Total Visitors

Web statistics

Locations of visitors to this page

Late Middle Paleolithic Bifacial Industries from Sesselfelsgrotte
in Southern Germany

by Jürgen Richter

The layer G stratigraphic complex ("G-Komplex") of Sesselfelsgrotte yielded one of the longest cultural sequences of late Middle Paleolithic bifacial industries (Micoquian) in Central Europe. Information from this sequence permits reconsideration of the internal structure and the dating of the Micoquian and its relationship to the Mousterian (RICHTER 1997). Evidence is presented for an earlier Micoquian stage with almost no Levallois technology and a later stage with Levallois technology, both occuring at the very end of the European Middle Paleolithic.

The Sesselfelsgrotte sequence
Fig. 1: The Sesselfelsgrotte site is situated in the lower Altmühl Valley (Bavaria, Southern Germany), a tributary to the Danube.

Fig. 1: The Sesselfelsgrotte site is situated in the lower Altmühl Valley (Bavaria, Southern Germany), a tributary to the Danube.

The paleolithic cave site of Sesselfelsgrotte is situated in the valley of the lower Altmühl river (Bavaria), a tributary to the Danube (Fig.1). The site is important because of its unique sequence of 22 Middle Paleolithic occupations (Fig.2). Field campaigns at the site were carried out from 1964 to 1977 and, again, in 1981, directed by G.Freund and collaborators (University of Erlangen; FREUND 1975).

Fig. 2: The Sesselfelsgrotte sequence (schematic section) covers the last glacial cycle. The lower layers are dated to the Early Weichselian before the first glacial maximum, the 	"G-Komplex	" and the late middle Paleolithic layer E3 are dated to the period between the first and the second glacial maximum.

Fig. 2: The Sesselfelsgrotte sequence (schematic section) covers the last glacial cycle. The lower layers are dated to the Early Weichselian before the first glacial maximum, the "G-Komplex" and the late middle Paleolithic layer E3 are dated to the period between the first and the second glacial maximum.

About 7 m of sedimentary deposit were excavated. The layers consisted mainly of limestone debris from the roof of the shelter and from the slope above the cave. Eight occupation units were uncovered from the lower part of the sequence ("Untere Schichten"). An early Weichselian date is suggested for these assemblages which are typologically and technologically similar to contemporaneous western European Mousterian industries (WEISSMÜLLER 1995). They can be classified as Mousterian with micro-size tools (assemblages Ses-U-A08 and Ses-U-A07), Charentian/Ferrassie type (assemblages Ses-U-A06 and Ses-U-A05), Charentian/Quina type (assemblage Ses-U-A04), and typical Mousterian (assemblages Ses-U-A03, Ses-U-A02 and Ses-U-A01). About 10.000 stone artefacts, found in the lower layers ("Untere Schichten"), were discarded during ephemeral occupations. These occupations took place under interstadial conditions (oxygen-isotope stade 5c and 5a) with forest and open landscape. Hunting of horses was an important subsistence activity. Only in the uppermost part of the lower layers (layers 3-West to M1), and quite close to the interface to the first glacial maximum (oxygen-isotope stade 4) of the Weichselian glaciation, does glacial fauna like Mammouth occur for the first time.

A series of layers follows upward, containing no archaeological material, but abundant rodent remains (layers L,K,I). They are dated to the first glacial maximum of the Weichselian glaciation (oxygen-isotope stade 4). The rodent bones (remnants of owl pellets) suggest several subsequent stages of environmental change from a steppe landscape towards an arctic tundra landscape.

The overlying "G-Komplex" (layers H, G5, G4a, G3, G2, G1) yielded 13 Mousterian and Micoquian assemblages (Fig.3). Some of them were recovered from virtual living floors (in particular the layers G4 and G2 with several fireplaces). 85.000 stone artefacts from the "G-Komplex" go along with abundant prey remains, mainly from mammouth, rendeer and horse. Men lived in a steppe landscape with some arctic elements, increasing towards the top of the stratigraphic series. The "G-Komplex" is presumed to be part of an evolved stage within the Oerel-Glinde interstadial complex. Men were present here between 50.000 and 40.000 14C-years B.P., based on preliminary radiocarbon dates.

Seperated by an archaeological sterile layer (layer F), the "G-Komplex" is overlain by another late Middle Palaeolithic horizon (layer E3). On top are loessy deposits of the second glacial maximum of the Weichselian and another two archaeological horizons with several late Upper Paleolithic and Late Paleolithic assemblages.

The "G-Komplex" problem

The study of the "G-Komplex" assemblages posed new questions (Table 1): A problem occurs in that most inventories can be attributed equally well to different "Micoquian" variants (if classification is derived from bifacial "type tools"), as well as to specific "Mousterian" variants (if classification is based on unifacial tool counts). "Micoquian" and "Mousterian" turn out to be multiple, interlaced phenomena, and not clearly seperated cultural units in time and space (Fig. 4).

Fig. 4a: General "Mousterian" component of the "G-Complex"

Fig. 4a

Fig. 4b: Upper Palaeolithic component of the "G-Complex"

Fig. 4b

Fig. 4c: "Microlithic" componentof the "G-Complex"

Fig. 4c

Fig. 4d: Bifacial component of the "G-Complex"

Fig. 4d

Fig. 4 a-d: The stone artefact assemblages of the "G-Komplex" are characterised by four components: A general "Mousterian" component, a component with upper palaeolithic types, a "microlithic" component, and a bifacial component.

Which parameters are responsible for the main characteristics of such stone artefact assemblages? The observation of changing raw material procurement patterns revealed four different occupation cycles, being also coherent in terms of technology and typology (Fig.5). The cycles start with small assemblages of broad spectrum raw material procurement ("Initialinventare"). The cycles end with mostly larger assemblages ("Konsekutivinventare") of more specialized raw material procurement confined to few resources. "Initialinventare" might originate from the beginning reconnaissance and exploitation of a region (the Altmühl valley). "Konsekutivinventare" document a specialized exploitation of resources and might arise from a time when people had already been present in the region for weeks or months.

Modes of stone artefact production

Characteristic modes of stone artefact production (TURQ 1992; BOEDA 1995) are: During Cycle 1 (layer I/H with assemblages Ses-G-A13 to Ses-G-A10) a Quina method of artefact production, during Cycle 2 a Quina and a Levallois method (layer G4 with assemblages Ses-G-A08 and Ses-G-A09), during Cycle 3 the centripetal-recurrent Levallois method (layers G3 and G2 with assemblages Ses-G-A07 to Ses-G-A04), and during Cycle 4 the parallel-recurrent Levallois method (layer G2/G1 with assemblages Ses-G-A03 to Ses-G-A01).

Typological variability

At the beginning of the cycles bifacial tools tend to be rare: at the end they tend to be more abundant. Corresponding change can be observed between the unifacial tools within the cycles. By increasing tool numbers, the Denticulate/other tools ratio is changed such that the Denticulate percentage decreases. Analysis of modification stages shows that bifacial tools can be subject to multiple reworking. Single forms, double forms and reduction forms can be recognized, as well as their relations to each other, should be known before interpreting formal tool counts.

As a result, the relationship between Mousterian and Micoquian and their position within the Weichselian chronology must be reconsidered. In order to characterize the specific relationship between conventional Mousterian and Micoquian, a new term is proposed as a typological and cultural unit: "Mousterian with a Micoquian option" (M.M.O.). "M.M.O." should not necessarily replace "Micoquian" as a term, but labels a new understanding of the conventional term "Micoquian". Presumably, the same people were the authors of "Micoquian" assemblages as well as of "Mousterian" assemblages within the "G-Komplex".

A new chronological model

The M.M.O.-A begins (simultaneous to the Quina horizon of Combe-Grenal) just after the end of the first glacial maximum and, thus, falls within an interstadial complex between 40.000 and 50.000 B.P. Stone artefact production is dominated by the Quina method (M.M.O.-A1) and by other non-Levallois methods (M.M.O.-A2). Larger assemblages contain bifacial tools (conventionally attributed to the Bockstein type of inventories; BOSINSKI 1967) such as Micoquian handaxes, simple bifacial backed knives, "Halbkeile", "Faustkeilblätter".

The M.M.O.-B is dated to a late stage of the same interstadial complex and with its latest variant, perhaps, to a subsequent interstade. It was contemporary to the Typical Mousterian/Denticulate Mousterian stratigraphic complex and the MTA at Combe Grenal. It is characterized by the exclusive application of several Levallois methods. The small assemblages resemble the Denticulate Mousterian (or Kartstein- type inventories), the larger assemblages contain type tools of the Klausennische- and Kxnigsaue-type inventories (convergent bifacial backed knives, bifacial scrapers, "Halbkeile", "Faustkeilblxtter"), and many "microlithic" elements. The Sesselfelsgrotte occupation cycles 3 and 4 are attributed to the M.M.O.-B (M.M.O.-B1 with centripetal-recurrent Levallois, M.M.O.-B2 with parallel-recurrent Levallois method). A presumably later variant M.M.O.-B3 with exclusively parallel-recurrent Levallois method and few bifacial tools is not present in the "G-Komplex" of Sesselfelsgrotte, but in Balve-IV. Perhaps some assemblages of the "Altmühl group" (leafpoint industries) may be placed in the same M.M.O.-B3.

The Sesselfelsgrotte "G-Komplex" assemblages have much in common with bifacial industries from other parts of central Europe, and, not mapped here (Fig.6), from the Crimea (GLADILIN 1976, 124, 125; KOLOSOV 1986, 123). Recent research in the Crimea confirmed comparably late datations for middle palaeolithic bifacial industries (CHABAI 1998, CHABAI & MARKS 1998).

The conventional cultural units of the western central European Mousterian ("Moustxrien x denticules" or "Inventartyp Kartstein", "Inventartyp Balve IV") and Micoquian ("Inventartyp Bockstein", "Inventartyp Klausennische", "Inventartyp Schambach", "Inventartyp Kxnigsaue", perhaps also "Inventartyp Rxrshain" and "Altmühlgruppe") are now suggested to be parts of the same system. The historic reality can possibly be understood in terms of an early Micoquian and a later Micoquian (M.M.O.), combined lasting not much more than 15.000 years and being part of the latest cultural heritage of Neanderthal man.


BOEDA 1994: E.Boeda, Le concept Lxvallois: Variabilitx des mxthodes. Monographie du CRA 9. Paris 1994.

BOSINSKI 1967: G.Bosinski, Die mittelpalxolithischen Funde im westlichen Mitteleuropa. Wien-Kxln-Graz 1967.

CHABAI 1998: V.P. Chabai, The History of Crimean Middle Paleolithic Studies. In: A.E.Marks and V.P.Chabai (eds.), The Middle Paleolithic of Western Crimea, I. ERAUL 84. Liège 1998, 1-15

CHABAI & MARKS 1998: Preliminary synthesis: Middle Paleolithic assemblage variability in Western Crimea. In: A.E.Marks and V.P.Chabai (eds.), The Middle Paleolithic of Western Crimea, I. ERAUL 84. Liège 1998, 355-368.

FREUND 1975: G.Freund, Zum Stand der Ausgrabungen in der Sesselfelsgrotte im Unteren Altmühltal. In: Ausgrabungen in Deutschland. Monographien des Rxmisch-Germanischen Zentralmuseums I/1. Mainz 1976, 25-41.

GLADILIN 1976: V.N.Gladilin, Problems of the Early Paleolithic in Eastern Europe (in Russian). Kiev 1976.

KOLOSOV 1986: Ju.G.Kolosov, Ak-Kaya Mousterian Culture (in Russian). Kiev 1986.

RICHTER 1997: J.Richter, Der G-Schichten-Komplex der Sesselfelsgrotte. Zum Verständnis des Micoquien. Quartxr-Bibliothek, Band 7. Saarbrücken 1997.

TURQ 1992: A.Turq, Raw material and technological studies of the Quina Mousterian in the Perigord. In: H.Dibble and P.Mellars (ed.), The middle palaeolithic: adaptation, behaviour and variability. University Museum Monographs 78. Pennsylvania 1992.

WEISSMÜLLER 1995: W.Weißmüller, Die Silexartefakte der Unteren Schichten der Sesselfelsgrotte. Ein Beitrag zum Problem des Mousterien. Quartxr-Bibliothek, Band 6. Saarbrücken 1995.


iII = Index of standard Mousterian tools

iIII = Index of "Upper Paleolithic" tools

iIV = Index of denticulated and notched tools

iBif = Index of bifaciallly retouched tools

iR = Index of all scrapers

iCh = Index of Charentian scrapers

N = Total number of tools


Table 1: "G-Komplex": ratios of important groups of tools.

  • iII
  • iIII
  • iIV
  • iBif
  • iR
  • iCh
  • N
  • A01
  • 35.90
  • 12.82
  • 27.88
  • 17.95
  • 34.62
  • 17.95
  • 312
  • A02
  • 36.61
  • 11.48
  • 26.78
  • 21.86
  • 34.42
  • 13.11
  • 183
  • A03
  • 36.14
  • 12.05
  • 31.33
  • 16.87
  • 36.14
  • 16.87
  • 83
  • A04
  • 26.00
  • 16.00
  • 32.00
  • 18.00
  • 23.00
  • 5.00
  • 100
  • A05
  • 41.03
  • 8.97
  • 26.28
  • 18.59
  • 39.75
  • 21.79
  • 156
  • A06
  • 32.09
  • 14.64
  • 28.04
  • 17.13
  • 30.22
  • 13.08
  • 321
  • A07
  • 32.31
  • 6.15
  • 43.08
  • 18.46
  • 32.31
  • 12.31
  • 65
  • A08
  • 31.84
  • 7.62
  • 27.80
  • 30.04
  • 28.70
  • 14.35
  • 223
  • A09
  • 40.72
  • 8.38
  • 23.35
  • 25.75
  • 39.52
  • 17.37
  • 167
  • A10
  • 55.13
  • 3.85
  • 17.95
  • 23.08
  • 55.13
  • 29.49
  • 78
  • A11
  • 51.11
  • 6.67
  • 31.11
  • 8.89
  • 51.11
  • 31.11
  • 45
  • A12
  • 53.85
  • 3.85
  • 34.62
  • 7.69
  • 53.87
  • 30.77
  • 26
  • A13
  • 0
  • 0
  • 0
  • 0
  • 0
  • 0
  • 0