J. Šarić, Paleolithic and Mesolithic Finds From Profile of the Zemun Loess Starinar LVIII/2008, 9-27 JOSIP ŠARIĆ, Institute of Archaeology, Belgrade PALEOLITHIC AND MESOLITHIC FINDS FROM PROFILE OF THE ZEMUN LOESS UDC Received: January 15, 2009 DOI Accepted: May 04, 2009 Short communication Abstract: Segment of the material from these two sites was published already in 1984. Owing to circumstances two new Paleolithic sites discovered in the territory of Serbia in recent times have made possible placing of the finds from the sites ‘Ekonomija 13 maj’ and Beljarica in the new context. This work, by revising already published material and also by presenting the artifacts discovered in the meantime, expands the data base related to the human settlements on the fringes of the Pannonian basin, i.e. in the territory of present-day Serbia during Middle and Late Paleolithic but also during Mesolithic period. Key words: Chipped artifacts, Middle Paleolithic, Late Paleolithic, Mesolithic, raw materials, chert, quartzite, handaxes, hand points, shouldered projectiles/points, sidescrapers, geometric microliths. Introduction he first indirect evidence for the existence of scull with the characteristics of the Neanderthal man the Paleolithic stations in the Belgrade city ter- as interpreted by N. Županić.1 This fossil skull, which ritory were provided in the works of H. Breuil, was found near the former Vidin kapija (Vidin Gate) in T a layer together with the teeth of Elephas antiquus in G. Mac Curdy, H. Obermaier and J. Skutil quoting the data about the existence of caves with the Auri- 1919, has been lost in the meantime, so the mentioned gnacian finds but on the other hand S. Brodar denied claims could not be verified today. The first chipped and doubted it in his work from 1954. The finds of the stone artifacts ascribed to the Paleolithic period were Pleistocene fauna in a few cave entrances and one rock the chance finds from the Sava river bank few hundred shelter on the southwestern slopes of Banovo Brdo, meters far from the layers with the Pleistocene fauna at between Čukarica and Žarkovo discovered in 1955 ac- Banovo Brdo and they were published in 1958.2 There tualized at that time once again the assumptions con- are also some unreliable data impossible to verify that cerning the settling of the Paleolithic populations in the vicinity of the present day Belgrade. To the men- 1 Gavela 1956. tioned assumptions was added the story of the human 2 Gavela 1958. * The article results from the project: Lepenski Vir Culture: Cultural processes and transformations during 9. to 6. Millennium BC (no 147009 D) funded by the Ministry of Science and Technological Development of the Republic of Serbia. Josip Šarić, e-mail: [email protected] 9 J. Šarić, Paleolithic and Mesolithic Finds From Profile of the Zemun Loess Starinar LVIII/2008, 9-27 the remains of Elephas cf. primigenius Blumb., frag- results were published in 2006.20 The most recent ex- ments of small singed(?) bones, resin, charcoal and cavations at the Petrovaradin fortress near Novi Sad21 triangular wedges of lajtovac3 have been found under that brought to light rich Mousterian (and somewhat the foundations of the palace ‘Albanija’ in Terazije in less abundant Late Paleolithic) assemblage of chipped Belgrade and indicating the existence of the Paleo- stone artifacts as well as already mentioned Mousteri- lithic, according to the author, probably the Gravettian an endscraper from Rušanj are the best proofs that we settlement.4 In the period between 1958 and 1984, i.e. should seriously count on the finds from the Middle for almost thirty years, there have been no new data Paleolithic stations in the Pannonia region in Serbia. about the Paleolithic finds in the territory of Belgrade. These finds are at the same time obvious confirmation However, in that year (1984) was published a text con- that the assumption about the existence of the Mous- cerning the segment of a large assemblage of chipped terian in the Zemun loess suggested already in 198422 stone artifacts gathered on the Danube river bank few was absolutely justified. The most recent results of the kilometers upstream of Zemun.5 Some articles about investigations of the Paleolithic in the territory of Ser- the excavations in the vicinity of Vršac and the test bia are the works of B. Mihailović and D. Mihailović trench excavations in the Smolućka pećina were pub- concerning the finds from Šalitrena pećina published lished in 1984 but also in 19866 and in the years fol- in 2007 and 2008 and the finds from the cave Baranica lowing the publishing of these texts the investigations also published in 2008.23 of the Paleolithic sites throughout Serbia have been The new information resulting from the inves- conducted, so there were more published texts with tigations conducted in the previous years, some new very significant results. The results of excavations in finds from the mentioned site in the vicinity of Zemun the Šalitrena pećina were published in 19857 and from as well as the need to revise the published material 1985 to 1987 were published the results of investiga- from Zemun are the reasons for writing this treatise. tions in the Smolućka pećina.8 New discoveries of the chipped stone artifacts from the Vršac Paleolithic sites Zemun loess plateau and location of the sites were published in 1989,9 and an accidental but very im- ‘Ekonomija 13. maj’ and Beljarica portant find of the Mousterian endscraper from Rušanj The mountain ridge of Fruška Gora was a barrier, was published in 1990.10 Kaludjerović published com- which caused the creation of the Srem loess plateau as prehensive work in 1991 concerning the Paleolithic a result of accumulation of the loess sand. The Zemun in the light of more recent investigations11 while the loess plateau is situated in the south east periphery of results of excavations of the Paleolithic sites in the vi- Srem, bordering on the Danube in the northeast, on cinity of Vršac were published in the following year.12 the Sava in the southeast and the line connecting Stari The data concerning excavations in Drenaića pećina Banovci and Boljevci in the west. The loess plateau on Medvednik,13 at Kremenac near Niš,14 in Mirilovska consists of alternating layers of light yellow loess and pećina,15 in the cave habitations in the Knjaževac ter- dark layers of the buried i.e. fossil soil. ritory16 as well as the results of surveying in the Soko banja valley17 were published in 1996. The synthetic 3 Lajtovac is a local name for a Miocene limestone occurs in vicinity work about the investigations of the Paleolithic con- of Belgrade. ducted so far in the eastern Serbia was published by D. 4 Stevanović 1977. Mihailović, Lj. Djuričić and Z. Kaludjerović in 1997.18 5 Šarić 1984. 6 Kaluđerović 1984; Radovanović 1986. In 2001 after the revision of the osteological material 7 Jež i Kaluđerović 1985. in the collection of the Institute for Regional Geology 8 Kaluđerović 1985, 1986, 1987. 9 Joanovič 1989. and Paleontology of the Faculty of Mining and Geol- 10 Kaluđerović 1990. ogy in Belgrade, the academic public was informed 11 Kaluđerović 1991. 12 Mihailović 1992. about a very important anthropological find. It is the 13 Kaluđerović i Jež 1996. fragment of right half of the mandible discovered some 14 Kaluđerović i Đurić 1996. time ago by Prof. V. D. Laskarev but never published. 15 Đuričić 1996. 16 Sladić i Jovanović 1996. On the basis of the morphological characteristics M. 17 Kaluđerović 1996. Roksandić and V. Dimitrijević came to conclusion that 18 Mihailović, Đuričić i Kaluđerović 1997. 19 Roksandić i Dimitrijević 2001. this was an individual belonging to the Upper Paleo- 20 Mihailović 2006b. lithic population.19 The investigations in Hadži Pro- 21 Mihailović 2006a. 22 Šarić 1984. danova pećina near Ivanjica suggested the existence 23 Mihailović B. and Mihailović D. 2007; Mihailović B 2008; Mihai- of Middle and Upper Paleolithic industry and these lović D. 2008. 10 J. Šarić, Paleolithic and Mesolithic Finds From Profile of the Zemun Loess Starinar LVIII/2008, 9-27 B. Laskarov is of the opinion that loess is the creation of interglacial period while A. Penk and B. Bula relate the creation of the loess plateau and sandy terrains to the glacial period. This assumption is cor- roborated by the field investigations of B. Ž. Milojević as well as of D. Mihajlović-Matić.24 Zeremski, Maruščak and Butrim define the ex- istence of two segments within the Zemun loess pla- teau – the bottom one, consisting of river-marshy sedi- ments dating from the Riss glaciation and the top one consisting of four layers of loess and four layers of fossil soils dating from the Riss and Würm periods. The problem of defining relative and absolute chronol- ogy of the Zemun loess plateau is a very complex one as it is confirmed by striking difference in number of layers of loess or fossil soils distinguished by differ- ent authors. In his first works about the Zemun loess plateau Gorjanović identified four layers of loess and three layers of fossil soils while V. Laskarov distin- guished five layers of loess and four layers of fossil soils.25 J. M. Marković identified eight layers of loess and eight layers of fossil soils on a profile near Bata- jnica, Maruščak and Butrim identified 10 layers of loess and 12 layers of fossil soils at the same location while Rakić and al. recognized four layers of loess and four layers of fossil soils (Fig. 1).26 Fig.1. Profile of the Zemun loess at Beljarica. The main reason for this problematic synchroni- Along the entire Danube bank from Zemun almost zation lies in the subjective stratigraphic assessments to Banovci the loess plateau is over 20 meters high as well as in the scarce paleontological material, which, (after the text of V.