during the late 13 t h , the early 15 t h ,the f ir st half of the 17 t h and the beginning of the1 9 t h centur ies3) .

Summary and Conclusions

The results of the above investigation m ay bebriefly summ arized below the headings of thedifferent climatic phases identif ied for the N earEast in the course of the discussion:

P o s t p l u v i a l I. D u ri ng th is p eri od ofunknow n duration following the Last Pluvial,wind-erosion was particularly pronounced, tem-peratures apparently a li t t le lower, and the rain-fall of the Near East was somew hat less thanthat of the present, probably reaching a minimumduring the 11 t h m illenium B. C.

S u b p l u v i a l I. (p ro ba bl y d ur in g th e 9 t h

millenium B. C.) A marke d, but te mp orary im-provement of moisture conditions is evident inte Near East, with a mean precipitation perhaps

3 ) The author wishes to express his s inceres t thanks toProfessor Carl Troll, Bonn , for his cont inu ed interes t ,valuable suggest ions and kind generosi ty. The s t imulat inglectures and cr i ticism of Professor F. K. Hare of McGi l lUnivers i ty, under whose supervis ion this s tudy was int ia-ted, are also grateful ly acknow ledged.

halfway between that of the Last P luvial andmodern conditions. I t appears that a last read-vance of the disappearing Wü rm glaciers took

place during this time, accompanied by lowertemperatures effective to at least as far south asla t i tude 33 ° N .

P o s t p l u v i a l I I a . (pe rh ap s 8000— 6500B. C.) . Following the tem pora ry humid phaseprobably associated with the last glacial relapseof the 9 t h millenium, tempe ratures rose andtypical p ostpluvia l conditions of aridity set inwith a rainfall a li t t le less than tha t of the pre-sent.

P o s t p l u v i a l l i b . (c. 6500 to 5500 or5000 B. C.) marke d an impro vem ent in precipi-tation, characterized by moisture conditions

closely resembling those os the last century ofour era.

S u b p l u v i a l I I . (c. 50 00 —2 40 0 B . C .)This Neolithic moist interval, as it has alreadybeen called, enjoyed a rainfall somew hat greaterthan that of the present despite indications ofhigher local temperatures. A higher rate of eva-poration consequently implies that the increasein precipitation was appreciable, again probablyon a p lane halfway between that of a p luvial andmodern conditions. There appear to have beentemp orary decreases in rainfall shortly beforeand after 4000 B. C. and again about 3000 B. C.

P o s t p l u v i a l I I I . (c. 240 0— 85 0 B. C .).A renewed decrease in precipitation during thelate 3 r d m illenium led to a longer period of aridconditions probably accompanied by greaterwa rmth , with an average rainfall below th at ofthe present, interrupted by at least one moisterinterval ( in the 12 t h century B. C.) .

P o s t p l u v i a l I I I . (c. 2400— 850 B. C .)700.) Ushered in by a short interv al of q uite moistconditions, the period was characterized by arainfall abo ut the same as, or rat her a trifle lessthan to-day. In general the winters s til l appearto have been a li t t le warmer. A very severedrough t between perha ps A. D. 590—645 can

actually be determined, which however, onlyrepresents a minor climatic f luctuation prior to achange to s lightly different conditions of rainfalland temperature.

P o s t p l u v i a l I V b . (A . D . 700 — .) A fterabout A. D. 650 an increasing number of quitecold winters were recorded in the Near East,during which the Black Sea was apparentlyfrozen ov er on two occasions (in A. D. 673 and800/01), and ice even formed on the Nile (inA. D. 829 and 1010/11). The rainfall was a li t t lehigher on the average, but marked by continuousshort-term fluctuations over a fair ly wide range.



34 Erdkunde Band XI

On ly since 1900 can one speak of a certain cli-matic deterioration which has been destructive toorganic life in such ma rgina l pastures as the ele-

vated plateaus of the Libyan Desert. This veryrecent clima tic fluctuation is associated with a10—15 % decrease in rainfall, a fall of level inlakes with interior drainage, a decreased Nilevolume and a retreat of the peripheral zone gla-ciers. There seems to be no overall tendency tohigher temperatures and one is inclined to see ama jor cause in the retreat of the glaciers in adecrease in precipitation, in itself largely con-fined to the winte r season, and no t so much in atemperature amelioration.

References

1. Alimen, Henriette: Préhis toire de l 'Afr ique. P ar is ,1955.

2. Amer, M. an d Huzayyin, S. A.: Some phys iographicproblems r e l a t ed to the pred ynas t i c s i te of Ma ' ad i , Proc .Pan-Af r i can Congres s on Prehi s tory , Nai rob i 1947. Oxford ,1952, p. 222—24.

3. Arnold, ]. R. an d Libby, W. F.: Radiocarbo n dates ,Science 113, 1951, p. 111—120.

4. Bate, D. M . A.: The foss il an telopes of Pales t ine inNatuf ian (Mesol i thic) t imes . Geol . Mag. 77, 1940, p.418—433.

5. Berg, L. S.: The level of the Caspian Sea in his tor ict imes ( in Russ ian) . Problem y f is izeskoj geograf i i (Lenin-grad), 1, 1934, p. 11—64.

6. Blake, G. S.: The Geology and Water Resources ofPales t ine. Jerusalem, 1928.

7. Bloch, M. R.: Wri t ten communicat ion, 6. 4. 1956.

8. Bobek, H.: Die Rol le der Eiszei t in Nordwest i ran,Z. Gletscherk. 25, 1937, p. 130—183.9. Bobek, H.: Die gegenwärt ige und eiszei t l iche Ver-

gletscherung im Zentralkurdischen Hochgebirge, Z. Glet-scherk. 27, 1940, p. 50—88.

10. Bobek, H.: Klima und Landschaf t I ran s in vor- u ndfrühgeschichtl icher Zei t , Geogr . Jhrb er icht a. Österreich 25,1953/54, p. 1—42.

11. Breuil, H.: Les gravures rupestres du Djebel Ouenat .Revu e Scient i fique 66, 1928, p 105—117.

12 . Brooks, C. E. P. Cl imate through the Ages . London,1926. Revised 1949.

13 . Brooks, C. E. P.: The changes in cl imate in the OldWorld dur ing his tor ic t imes . Quar t . J . R. Meteor . Soc. 57,1931, p. 13—30.

14. Brückner, £. : Kl imaschw ankun gen seit 1700. Geogr .Abhdl . 2, 1890.

15 . Caton-Thompson, G. and Gardner, E. W.: Recentwork on the problem of Lake M oer is , Geogr . Jour . 73 ,1929, p. 20—60.

16 . Caton-Thompson, G. an d Gardner, E. W.: The De-ser t Fayum. London, 1934. 2 Vol .

17 . Caton-Thompson, G., Gardner, E. W. an d Huzayyin,S. A.: Lake Moeris . Re- invest igat ions and some comments .Bul l . Ins t . Egy pte 19, 1937, p. 243— 303.

18 . Caton-Thompson, G.: The Kharga Oas i s i n Prehi s -tory, London 1952.

19. Clayton, H. H.: Wor ld Weather Records . Smi th-sonian Miscellaneous Coll., 79, 1927; 90, 1934; 105, 1947.

20 . Coon, C. S.: Cave Explora t ions in I r an , 1949. Phi l a-delphia, 1951.

21 . Coon, C. S. an d Ralph, E.K.: Radiocarbon datesfor K ara K am ar , A fghanis tan, Science 121, 1955, p. 921bis 922.

22 . Cornwall, P. B.: Ancient Arabia : explora t ions inHa sa 1940—41, G eogr . Jo ur . 107, 1946, p. 29—50.

22a. Desio, A.: Ap pun t i geografici e geologici sul la ca-tena del lo Zardeh Kuh in Pers ia. Mem. Geol . Geogr . ,Firenze, 4, 1934, p. 141—167.

23 . Enge, J.: De r Anst ieg des To ten M eeres 1880—1900und seine Erkläru ng. D r . Phi l . Disser ta t ion, Leipzig, 1931.

24 . Erinç, S.: Glacial evidences of cl imat ic var iat ion s inTurkey. Geogr . Ann. 34, 1952, p. 89—98.

25 . Ewing, ]. F.: Comment s on the r epor t of Dr . H . E .Wrig ht , J r . on his s tudy of Lebane se marine terraces . Jour .Near East Studies , 10, 1951, p. 119—122.

26 . Fourtau, R.: Cont r ibut ion à l ' é tude des dépot s n i lo-t iques . Mém. Ins t . d 'Egypte 8, 1915, p. 57—94.

27. Frobenius, L.: Ekade Ektab , d i e Fel sbi lder Fezzans .Leipzig, 1937.

28 . Gabriel, A.: Die Lut und ihre Wege. Zei t . Erdkunde,Frankfur t , 1942, p . 423—442.

29. Graziosi, P.: Les problèmes de l ' ar t rupestre Libyqueen relat ion a l ' amb iance Saha r ienne . Bul l . Ins t . Déser t 2, i ,1952, p. 107—113.

30 . Gross, H,: Das Al l eröd- Inter s t adia l a l s Lei thor i zontder letzten Vereisung in Europa und Amerika, Eiszei tal teru. Gegenw. 4, 1954, p. 189—209.

31 . Gross, H.: Per sonal communicat ion , 14 .5 .56 .32 . Hennig, R.: Kata log bemerkenswer t er Wi t t erungs -

ereignisse. Ab hdl . K. Preuss . Me teor . Ins t . 4, i i , 1904,p. 1—93.

33. Hövermann, ].: Über d i e Höhenlage der Schnee-grenze in Äthiopien und ihre Schwankungen in his tor ischerZei t . Nachr . Akad. Wiss . Göt t ingen I Ia, Nr . 6, 1954,p. 112—137.

34 . Huzayyin, S. A.: The Place of Egypt in Prehi s tory .Mem . Ins . t Egy pte 43, 1941.

35 . Huzayyin, S. A.: Recent physiographic s tages inthe Lower Ni l e Val l ey and thei r r e l a t ion to hydrographieand cl ima t ic changes in Abyssinia and E ast Afr ica. P roc.Pan-Af r i can Congres s Prehi st . 1947, (Oxford , 1952) p .

75—78.36 . Huzayyin, S.A.: Ne w l ight on the Upp er Palaeol i -t h i c o f E g y p t . Ibid. p. 202—204.

37 . Knetsch, G.: Beobachtungen in der l ibyschen Sahara,Geol . Rund. 38, 1950, p. 40—59.

38 . Koppel, R.: U fers tudien am Toten M eer . Bibl ica 13,1932 a, p. 6—27.

39 . Koppel, R.: U fer s tudien am Genesare th . Das Hei -l ige Lan d 76, 1932 b, p. 65— 69.

40 . Koppen, W.: Schwankungen des Kaspischen Meeres .Ann. Hydrographie u . mar i t . Meteor . 46 , 1936, p . 47—49.

41 . Leontyev, O. K.: an d Fedorov, P. V.: Zur Geschichtedes Kaspimeeres im Spät- und Postglazial ( in Russ ian)Nachr . Akad. Wiss . USSR, geogr . Ser . 1953, 4, 64—74.

42 . Leutelt, R.: Glazialgeologische Beobachtungen imLasis tanischen Hochgebirge. Z. Gletscherk. 23, 1935, p.67—80.

43 . Libby, W. F.: Radiocarbon dates I I , Science 117,

1951, p. 291.44 . Little, O. H.: Recent geological work in the Faiyum

and in the adjoin ing por t io n of the Nile Val le y. Bul l. Ins t .Egypte 18, 1936, p. 201—240.

45 . McBurney , C . B„ Trevor, J . C . an d Wells, L. H.:A foss il human mandible f rom a Leval lo i so-Mous ter i anhorizon in Cyrenaica, Nature 172, 1953, p. 889—891.

46 . Murray, G . W. : The Eg ypt ia n cl im ate: an his tor icalout l ine. Geogr . Jour . 117, 1951, p. 422—34.

47 . Myres , O . H . : The Si r Rober t Mond Expedi t ion ofthe Egy pt ian Ex plora t ion Society. Geo gr . J . 93, 1939,p. 287—291.

48 . Passarge, S.: Geographi sche Völkerkunde. Ber l in ,1951.

49 . Petrocchi, C.: Richerche prehis tor iche in Cirenaica(Napol i ) 7, 1940, p. 8—34.



35

50 . Philby, H. St. ] . B.: Ru b al-K hal i . G eogr . J . , 82,1933, p. 1—33.

51 . Pietsch, W.: Das A bflußgebiet des Nils . Dr . Phi l .

Disser tat ion, Ber l in, 1910.52 . Picard, L.: Z ur G eologie der Besan Ebene. Ze i t . dt .Paläs t ina Ver . 52, 1929, p. 24—90.

53 . Picard, L.: Zur Geologie des mit t leren Jordantales ,ibid. 55, 1932, p. 199—236.

54 . Ralph, E. K.: Univer s i ty of Pennsylvania r adio-carbon dates I , Science 121, 1955, p. 150—51.

55 . Rathjens, C.: Löss in T r ipol i ta nien , Z. Ges . Erd -kunde Ber l in, 1928, p. 211—228.

56 . Reinhard, A. v.: Glazia lmorphologi sche Studien imwest l ichen und zentralen Kaukasus . Z. Gletscherk. 14, 1925,p. 81—148, 21—235.

57 . S an dfo rd , K . S . and Arkell, A. ].: Palaeol i th i c manand the Ni l e Val l ey in Lower Egypt . Univ . Chicago Or ientInst. 46, 1939.

58 . Schaeffer, C. F. A.: Enkomi-Alas ia. Par is , 1952.59 . Schultz, A.: Morphologische Studien in der ös t l ichen

Kara-kum Wüs te . Z . Geomorphologie 3 , 1927, p . 249—294.60 . Shukri , N. M. and Azer, N.: The mineralogy of

Pl iocene and more recent sediments in the Fayum. Bul l .Inst. Desert II, 1952, p. 10.

61 . Stekelis, M. an d Haas, G.: The Abu Usba Cave (MountCarmel) . Is rael Explorat ion Jour . I I , i , 1952, p. 15—47.

62 . Suess, H. £ . : U. S . Geological Survey r adiocarbon

dates I. Science 120, 1954, p. 467—73.63 . Stein, A.: The Indo- I r anian Border l ands . Jour . Roy.

Anthropol . Ins t . 64, 1934, p. 179—202.64 . Taskin, G. A.: The f a l li ng l evel of t he Caspian Sea

in relat ion to Soviet , economy. Geogr . Rev. 44, 1954,p. 508—527.

65 . Thesiger, W.: A camel journey to Tibes t i , Geogr .Jour . 94, 1939, p. 433—446.

66 . Tolstov, S. P. , Kess, A. S. an d Shdanko, T . A . : G e-schichte des Sarykamishsees im Mit telal ter ( in Russ ian) .Nachr . Akad. Wiss . USSR geogr . Ser . 1954, 1, p. 41—50.

67 . Wagner , A .: K l i m as ch w an k u n g en u n d K l i m aän d e -rungen. D ie Wissenschaf t (Braunschweig) 92, 1941.

68 . Werdecker , ] .: Beobachtungen in den HochländernÄthiopiens auf e iner Rei se 1953/54. Erdkund e IX, 1955,p. 305—17.

69 . Woolley, L.: Excavat ions a t Ur . London, 1954.

70 . Wright , H.E., Jr.: The geological set t ing of fourprehis tor ic s i tes in North we stern I raq . Bul l . Am er . SchoolsOrient . Res . 128, 1952, p. 11—24.

S O Z I A L G R U P P E N A L S F O R S C H U N G S G E G E N S T A N D

D E R G E O G R A P H I E

Gedan ken zur Systematik der Anthrop ogeog raphie

Helmut Hahn

Social groups as sub ects of research in geography

Summary : G e r m an g eo g r ap h y h a s r eco g n i s ed o n l y r e la -t ive ly l a t e the bear ing that socia l groups have on theshap ing of t he cul tura l l andscape. D ur in g the l as t few-d ecad es , h o w ev e r , a g o o d d ea l o f p r e li m i n a r y w o r k h a sbeen done, and i t s eems that t he t ime has now come togive socia l geo grap hy i ts l ogica l p l ace wi th in the f ie ld ofgeography as a subject i n t eaching and r esearch . A br i efr ev i ew o f t h e d ev e l o p m en t o f h u m a n g eo g r a p h y s in ceRatêl and an analys i s of the co nten t and t ask of socia lgeography shows that i t i s by no means ident i ca l wi thhuman geography as i s c l a imed here and there in morerecent publ i ca t ions .

Like a l l geographical r esearch socia l geography mus tbe pur sed analy t i ca l ly and synthet i ca l ly . S ince man shapesthe cul tura l l andscape a lmos t exclus ively merely as a mem-

b e r o f a g r o u p , a s a so c i a l b e i n g , s y n t h e t i c h u m a n g eo -g r ap h y can , h o w ev e r , o n l y b e p u r s u ed i n t h e fo r m o fsocia l geo grap hy. T he es t abl i shed branch es of analy t ica lh u m an g eo g r ap h y , i . e . p o p u l a t i o n , s e t t l em en t , e c o n o m i ca n d c o m m e r ci a l g e o g r a p h y , g e o g r a p h y o f c o m m u n i c a -t ions and in par t i cular phys ica l and psychical anthropo-g eo g r ap h y ( a s u n d e r s t o o d b y Ratêl complete ly r e t a inthe jus t if i ca tion of t he i r exi s t ence bes ide the analy t i ca lsocia l geography. ( In order to make the d i s t inc t ion c l eareri t should r a ther be ca l l ed socio-geography. ) The d i f f er entmani f es t a t ions o f hum an act iv i ty ar e mater i a l ly so c lose lyt i ed to the laws of t he inorga nic and b io logica l sphe res ,tha t we should n ot cut our se lves off f rom the pos s ib i l i t yo f a p en e t r a t i n g an a l y si s b y a o n es i d ed em p h as i s o f t h esocia l sphere .

Socia l geography merely a ims to juxtapose to the cau-sa lly determ ined com plex of phys ica l na tu re a cul tura l

spat i a l comp lex shape d by hum an forces and thu s to f ac i-l i ta t e deep er penet r a t io n in to the in t er r e l a t ionship exi s t ingwi th in the cul tura l l andscape (habi t a t an d h u m an s o c i e ty ) .

Kürzlich hat H. Overbeck einen Überblick überdie Entwicklung der Anthropogeographie veröf-fentlicht, der in übersichtlicher und klarer Da r-stellung die methodischen und sachlichen Fort-schritte dieses Zweiges der Geographie beleuchtet,aber auch zur Besinnung anregt1 ). Gilt es doch,die Leistungen und Irrtümer von drei Forscher-generationen zu beherzigen, wenn wir selbst be-stehen wollen. Wie notwendig eine solche Besin-nung ist, wird allein schon durch die Tatsac he

bewiesen, daß z. B. die niederländische Geogra-phie eine ganze Gene ratio n früher als die deutschedie landschaftliche Gestaltungskraft sozialer O rd-

*) Hermann Overbeck, Die Entwicklung der Anthropo-geographie ( insbesondere in Deutschland) sei t der Jahr-hun der tw end e und ihre Bedeutu ng für die geschicht licheLandes for schung, aus : Blä t t er für deutsche Landesge-schichte, 91. Jg . 1954, S. 182— 244.

Auf ein umfangreiches Li teraturve rzeichnis sol l hier ausRaumgründen verz i chte t werden. Die in den Fußnotenzi t i ert en A utoren — insbesondere Overbeck, Winkler,Houston, Hottes un d Lautensach — br ingen Schr if ttums-verzeichnisse, welche die wesent l iche einschlägige Li teraturerfassen.

Documents

Boizer_Late glacial and postglacial climatic variation in the Near East.pdf