View metadata, citation and similar papers at core.ac.uk brought to you by CORE
provided by UNL | Libraries
University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln
Transactions of the Nebraska Academy of Sciences and Affiliated Societies Nebraska Academy of Sciences
1978
The Classical European Glacial Stages: Correlation with Deep-Sea Sediments
George Kukla Columbia University
Follow this and additional works at: https://digitalcommons.unl.edu/tnas
Part of the Life Sciences Commons
Kukla, George, "The Classical European Glacial Stages: Correlation with Deep-Sea Sediments" (1978). Transactions of the Nebraska Academy of Sciences and Affiliated Societies. 334. https://digitalcommons.unl.edu/tnas/334
This Article is brought to you for free and open access by the Nebraska Academy of Sciences at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Transactions of the Nebraska Academy of Sciences and Affiliated Societiesy b an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. Transactions of the Nebraska Academy of Sciences- Volume VI, 1978
THE CLASSICAL EUROPEAN GLACIAL STAGES: CORRELATION WITH DEEP-SEA SEDIMENTS
GEORGE KUKLA
Lamont-Doherty Geological Observatory Of Columbia University Palisades, New York 10964
Four glacials and three interglacials, recognized by classical sea record revealed that this is not the case. In order to learn Alpine and North-European subdivisions of the Pleistocene, were cor the reasons for the discrepancy, we identified the type units related with continuous oxygen-isotope records from the oceans using and their boundaries at the type localities, correlated them loess sections and terraces as a link (Fig. 15). It was found that the Alpine "glacial" stages are represented by sediments formed during with the standard stratigraphic system, and tried to recognize both glacial and interglacial climates, that the classical Alpine "inter their true climatic characteristics. The record obtained from glacial" stages do not represent episodes of interglacial climate but the sea bottom, a region of practically uninterrupted deposi probably intervals of accelerated crustal movements, and that the tion which is the most complete system available, was used as physical evidence on which the North-European classical subdivision is a standard. based is misinterpreted due to lengthy gaps in the record. It is recommended to discontinue the use of classical terminol ogy in all interregional correlations and to base the chronostratigraphic subdivision of Pleistocene on the 0 18 record of deep-sea sediments. THE OXYGEN ISOTOPE RECORD It is believed that, as in Europe, the loesses and terraces in Ne braska will eventually provide the link needed for correlation of deep The ratio of oxygen isotopes 0 18 and 0 16 in foramini sea sediments with the classical American glacial stages. feral tests provides the most complete record of global climatic t t t changes known to date (Emiliani, 1955). It is ideally suited for a standard with which all other records can be compared. The isotopic variations in benthic foraminifera is controlled INTRODUCTION predominantly by fluctuations in the total amount of ice deposited on land (Shackleton and Opdyke, 1973, 1976). This There are two classical systems of the land-based Pleisto record is almost independent of the geographic location of cene in Europe namely the Alpine and the North-European. studied cores. It was also shown that the benthic 0 18 record Table I summarizes the principal information on their units. is closely paralleled by that of planktonic foraminifers, first Recently a new stratigraphy of Pleistocene has emerged, classi described by Emiliani (1955, 1958, 1966, 1972; Emiliani, fying deep-sea sediments after the isotopic composition of et al., 1961). Emiliani used it to subdivide deep-sea sediments oxygen in the calcareous shells of foraminifers. Because the in several Caribbean and North Atlantic cores into numbered latter is to a significant degree controlled by the volume of stages. His system, commonly referred to as the "oxygen iso land·locked ice, the deep-sea record closely relates to Pleisto tope stratigraphy" or "delta 0 18 stratigraphy," is now widely cene deposits on land. used by deep-sea researchers.
The three Pleistocene subdivisions are climatostrati Piston core V28-238 can conveniently serve as a type graphic systems based on sequences of strata or geomorphic locality of the 0 18 /016 record (Shackleton and Opdyke, features with specific climatic signatures. As the gross changes 1973; Geitzenauer, et aI., 1976). It was taken from the equa ~f climate are globally synchronous, the classifications should torial Pacific at Long. 01 °Ol'N and Lat. 1600 29'E, from a e rather similar. water depth of 3120m. It is deposited at the Lamont-Doherty Geological Observatory of Columbia University, Palisades, Correlation of the classical land systems with the deep- New York. As seen in Figure 2 the upper 14m of the core are
57 TABLE I
Type Units of the Classical Subdivisions of Pleistocene
Interpreted System Interval Climate Type Unit Type Locality Reference
Alpine Postglacial Temperate Wurm Glacial Low Terrace Riss-Wurm Interglacial Riss Glacial High Terrace PenckI and Mindel-Riss In terglacial Bruckner (1909) Mindel Glacial Younger Cover Iller-Lech Platte, Gravel Germany Gunz-Mindel Interglacial Gunz Glacial Older Cover Gravel 1 Donau-Gunz In terglacial Eberl (1930) Donau Glacial Donau Gravel Eberl (1930) Biber-Donau In terglacial Schaeffer (1953) Biber Glacial Highland Gravel Schaeffer (1953) ......
North European Flandrian Temperate Marine and Brackish Netherlands Dubois (1925) Brackish deposits Weichsel Glacial End-Moraines North Germany, Poland Keilhack (1926) Eem In terglacial Marine and Vicinity of Amersfoort, Madsen (1929) Brackish deposits Netherlands Warthe Glacial End Moraines Central Germany Keilhack (1926) Holstein In terglacial Marine deposits Schleswig-Holstein Penck (1922) Elster Glacial Glacial-lacustrine Central Germany Keilhack (1926) Cromer In terglacial Marine and East Anglia, G.B. Reid (1882) Brackish deposits
••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• •• , Of
American Recent Temperate Lake, alluvial and North-Central USA For American usage: colluvial deposits Kay and Leighton on top of Wisconsin (1933) Wisconsin Glacial End Moraines Eastern Wisconsin Geikie (1894) (originally East Wisconsin) Sangamon In terglacial Black mud with Vicinity of Rochester, Leverett (1898) (alluvial clay?), Sangamon County, gununy clay (accre Illinois tion gley?), sand, gravel and peat, ferreto (red soil).
58 T ABLE I (Continued)
Interpreted Interval Climate Type Unit Type Locality Reference ---System -- Illinoian Glacial Tills Southwestern Illinois Chamberlin (1896) (originally Illinois) Yarmouth In terglacial Sand, clay and Well in Yarmouth, Leverett (1898) peat with twigs exposure in West and bones of hare Point, Iowa on till Kansan Glacial Tills Northeastern Kansas Geikie (1894) Afton Temperate Gravel and sand, Exposures in Afton Chamberlin (1895) muck with vegetation Junction, Iowa (= organodetritic silts) Nebraskan Glacial Tills Exposures in the Shimek (1909) vicinity of Omaha, Nebraska
subdivided into 22 oxygen isotopic stages numbered in order such as faunal assemblages (Ericson and Wollin, 1968; Imbrie of increasing age. Odd-numbered units are relatively deficient and Kipp, 1971), coccoliths (Ruddiman and McIntyre, 1973, in 0 18 and correlate with cold intervals characterized by in 1976), ice rafted detritus (Kent, et aI., 1971), and total car creased volume of land-based ice. The boundaries, defined by bonate content (Gardner and Hays, 1976; Kellogg, 1976). depth in the core, are placed near the level of fastest change of the isotopic ratio, mostly halfway between the neighbor In theory the 0 18 ratio in forams should provide an ac ing maxima and minima. Boundaries separating especially curate measure of past ice volume. However, the resolution of pronounced isotopic maxima from exceptionally pronounced the method is limited by a number of factors such as delay in minima were called "terminations" by Broecker and Van Donk ocean mixing, varying sedimentation rates, different sampling (\970). The segments bounded by two consecutive termina intervals, burrowing, redeposition of shells into younger sedi tions and composed of commonly two or up to four basic ment, selective dissolution, etc. (Shackleton and Opdyke, stages represent units of higher rank labeled "glacial cycles." 1976). As a result the precision of the method greatly de Terminations are numbered by Roman numerals and glacial creases with age, and the amplitudes of correlative peaks in cycles by capital letters in order of increasing age. different cores diverge to a significant degree. At present, the shapes of 0 18 curves, rather than the absolute values of iso Stage 5 was further subdivided by Shackleton (1969) topic ratios, are used for stratigraphic interpretations. Records into five substages labeled 5a to 5e in order of increasing age. of several cores must be compared before confident climato Substage 5e has the lowest isotopic ratio of the five. It corres stratigraphic correlations are made. Ponds to the minimum volume of ice and is considered to represent the last interglacial. Similarly, stage 7 was subdivided The reversal of the earth's magnetic field, K/ Ar dated in mto substages 7a, 7b and 7c (Ninkovitch and Shackleton, land-based volcanics to about 0.7 million years (Dalrymple, 1975). Cores where the subdivisions were described are listed 1972), was recognized in a number of deep-sea cores (Opdyke, in Table II. The last termination, within the limits of dating 1972). In V28-238 it was located at the base of isotopic stage preciSion, correlates with the last deglaciation (Ruddiman and 19. In V28-239 it was found below termination IX in isotopic MCintyre, 1973). Correspondingly earlier terminations are stage 20. Undoubtedly the reversal occurred in the vicinity of mterpreted as representing earlier deglaciations. The climatic stage 19/20 boundary. Younger sections of deep-sea sediments Interpretation of 0 18 record is abundantly supported by inde are extenSively dated by radiocarbon (Ericsor., et aI., 1961; pendent climatic indicators present in deep-sea sediments Broecker and Van Donk, 1970; Ruddiman and Glover, 1972).
59 0\ 0 TABLE II
Key Sites of Oxygen Isotope Stratigraphy
Core Ocean or Sea Latitude Longitude Water Depth (m) Length (em) 0018 Stages Remarks References Place of Deposit Site No.
AI72-6 Caribbean 14°59'N 68°57'W 4290 935 1-14 Orig. definition Emiliani,1955 Lamont 4 of 0 18 stages 1-14
P6304-8 Caribbean 14°59'N 69°20'W 3927 1054 1-11 Defmition of 15-17; Miami 5 Radiometric ages of Rona-Emiliani, P6304-9 Caribbean 14°57'N 68°55'W 4126 1429 1-17 stages 2-7 1969 Miami 3
V12-122 Caribbean 17°00'N 74°23'W 2800 1095 1-14 Orig. definition Broecker and Lamont 6 of terminations & van Donk, 1970 Glacial Cycles
V16-205 Atlantic 15°24'N 43°24'W 4043 1232 1-14 Orig. definition van Donk, 1976 Lamont 2 of stages 2341
V28-238 C. Pacific 01°01'N 1600 29'E 3120 1609 1-22 Orig. defmition Shackleton & Lamont 8 of stages 17-22 pro- Opdyke, 1973 posed 0 18 type core
V28-239 C. Pacific 03°15'W 159°11'E 3490 2102 1-23->- Detailed 0 18 record Shackleton & Lamont 9 for ± 2 m.y. Opdyke, 1973
RCII-120 S. Pacific 43°31'S 79°52'E 3135 954 1-9 Cycle analysis and Lamont 11 correlation with Hayes et al., E49-l8 S. Pacific 46°03'S 900 09'E 3256 1459 5-13 astronomic chronol. 1976 Lamont 10
K708-7 N. Atlantic 53°56'N 24°05'W 3502 1280 not defmed Orig. names of Ruddiman and Lamont Glacial cycles A-H McIntyre, 1976 in deep-sea sed.
Sw Exp-58 Eq Pacific 6°44'N 129°29'W 4440 991 not defined First downward Arrhenius, 1952 Uppsala 7 numbered stages in a deep-sea core. Concept followed by Emiliani. "11 10
w~ w
90 120 150 180 150 120 90
Figure 1. Location of key Pleistocene sites with continuous stratigraphic records. 1-11: deep sea cores as listed in Table II; 12: deep-sea core V28-179; 18: Tenaghi Phillip on in Macedonia; 19: DunafOldvar; 20: Paks; 21: Red Hill; 22: Krems; 23: Karlich; 24: Grande Pile; 25: Padul; 26: Coral reefs at Barbados; 27: Camp
0\ Century; 28: Sabana de Bogota; 29: Buzzard Roost; 30: Clear Lake; 31: Coral reefs in New Guinea. 8 0 '°to PDS ..'"w I- w '" V28·238 V28·239 V16·205 .. -0 ., ·2,. 0 ., II :~_ I=' I 2 -I==-A = 0 --=='==A= '. • 128:~_ ~_Z•
_1,700 j
V28 -179 -1,850 -t=-- il -+;::~:~Q~ ',1,] L.,o'I_-' II (41) 4-,
f--______'''I_f __-II-_? 2.4~O - ,I ' "1J
3,000 -
3.400 -
Figure 2. Key oxygen isotope records. Location of the cores in Figure 1. Fonnal numbering of 8018 stages in Arabic numerals after Emiliani (1955, 1966) and after Shackleton and Opdyke (1973, 77), infonnal below stage 22 after Van Donk (1976), terminations in Roman numerals after Broecker and Van Donk (1970), major cycles in capital letters after Ruddiman and McIntyre (1976). Faunal and floral markers mainly after Geitzenauer, et al. (1976), Thierstein, et al. (1977), and Briskin and Berggren (1975) showing: A: beginning dominance of Emiliania huxleyi over Gephyrocapsa caribbeanica in transitional waters; B: first appearance of coccolith Emiliania huxleyi; C: last appearance of radiolaria Stylatractus universus; D: last appearance of coccolith Pseudoemiliania lacunosa; E: appearance of abundant foraminifer Sphaeroidinella dehiscen; F: last appearance of foraminifer Globigerinoides /istulosa; G: last discoasters; H: emergence of Globorotalia menardii-tumifll group and of the foraminifer Globorotalia truncatulinoides; I: last appearance of Globigerinoides obliqua; J: last appearance of foraminifer Globorotalia exilis; Y-Q: Faunal zones of Ericson defmed by the lack of the Globorotalia menardii-tumifll group in the Atlantic. Wann oxygen isotope stages stippled, estimated age of terminations from Table III. Normal polarity in black, reversed blank-
62 ium-decay series were used to calibrate uneven sedimen eleven terminations, I to XI, were completed in the past one uran rates back to approximately 0.3 million years (Rosholt, million years from the Jaramillo normal polarity event. wHO n a1 1961; Rona and Emiliani, 1969; Ku and Broecker, e~66)'. The ages derived by this method are subject to large The ice-volume oscillations as recorded by 0 18 ratio : . tematic uncertainty. Oxygen isotope variations in corals within each glaCial cycle seem generally to follow a sawtooth )ysde it possible to accurately date, by the uranium/thorium pattern, progressing from an early minimum to a late maxi ma 18 . ad the 0 peaks of stages 5, 7 and 9 1ll coral reefs at mum. Apparently the glaciers have grown in steps interrupted me th ' Barbados (Shackleton and Matthews, 1977). The U/Th dates by temporary regressions. The 0 18 record does not show corals are considered accurate because the aragonitic where the ice was deposited and where the retreats took place. O;eleta of corals are almost impermeable to radioactive iso It remains unclear whether the changes of ice volume pro :opes and maintain a closed geochemical system. Age deter ceeded in phase or out of phase between the two hemispheres minations of corals were extensively cross-checked by repeti or between the North American and Euro-Asian continents. tive sampling at a number of localities by several laboratories_ The number of secondary isotope fluctuations recog Table III summarizes the recently published estimates of nized within each glacial cycle, depending on the sedimenta the age of terminations within the last one million years, ob tion rate and sampling interval of the particular core studied, tained by interpolations to the Brunhes/Matuyama reversal varies commonly between 4 and 6. In each of the well-known and by radiometric dating of corals. last two glacial cycles, Band C, five secondary oscillations took place. A typical 0 18 curve of cycle B is shown in Figure The 0 18 record shows eight completed glacial cycles, 5. namely B through I, and nine terminations, I through IX in the last epoch of normal polarity which lasted approximately In general it was observed that during the past approxi 0.7 million years. Ten glacial cycles, B to K, separated by mately one million years the general shape and amplitude as
TABLE III
Age of Terminations-Overview of Published Estimates
Broecker Shackleton Ruddiman Hays, Imbrie and Van Donk and Opdyke and McIntyre and Shackleton Termination 1970 Kukla 1970 1973 Kukla 1975 1976 1976 Average (103)
11,000 10,500 13,000 11,000 13,500 10,000 11.5 -1 +2 II 127,000 128,000 128,000 128,000 127,000 127,000 128.0-1 +0 III 225,000 243,000 251,000 245,000 220,000 247,000 240.0-20 +10 IV 300,000 333,000 347,000 335,000 315,000 336,000 330.0-30 +20 V 380,000 370,000 440,000 415,000 380,000 425,000 400.0-30 +40 VI 485,000 502,000 480,000 455,000 480.0-20 +20 VII 600,000 592,000 575,000 515,000 570.0-55 +30 VIII 647,000 635,000 605,000 630.0-25 +20 IX 726,000 706,000 700,000 710.0 -10 +20 X 845,000 782,000 810.0-30 - +35 63 NORTH ATLANTIC M A C E DON I A (j) (j)0 z wW 0 ~ ~ <..91-- FORAMS z P 0 L LEN -- W W 0/0 W 0/0 ~a:-.J rou U I-- 0 -.J I ~ ~ PANGAION 6 C 7 bO SYMVOLON c TII ~-~ 8 0 9 N I , LEKANIS 10 COMPLEX E 11 12 F 13 BOZ DAGH 14 COMPLEX G 15 16 H PHALAKRON 17 Figure 3. Foraminiferal assemblages in North Atlantic core K708-7 after Ruddiman and McIntyre (1976), compared with till pollen record in the Tenaghi-Phillipon peat bog in Macedonia, after Van der Hammen, et al. (1972). Percentage of subtropiC8 and transitional foraminifera in K708-7 and oak pollen in Macedonia shown in black. Forams other than a polar typ~ an' total tree pollen stippled. Position of terminations, correlation with oxygen isotope stages and glacial cycles modified a~e Ruddiman and McIntyre (l976). Recently confirmed by an ao 18 analysis of the core by Shackleton (personal communlC8 tion). 64 + WARSZAWA t + ~ "' ""::> ~'" + + BUDAPEST igure 4. Key loess areas around Brno, Praha and Krems (3) and the maximum extent of Pleistocene glaciers in Europe (stippled). Rivers, which the north European glacial stages were named after, or those with dated terraces are also shown. 1: type local ity of the Alpine glacial stages (crosshatched); 2: type region of the north European glacial stages. well as internal structure of the 0 18 record did not markedly least nine episodes with global climate comparable to the change. This conclusion takes into account that the limited present one, which can be labeled "interglacial." Stage 7 shows :limate history during each one of the glacial cycles C-J is two pronounced warm peaks of interglacial rank. This is a ikely to have displayed about the same degree of complexity characteristic difference from other odd-numbered isotopic s the last glacial cycle B. stages which only allow recognition of single 0 18 peak. Thus the interglacial of stage 7 may have been doubled. Whether Cold stages 2, 6, 10, 12, 16 and 22 were found to be some of the warm stages older than 13 displayed similar two marked on the average by deeper and/or longer-lasting isotopic fold structures is impossible to determine because of insuffi highs than were the remaining cold stages. On the contrary, cient resolution of available cores. cold stage 14 is poorly expressed and points to an exception ally low ice volume. The warm peaks of odd-numbered stages In summary, the isotopic record of deep-sea cores pro II to 21 (except stage 3) all seem to approach a closely similar vides the most accurate hitherto known information on evel, at least within the precision limits of the 0 18 method. Pleistocene climates, unique in its continuity and global hi . d' s III lcates that the global ice volume and sea surface tem- extent. Validity of any other climatostratigraphic system of ~ratures in peak interglacials were similar to those of today Pleistocene must be tested by comparison with the deep-sea ld also that during the past 0.9 million years there were at record. 65 CHARACTERISTIC ENVIRONMENT DEEP SEA CORRELATlON~ SNAILS ",- G LAC IAL INTERGLAC 12 >- I- ~ ~~ ~ ~ ~. '8 (j) C PMG ,.~'" C,)~ t) ~ ~ ...... ,,~ STEP 0 0 S TAG E S '4 z ",,,, ::J c').." ~ ;;: ..... ~ e t::I z FAUNAL w Q <:) ---.J Figure 5. Summarized loess stratigraphy of the last 130,000 years in the area around Praha and Bmo. PMC: polarity of remaneni magnetization, black: normal, stripped: interpreted as reversed (after Kukla and Kocl, 1972). Radiocarbon dates on charC08 and soils after Kllina, et al. (1962) and Kukla (1975), abbreviations of type localities of the soils after Kukla (1975), litho logic symbols same as in Figure 9. Roman numerals designate marklines and terminations. Vertical scale proportional to relative thickness, not time. CU01 of reconstructed environment shows (C) cold, (N) normal, (M) mild loess steppe; barren badlands; (R) sparse marginal grass lands; (T) thick grasslands; (BE) coniferous or short-lived mixed forests on braunerde; (PB) deciduous, long-lasting forests Ol parabraunerde. Cold and normal loess steppe in studied area is characteristic for full glacial (GLAC); parabraunerde for tiJI interglacial (INTGL). Tentative correlation with ~018 record also shown. After Kukla (1975). 66 THE LOESS RECORD eolian deposit of contemporary origin because of its fme net IN CENTRAL EUROPE work of secondary carbonate, which coats the rock detritus and sheets of clay minerals, partly fIlling the dense system of Loess is a yellow-calcareous, porous, wind-blown silt, fine grass rootholes. Stratigraphic relations and micromorphol 'haracteristic of Pleistocene sequences all over the world. Its ogy prove beyond doubt that the calcareous network is syn chickest deposits, several tens of meters thick, accumulated in genetic. A special type of arid weathering called "Loessifica t eriglacial areas along the course of major rivers (Smalley, tion," acting in both the source area as well as in the accumu i975). The bulk of deposits is of glacial age. Largest are in lation zone, is believed responsible for this feature (Ganssen, China (Liu and Chang, 1961) and Ukraine (Veklitch, 1969; 1922; review by Marosi, 1970). Velichko, 1969). The best studied and the most important for Pleistocene stratigraphy are deposits lying closest to for The depositional environment of loess can be fairly well merly glaciated areas of Europe. reconstructed. Extensive, barely or poorly vegetated, season ally dry surfaces exposed to loessification must have existed The advantage of loess sequences is that they are essen in the deflation areas. Sparse patches of grass and sedges in tially continuous. With the help of magnetostratigraphy and the accumulation areas were necessary to pick up the dust and observed parallels in climatic history, they can be directly cor prevent further erosion of the surface. Indeed, pollen of related with deep-sea sediments (Kukla, 1970, 1975). They Gramineae and Cyperaceae were found in the loess by Frenzel are also stratigraphically related to the terraces of rivers flow (1964). Sandy interlayers and wind-moved rock fragments on ing through formerly glaciated areas. In this way a link is pro the slopes prove that winds were occasionally very strong vided for a direct correlation of the deep-sea record with (Zebera, 1949). classical glacial stages. Valuable information on the loess environment is pro The periglacial area around Prague, Bmo and Nitra in vided by gastropodes. These are found largely intact, and there Czechoslovakia, as well as around Krems and Vienna in Austria is no doubt that animals lived at the time of loess deposition. (Fig. 4) was never glaciated. When the Alpine and Fenno scandinavian ice sheets reached their maximum extents, loess The typical autochthonous assemblage of central Euro accumulated in this region, influenced by highly continental pean loesses is dominated by the Pupilla group and called the periglacial climate. During the interglacials the climate was Puppilla fauna (Lozek, 1964). Some of its elements still live of the Atlantic type, similar to the present one, or warmer in the steppes of Europe, while others such as Pupilla loessica and wetter. Deciduous forests spread, leaving behind the are extinct. No place is known today where the living elements characteristic parabraunderde soils. Thus in favorable deposi of Pupilla fauna dwell together. The environment indicated by tional basins, glacial-interglacial cycles were recorded by re Pupilla fauna is sometimes referred to as a "cold loess-steppe" peated alternations of loess and forest soils. The loess is fre (Lozek, 1964, 1969). quently deposited on river terraces behind steep cliffs of abandoned river banks. These were ideal sedimentary traps for Another loess assemblage, dominated by a cold resistant the wind-blown dust. The terraces around Prague were formed species Columella columella, is called Columella fauna. It is by a tributary of the Labe (Elbe) River which flows into the frequently accompanied by Vertigo parcedentata, Vallonia region of Fennoscandinavian glacial deposits. The terraces tennuilabris, or Pupilla loessica. Elements of wet and cold around Brno, Nitra, Krems and Vienna belong to the tribu environments such as Succinea putris, Vertigo arctica, and taries of the Danube River which also drains the Alpine foot Vertigo pseudosubstriata are also present. Elements of this hills where classical Alpine glacial stages were defined by assemblage today live in the Alpine zone of the Tien Shan Penck and Bruckner (1909). Thus, the correlation of loss Mountains of central Asia, under an extremely cold and sequences with glacial deposits in northern Europe and in the highly continental climate. At the paleolithic sites of Veston Alps is possible. ice and Pavlov, the wet Columella fauna is accompanied by lynx, wolverine (Gulo gulo), arctic fox (Alopex lagopus), More than one hundred selected localities, with espe reindeer, mammoth and wooly rhinoceros (Musil, 1959), Cially well developed sequences, have been described in Czech and in Bojnice by lemming (Dicrostonyx torquatus) (Pro~ek osluvakia and Austria during this century (Penck and Bruck and Lozek, 1957). Also related to this assemblage are the frost ner, 1909; Gatzinger, 1936; Sch5nhals, 1951; Zebera, 1949; gley soils indicative of permafrost. Obviously the Columella f IIlk 1954, 1969; Musil, et al., 1955; Prosek and Lozek, fauna was able to persist in a subarctic environment with a 1957; Kukla and Lozek, 1961; Demek and Kukla, 1969; very low winter temperature, sometimes referred to as loess KUkla, 1975). The two most important key localities are tundra. Cerveny Kopec (Red Hill) in Bmo and the Kremser Schies statte in Krems. The Striata fauna is climatically the most demanding assemblage known from loess (Lozek, 1964, 1969). It is repre Glacial loess in central Europe differs from any known sented, among others, by Helicopsis striata, Chondrula tridens, 67 Pupilla muscomm, Vallonia costata and in some areas by Leached soils, called parabraunderdes, with claY-en. Abida fmmentum All its elements today liw in central riched B-horizons have the highest stratigraphic value. They Europe, and the assemblage as a whole is closely related to the are the product of deciduous forests, which in this part 01 Holocene steppe faunas of southeastern Europe. The Striata Europe are, by defmition, interglacial. Today they border the fauna is characteristic of what could be called a "warm loess· southern periphery of the loess belt and reach their present steppe" (Lozek, 1969). Presence of Striata fauna does not northern limits in central Europe. necessarily imply that the climate was mild. The assemblage may resist much harsher winters than those occurring today in Correlative recent snail faunas of the forested borden eastern Europe. Nevertheless, it requires relatively warm and of the loess belt are richly diversified with Monachoides inC/lr. dry summers. nata, Ena montana, Discus perspectivus, Helicodonta obvoluta, and Helix pomatia. Lozek (1964) refers to these assemblages Obviously the loess of central Europe was deposited un· as the Helix faunas. der a variety of climatic conditions which have only this in common: expressed continentality, low precipitation and a The parabraunderde of Czechoslovakia and Austria short, if any, rainy season. Mean annual temperatures may which developed on calcareous loess of the last glacial age, h have been as low or even lower than in northern Siberia today in undisturbed environments, always associated with mixed but also as high as they are now in the Black Sea Basin. hardwood forests and Helix faunas. Fossil parabraunderdel are also associated with Banatica snail assemblages. These, in Two specific types of sediments are commonly found in addition to forms known from recent broadleaf forests, in. the loess-filled depressions of Czechoslovakia, the "pellet clude species such as Helicigona banatica, Soosia diodonta, and sands" and the "marker." The pellet sands, sediments com Aegopinella rossmani, requiring higher winter temperature! posed of sand-sized pellets made of silt or clay, are deposits and wetter summers than present. Gastropods are typicall) of torrential rains that followed a prolonged warm and dry found in krotovinas reaching into the Oca horizons of thil season. They are at least partly contemporaneous with gravel soil or in carbonate-rich, hillwash loams, immediately under aggradation in neighboring rivers. "Markers" are thin bands lying such soils. They are not present within the soil itself of light-colored silt probably deposited by dust storms of con since it is strongly leached. In the parabraunerde of the lasl tinental magnitude (Kukla, 1961 b). Snail assemblages of the interglacial, Frenzel (1964) found pollen of oak, linden pellet sands and of markers are closely related to the Tridens elm, and hazel, providing independent proof of an inter and Striata faunas. glacial origin for this soil. Stones of Celtis (cherry) are com mon in the older parabraunerdes and braunlehms (l.ozek Soils interstratified with loess and hillwash sediments 1964). record intervals of dense vegetation established at a locality for a period of at least several centuries. The quality of soil Braunlehms are fossiliferous, strongly-weathered, argillil horizons depends directly on the type, density, and duration forest soils and frequently have large carbonate concretioru of vegetational cover and in such a way indirectly on climate. with a high proportion of clay plasma. They are interpreted ill Because most of the studied localities lie on slopes, the soil polygenetic formation produced by alternations of warm-wei surfaces have not become fully stabilized, and the soil material and warm-dry climates. Their snail assemblages are essentiall) from higher elevations continued to be slowly redeposited the same as those of the parabraunerde. into the depressions. Chernozems, rich in humus, calcareous, well-granulate, The soils in the loess belt of central Europe are of two biogenic soils are the postglacial climax soils of the driesl main groups: (1) biologically reworked steppe (prairie) soils, loess belts. They are preserved north of Prague, south of Brno showing an accumulation of organic matter but little chemical and east of Vienna and Bratislava. Chernozems stretch tada) change of mineral matrix; and (2) leached soils of braunerde from central Germany and Czechoslovakia to the Urals ani type with evidence of in situ redeposition of carbonate, iron, central Asia, reaching as far north as 57° latitude. They are al and manganese as well as clay plasma. These are not reworked present soils of continental climates with cold to very coli to any Significant degree by pedofauna. winters, rainy springs, and dry summers. Under unaltered conditions, the loess belt in central Recent snail faunas in the driest parts of the basinS Europe today would be thoroughly vegetated, its surface sta with thick chernozems, are dominated by such species aI bilized, and the occasional limited airborne increments reach Chondmla tridens and Abida frumentum. They differ fron: ing the surface would be assimilated in soils without inter similar Pleistocene Tridens faunas of buried chernozerns b) rupting the pedogenetic process. Actually, however, large areas the presence of exothermic elements like Cepea vindobonensiJ have been denuded by man and are now subject to extensive or Oxychilus inopinatus and by the presence of southet1 soil erosion and redeposition. Thus, modern allochthonous immigrants like Helicella ceciloides, whose expansion WSl soils and loess-derived floodloams are frequent in the region. facilitated by man (Lozek, 1972). 68 The so-called frost-gleys or pseudogleys are fossil soil naming the soils according to their cyclic position in the izons that differ from the raw loess only by the greenish sequence (Kukla, 1975). Cycles of the first order are called ho r stains and frequent ochre-brown limonitic concentra- "glacial cycles" and are delimited by marklines. Those of the ,reY . ~, s These features are beheved to be due to repeated satura- second order are "stadial cycles" or "sub cycles" and are de (lOn,. . , of the soil by groundwater. As the soIls are often asso limited by submarklines (Fig. 5). Each cycle and sub cycle (lOn , ted with cold, tolerant Columella faunas and since the gley starts with a thin deposit of hillwash loam (phase 1), suc c11~enomena run parallel to the surface and are accompanied ceeded in turn by a forest soil of braunerde type (phase 2), ~ small involutions and signs of solifluction, it is assumed steppe soil of chernozem type (phase 3), marker (phase 4), t1;at they formed in the summer melt on top of the permafrost pellet sand (phase 5), and loess (phase 6). Some members of (Klima, et aI., 1962). the sequence may be missing, but within the studied area the successive order of the phases always remains identical. The Common braunderde soils are also found in the loess detailed, cyclic repetitions of chernozems, markers, and pellet sequences. They are leached and oxidized but do not show sands have only been reported from the last three glacial cycles, reflecting the fact that relatively few localities with ,I,,'ans of clay illuviation. Temperate snails were not found ac ,~mpanying such soils in any of the studied localities. Smolr- older sediments are known. kava (1971) studied the micromorphology of fossil braunerdes in the neighborhood of Brno and Prague and compared them Marklines are the boundaries between a thick layer of to arctic braunerdes, commonly found under boreal coniferous loess, commonly containing Columella or Pupilla faunas, and forests. the overlying parabraunerde or braunlehm. Where the hill wash of phase 1 is present beneath the soil, the markline is drawn at the boundary of the hillwash with the loess. Climatostratigraphic Value of Loess and Soils Each glacial cycle delimited by two successive mark lines is labeled by a capital letter. The elapsed part of the As the preceding discussion indicates, the paleoclimatic Holocene glacial cycle is labeled A, the last completed Pleisto information that can be derived from loess sequences comes cene cycle B, the next order one C, and so on, in order of from characteristic types of sediments or soils and, indepen increasing age. dently, from snail assemblages. Both soils and snails depend upon the type of established vegetation cover and on its per Within each glacial cycle an oscillatory but gradual sistence. They are specific to the loess belt, which is relatively change is evident between environments of a pronounced cli narrow in Europe. Thus, unlike marine faunas which can be matic optimum and of the most severe continental climate. followed moving north and south during the glacial cycles, Whereas the optimum occurs at the beginning of each cycle, the characteristic soils and gastropod assemblages of the loess the coldest environment is recorded shortly before its end. In belt alternately appear and disappear. this respect the loess cyclothems resemble the sawtooth glacial cycles in marine 0 18 records (Broecker and Van Donk, 1970) Interglacials are defined as intervals of warm and wet or the oscillations of foraminiferal assemblages and coccoliths climate during which temperate deciduous forests were estab in the North Atlantic (Ruddiman and McIntyre, 1976). lished in northwestern Europe (Turner and West, 1968; Fair bridge, 1972; Kukla, et aI., 1972; Turner, 1975). In the neigh Submarklines (Kukla and KoCl, 1972) are the boun borhood of Prague, Brno and Krems the development of hard daries of loess of any kind with overlying parabraunerde, wood forests was practically synchronous with the type braunerde, or chernozem. Each unit delimited by two suc region, Thus, parabraunerdes, braunlehms, and the accom cessive submarklines is called a "stadial cycle" or a "subcycle" panying Helix and Banatica faunas are the approximate and labeled with the capital letter of the corresponding glacial equivalents of the temperate phases II and III of interglacials cycle followed by an Arabic numeral (Fig. 5). as described in pollen records. Their presence in the sequence proves an interglacial proper. On the other hand, the full Individual lithologic units recognized in the studied glacial conditions, during which the glaciers in the Alps, area are so numerous that the original attempt to name them northern Germany, and Poland reached their last maxima, are after the type localities (KlIma, et aI., 1962) was abandoned. reCorded in the Prague and Brno area by frost-gley soils and by Instead, the soils are labeled by a lower case letter combined COlumella snail faunas. with the designation of the corresponding stadial or glacial cycle. The intention is to use the same designation for hori . A surprisingly close correspondence of sedimentary zons considered to be of equal age throughout the region. ,equences of different ages has been observed in the region However, because a correct correlation cannot always be guar arOund Prague and Brno. This observation led to the defini anteed, the complete reference needs to include the abbre ;~n of sedimentary cycles of the first and second order viated symbol of the locality. Also, where correlations are ukla, 1961 a) and later to a regional lithostratigraphic system made with units from outside the loess system, the prefix L 69 (for loess) is used to avoid confusion with similar labels of ELEVATION ABOVE SVRATKA RIVER .--r-,-r--r-, -r--r--r--., -'-'m o' -'--'1 ~ b- other authors. o 0 ~ g~~ Thus the designation L-CK-C3j refers to the soil labeled 1 as j observed within the sub cycle C3 (youngest section of gla () cial cycle C) at the locality Cerveny Kopec (symbol CK). ~ Correspondingly, the soil L-DK-C3j in Dolni Kounice (symbol DK) shows the same stratigraphic position and presumably is of the same age. 1 Apart from dividing the glacial cycle into three or five stadial cycles, there also are good reasons for a twofold subdi vision. Under such a scheme, a sedimentary sequence of every glacial cycle is divided into a "Lower" and "Upper" series. .o In the lower one, dark-colored soils and soil sediments pre o o dominate, while in the upper one, light-colored loess is the predominant deposit (Fig. 5). Typical arctic vertebrates like arctic fox, lemming, or wolverine are either missing or very rare in the lower series but are abundant in the upper one, ;:: giving it a pronounced glacial character. Thus the composi '".... ::u'" tion of vertebrate faunas in this part of Europe parallels the U> twofold rather than the multifold subdivision of the glacial cycle. The twofold subdivision is applicable for young as well as old deposits and is useful for correlation with deposits o outside Czechoslovakia and Austria. Similar twofold units are o common in loess areas of Hungary, Yugoslavia, Bulgaria, Romania, the Soviet Union, or in the United States, and are widely utilized in geologic mapping. o o The Time Control .o o The chronostratigraphic setting of the loess sequences is based on radiocarbon dates of charcoal and soil humus ex tending back over 50,000 years (Klima et al., 1962; Fink, 1969, 1976; Vogel and Zagwijn, 1967); positions of magneto m o stratigraphic boundaries correlated with the Brunhes/Matu o 1~ yama, the Olduvai/Matuyama reversals, and the Jaramillo and the Blake event. Time coincidence of marklines with the o o o 1T o () ------> o ~ Figure 6. Cerveny Kopec (Red Hill-CK) section at Brno, o o r Czechoslovakia. Exposed in an excavation front of a brickyard pit and in boreholes (vertical bars). Ter T () o races CK 1-5, the fossil-bearing section of Stn!nska "o ~ Skala (SS) and the Zidenice brickyard (ZD) are also shown. Horizontal scale shows distance from the Svratka .o o T River. 1: Brunhes/Matuyama reversal; 2: soil with n Biharian vertebrates; 3: cave fill with Biharian fossils; ~ A-K: baselines of corresponding glaCial cycles. Location ~ (II I 9 ! ~ I ? I I V> :E ~ in Figures 1 and 4. Detailed stratigraphic column in 131\31 '0'35 31'108'0' NOI1VI\313 V Figure 7. 70 unations in oceanic sediments makes it possible to use the abandoned river terraces and fIll sheltered depressions cut tern estimates of terminations directly as the age estimates of deep into the subsurface by meandering streams. As the :g~responding marklines (Table I). The summary of the avail streams gradually cut deeper, the high lying relief became in ~~le radiocarbon control is shown in Figure 5. creasingly exposed to denudation_ As a result, loess and soils immediately overlying fluvial sediments are thicker, better The record of the Brunhes/Matuyama reversal was 10- subdivided, and less affected by retrograde weathering or , ted in two sequences in Krems, Austria (Fink and Kukla, destruction. Each time the river cut to a deeper terrace level, L,a .,!,.J 1977), two at Cerveny Kopec (Figs. 6 and 7), C.zechoslovakia space was provided for the well-stratified younger fIll. Thus, a (BUcha, et al., 1969; Bucha, 1973; KoH, unpubhshed), one on "telescopic superposition" of strata (Kukla, 1961 b), exempli stranska Skala at v Brno and one in Unetice (Fig. 8) near fied in Figure 6 is commonly observed in the downwind Prague (KOCl and Sibrava, 1976). Independent paleontologic western banks of Czech and Moravian rivers. and morphostratigraphic correlations confirmed the same rela tive age of five out of the six locations mentioned (Kukla, 1975). The reversal occurs in sediments of late glacial J, or Stratigraphy possibly in the earlie~t layers of the interglacial I, in a very close vicinity of markline IX. Stratigraphy of several hundred loess sections in Europe has been described in the literature. Stratigraphic columns of Global synchroneity of gross climate changes results in some of them with brief descriptions are contained in special the synchroneity of terminations with marklines (Kukla, volumes edited by Fink (1969) and by Demek and Kukla 1970). Independent observations support such correlation. (1969). The age of markline I around Bmo falls close to 10,000 years B.P. judging from the final occurrences of autochthonous Figures 6 and 7 show the stratigraphy in one of the key wind-blown loess at C14 dated Magdalenian occupation levels sites exposed in the brickyard of Cerveny Kopec (Red Hill), in local caves (Valoch, 1968). Termination I, marking the last Czechoslovakia. Sediments and soils of ten completed glacial deglaciation, was dated in the cores of the equatorial Atlantic cycles B-K and of the Holocene rest in telescopic superposi to about 11,000 B.P. (Broecker, et al., 1960). Thus, the ages tion on top of five terraces of the river Svratka labeled locally of markline I and termination I are very close. Tennination II, as CKI-CK5. The Brunhes/Matuyama (B/M) paleomagnetic 127,000 years old according to Barbados coral reefs, predates boundary, 0.7 m.y. old, is on top of the high terrace CK4 and by about 5,000 years the interglacial optimum in the 0 18 is overlain by a soil with remains of Biharian vole Pitymys. record. Markline II which is more than 115,000 years old On the nearby locality of Stranska Skala in a similar position because of its relation of the reversed magnetostratigraphic above a high terrace, the B/M reversal is overlain by soil sedi unit in Modrice correlated with the Blake event, predates by a ments with rich Biharian vertebrate assemblage (Kurten, few thousand years the climax soil BIB of the deciduous 1960; Musil, 1969). The youngest part of the Red Hill se forests of the last interglacial. Thus the synchroneity of mark quence is paralleled at another nearby locality, Modrice, where line II with termination II is highly probable. an anomalously magnetized unit correlated with the Blake event was identified above the forest soil of the last intergla The number of distinguishable layers developed and pre cial (Kukla and Kocl, 1972). Chemozerns are the most abun served in the sequence, and thus the quality of climatostrati dant soils of the last five glacial cycles A-E, The climax graphic information, depends critically on the local morphol interglacial soils are parabraunerdes in the older cycles F-I; the ogy. Serious miscorrelations were frequent in the loess belt in soils are clayey braunlehms rich in large carbonate. Soils of former decades when Pleistocene stratigraphers did not pay cycles F and K are distinguished by strikingly red hues. sufficient attention to this fact. In the absence of independent correlation criteria, the authors dated the strata by counting Numerous boreholes and exposures in the vicinity of them down from the surface and labeling them in terms of Bmo demonstrate the stratigraphic relation between the loess Soergel's and Milankovitch's chronology. This "count from sequence and the terraces. Each terrace is a morphostrati top" method produced, for instance, the notorious hypotype graphic unit identified by relative elevation of its fluviatil soil of the Riss/Wurm interglacial, the Krelnser paleosol sediment incorporated into the terrace body. In order to date (Gross, 1957; Brandtner, 1956; Movius, 1960), which as we the terrace with respect to the loess sequence, exposures from now know is about one m.y. old, older than the whole of several localities have to be compared because deposition and Penck's classical Pleistocene, Also, by similar miscorrelation, erosion in a river bed occur simultaneously and give each a polygenetic soil at Paudorf was chosen to represent a Late terrace body a complicated internal structure. Sediments of Pleistocene interstadial 30,000 years old. It later yielded one greatly divergent ages may coexist at the same level. of the richest known snail assemblages of the last interglacial (Fink, 1976). It was found in several localities that the deposition of river gravels was contemporaneous with the pellet sands. The Most of the studied sections in central Europe rest on earliest recognized deposit of the lowermost terrace in the 71 • I CERVENY KOPEC CK ENVIRONMENT ENVIRONMENT 1 NTGL GLAC I HTGL ~ FOREST >- ., >- If) -' If) '"Vi I- U)" w '"0 U) a:: 0 ..J a:: a::'" w 0 "§ 6 <: " ~ 8 "§. "- 10 <- 12 to I I I I ~4 en 0 16 z '(), >- w -' -' W Q U W 0: ~ <: ~ ~ ·3 .' l;: ~'" .,0. ~ ::: i'"J 30 " F ~ ;;: ~ ~h j il5 ~ I. I I : ~i••! :'.'.• ' I Figure 7. Stratigraphy of the Cerveny Kopec (Red Hill) sequence in Czechoslovakia (cf. Fig. 6). Composed of segments with high est resolution. Lithologic symbols same as in Figure 5. BL-braunlehm. Levels of deep erosional incision in cycles C, F, I and K marked as breaks. VF 1-VF 3: vertebrates of Biharian faunal wave; VF 4: vertebrates of Steinheim faunal wave. ReCOn structed environment: same symbols as in Figure 5. Details in Kukla (I 975). 72 VLTAVA + ELBE Lb 10 lb IIo IIb IIIc No Nb METERS rno rnb ABOVE RIVER ILM + SAALE III II I 2 3 4 5 6 100 - 80 - 60 - 40 - 4- SAALE CORBICULA ...... :: ... 20 - o - RHINE UPPER LOW. HT HT MTI MTll MTm MTN NT 100 - I B/M REVERSAL-~~~? 60 - FRIMMERSDORF INTER 40 - ELSTER GLACIAL + PTEROCARYA / +JUGLANS 20 - o - ~~8NC; Figure 8. Terrace sequences of Vltava and Elbe around Praha, Czechoslovakia. Dm and Saale around Leipzig, Germany and lower Rhine around Koln (Cologne), Germany. Designation of Vltava terraces after Zaruba (1942), Dm after Soergel (1939), and Rhine after Brunnacker (1975). Terrace gravel stippled, floodloam marked with horizontal stripes, interglacial soils and silts black, ice deformed sections zig-zagged, loess blank, B-D glacial cycle baselines. After Kukla (1975). 73 vicinity if Bmo, CK1, correlates with the pellet sands of sub The Krems soil complex consists of rubefied braun. cycle C2 (Kukla, 1975). Correspondingly, the time span during lehms, representing the most intensely weathered Pleistocene which CKl terrace has been deposited is bracketed between polygene tic soils known from central Europe. The Cea hOri. the next-to-the-last glacial of cycle C and the present. Gravels zon, at the base of KR7a, is a hard, compact limestone crust and floodloams are still being incorporated into the body of penetrated by vertical root casts filled with red braunlelun CKl (Fig. 6). It should be noted that during sub cycle B2 the plasma (Lehmstangen). Almost no loess, only pellet sand, is river cut to a considerable depth. But because of subsequent preserved from the end of the glacial cycle M. It is unclear aggradation returned to previous and even higher levels, a whether the soil group KR7a-8 represents three glacial cyclea separate terrace unit did not develop. as shown in the diagrams or only two. However, because of the high degree of weathering of the soils and because of the In a similar manner the CK2 terrace, whose surface is exceptional thickness of the basal carbonate horizon, the about 20 to 30 meters above the present river, is younger than former alternative was considered the more probable. the floodloams of sub cycle Gl and older than the pellet sands of cycle C. The oldest-known loess covering this terrace is The loess separating the soils KR9 and KR10 and the that of cycle F (Fig. 6). The CK3 terrace, whose surface is 40 soils KR12 and KR13 seems to be, on the average, thicker to 50 meters above the river, is younger than sub cycle 13 and than other early Pleistocene loess units in Krerns. older than the loess of sub cycle F3. The oldest-recognized fluviatil sediments of this terrace were deposited either during A relatively thick, positively magnetized unit found cycle H or I. CK3 is the youngest terrace in the vicinity of under soil KR13 is interpreted as correlative with the top of Brno bearing highly weathered braunlehms, including the the Olduvai magnetozone. A unit with anomalous polarity in markedly red soils of cycle F The CK4 terrace lies 60 meters the KR 7 soil was interpreted as a result of normal fossil reo above the river. Reversely magnetized, water-reworked loess magnetization of an originally reversed layer. It is understood of cycle J with fluvial gastropodes covers the terrace gravels that the latter process took place when the soil weathered at both Cerveny Kopec and Stranska Skala. The age of this during the Jaramillo normal event. oldest terrace could not be bracketed because of the lack of suitable localities, but it is probable that downcutting from the Krems is the only hitherto known site where the glacial CK5 to CK41evei occurred during cycle K. cycles of Mid-Matuyama age are exposed in an apparently continuous sequence. At least 9 full glacial cycles were com· Thus we recognize in the vicinity of Brno three terraces pleted in the Middle and Upper Matuyama Epoch between of Brunhes age. Each one of the three was deposited during the end of the Olduvai and the Brunhes/Matuyama boun· glacial as well as interglacial climate and represents deposition dary. The combined record of Cerveny Kopec and Krems of more than one glacial cycle. The oldest of the three Brunhes therefore documents 17 full glacial cycles which occurred in terraces is capped by a highly-weathered, clayey-red polygene central Europe in the last 1.7 m.y. During each of them the tic soil of cycle F. The next older terrace CK4 is overlain by environment shifted from a temperate broadleaf forest to a reversely magnetized floodloam derived from reworked loess highly continental loess tundra. As unrecognized gaps may and therefore dates from Matuyama time. exist in the Krems stratigraphic column, this number has to be considered a minimum. The Kremser Schiesstatte (Krems Shooting Range) Paleoclimatic Information Derived The Kremser Schiesstatte (Krems, symbol KR) is the ex from the Loess Record in Central Europe posure in the loess fIll of an abandoned river meander elevated 60 meters above the Danube (Fig. 9). The site was described The main conclusions on European climates, derived among others by Gotzinger (1936), Fink (1955), and Bradtner from numerous loess sequences studied in detail (Demek and (1956). The two best-developed soil groups were considered Kukla, 1969), can be summarized as follows: to represent the R/W interglacial' (soils KR7-9) and the early Wurm Gottweiger interstadial (soil KR4). Demek and Kukla 1. The complexity of climate history on land is similar (1969) and Fink and Kukla (1977), showed that the main soil to that in the oceans. The number of full inter· complex is embedded in reversely magnetized sediments of glacials of Pleistocene age approaches twenty. Al· Matuyama age. The interglacial snails at the base of KR4 over though the well subdivided early Pleistocene se lying the Brunhes/Matuyama reversal include Biharian Heli quences are rare, there are no indications that ~ cigona eapeki and are closely related to the assemblage in the detailed climatic development of any glaCi f basal interglacial soil at Stranska Skala. Another biharian interglacial cycle was less complicated than that ~ interglacial snail assemblage is found in KR5. The soils KR7, the last one. Thus, the total number of major PlelS' KR9, and KR10 contain Late Villafranchian voles Mimomys tocene stadial-interstadial cycles can be estiJll8~ plioeaenieus, Mimomys hungarieus, and Lagurus sp. (Fink, to be on the order of a hundred. Each such cy e 1976). probably included several climatic oscillatiolll 74 KREMS ENVIRONMENT ~. z ~ ~ m ~ m , o· REMARKS <..l Z ~~ o:r~!-,:lS ~ i I o I'~' 'R' ~~I ,R 2 (/) Pupilla fOJIJO 'R' KR4p ,?; Cllondrula tndefiS 'R4m, ! I 10 t: 12 I, 14 16 IX J 20 22 ~ Yl ;rLEGEND N " 32 0 34 IIRlO f 1 'LOCDL:J4~ P KR11 36 R JEBRIS, c u\,JSL IDE SOLI cL UC T 38 40 'z 42 Me)S' 0 T 44 =, co CRUST ~ >- SOLIFl UCTION " ~ >- ~ RLJ"IROW ~:5" Figure 9. Stratigraphy of the Krems sequence, Austria. KRl-16: local soil designations. Inclination and declination plots after alf demagnetization to 150 oersted. Levels studied for gastropods marked with a snail-like symbol giving characteristic species after l.ozek (unpublished). Symbols for environment same as in Figure 5. BL-braunlehm. Vertebrates after Fink (1976); K and U breaks: levels of deep erosion. Krems 010 to 0100 MHZ: local designation of magnetohorizons. Further details in Fink and Kukla (1977). 75 significantly exceeding the range of variability ob THE ALPINE GLACIAL STAGES served in meteorological records of the past two or three centuries (Fig. 10). The four classical glacial stages, Wurm (W), Riss (Ro) Mindel (M), and Gtinz (G), and the three interglacials 2. The sequence of environmental changes recorded in R.isa: WUrm (RW), Mindel-Riss (MR) and Gunz-Mindel (GM), Were central Europe during the last three completed named by Penck (in Penck and Bruckner, 1909) after the fOIll glacial cycles B, C and D closely followed a uniform, Bavarian tributaries of the Danube (Donau) and Isar. The sY1- repetitive pattern. tern was later extended by adding two older glacial stages, 3. Environments of the last four interglacials of cycles Donau (Eberl, 1930) and Biber (Schaefer, 19S3)-and two B to E resemble that of the Holocene. Mixed decid corresponding interglacials-Donau-Gunz (DG) and Biber. uous forests were present in the highlands of central Donau (BD). The type locality of all the stages is the Iller. Europe, while temperate grasslands on top of Lech Platte in the northern foothills of the Alps, south of chernozem soils stretched further east. Ulm, Augsburg, and Munich (Fig. 4). 4. Two oldest soils of glacial cycle C were formed under deciduous forests. By comparison, only a The type units of the glacial stages are gravel terraces single basal soil in each of the cycles Band D cor called Niederterrasse (Lower Terrace, NT) for Wurm, Hochter. responds to such conditions. Thus the interglacial rase (High Terrace, HT) for Riss, Jtingerer Deckenschotter of cycle C was doubled. The timely correlative (Younger Cover Gravel, IDS) for Mindel, Alterer Deckenschot. 0 18 stage 7 of the deep-sea sediments shows the ter (Older Cover Gravel, ADS) for Gunz, Donauschotter same character. (Danube gravel) for Donau, and H{)henschotter (Heights S. Basal polygene tic soil of cycle F, as well as soils of gravel) for Biber glacial. The Lower Terrace is connected with cycles G-L with the exception of cycle J, are more morphologically fresh-looking Young Moraines (Jtingmorlinen) strongly weathered than the younger soil groups whereas the High Terrace is connected with poorly preserved and contain large carbonate concretions, partly Old Moraines (AItmorlinen). The surfaces of Mindel and filling rootcasts. This indicates longer and/or repeti Gtinz terraces, contrary to the younger two, are strongly tive occurrence of forests at the sites in central weathered into the red-clayey zone called "Ferretto." While Europe alternating with warm xerophytic steppes the terraces represent the glacials, the interglacials are repre or savannas. sented by the erosional steps separating the terraces and inter· preted as intervals of down cutting and removal of sediments. 6. The sedimentary features and faunal assemblages The vertical distance of Riss and Mindel terraces is in many point to the exceptionally mild glacial intervals of areas especially great. Because the depth of the incision from cycles G, L, and M. This correlates well with the one terrace to the next was considered to be a measure of poor expression of the 0 18 cold peak of stage 14 time, it was concluded that the M/R was substantially longer and with the 0 18 record below stage 23. than the R/W interglacial. 7. Intensive and widespread rearrangement and down cutting of the river network during cycles F and K Although Penck clearly stated that each of the four lead to separation of distinct terrace levels. Because glacials is documented by the corresponding "glacial series" both episodes do not seem accompanied by any which includes moraines and glacial outwash deposited during indications of an exceptionally cold climate, it is the outermost stands of piedmont glaciers in the type area, speculated that they may have been caused by a his stratigraphic system rests entirely on the four gravel ter· temporary intensification of crustal movements. races. Penck believed that their formation was fully contem Similar but less pronounced episodes occurred in poraneous with the glaciations (Penck and Bruckner 1909). cycles C and I (or H). As the terraces are well-dermed and mappable morphostrati 8. The Late-Villafranchian faunal elements were re graphic units, they proved ideal for regional extension of the placed by the typical biharian faunal elements in system all over the Alps. Thus, in practice, Penck and Bruckner cycle K. These in turn gave way to a successive and their numerous followers established and used the terraces young Pleistocene faunal wave during or before as type units of Wurm, Riss, Mindel, Gtinz, Donau, and Biber cycle F. glaciations (see also discussion by Zeuner, 1946). 9. The Pleistocene climatic cycles of Matuyama age in central Europe had basically the same amplitude The four terraces of Penck and Bruckner represent widt as those of the Brunhes Epoch. The environment spread morpho stratigraphic correlation units which can be repeatedly alternated between a continental cold readily differentiated over much of the Alpine foreland. An: loess steppe or semi-desert and a temperate inter several decades of intensive study and revisions, the value glacial forest. Mean duration of the cycles was close Penck's Alpine stages as first-class morphostratigraphic correla to 0.1 m.y. tion units remains intact. 76 SNAIL ILOCAL FAUNAL' Cf) w SSEMB IOCCURENCES ENVIRONMENT ..-J ..-J U Cf) « Cf) Cf) w w GLACIAL INTERGL >- U f- W ..-J ~ >- >-U « « Z U n:: STEPPE f- >- ~..-J >- Cf) ..-J ..-J FOREST (.? Cf) en ..-J ~ n::« ..-J U -",:: « -",:: ..-J W w« «U « « U n:: - f- o..U W ..-J« n:: (.? w « w C N M PB BL RL (.? f- « n:: BL R T BE >-0 O..-J « w n:: ..-J Z W I f-..-J « o..(.? ~ ~ en (.? z Cf) > I I II I I I I I W A I L> .... 0 >Q: B 0 0.11 0 II :!: C \j C rv .... • i:: ill 0 ~ a 77 Under the influence of Milankovitch's insolation chron Present knowledge on the stratigraphy of the forealpinl ology, Eberl (1930) described the so-called "Vollgliederung" terraces in Bavaria, Switzerland, and lower Austria is sYn or "full delineation" of the Alpine Pleistocene system. Within thesized in the composite diagram shown in Figure 11. T~ the original type area he recognized three substages of the numbered features show: Wtirm, two of the Riss, two of the Mindel, two of the Gtinz, and three of the Donau, and differentiated them by numerals. 1. Rolled Roman brick dredged from the submergel The substages are numbered in order of decreasing age. Eberl's base of coarse gravels of the Niederterrasse h approach combined geomorphic mapping with lithostrati Vienna during recent engineering operations (Fink graphic observations. While some of his conclusions were later 1976). found to be based on miscorrelations, the finding on at least 2. Logs at the base of Niederterrasse, C 14 dated a two generations of Riss gravels separated by a weathering 300 ± Y.B.P. in Frauendorfa.d. Au, (Piffl, 1971). interval of possibly interglacial rank was supported (Wold 3. Logs of oak (Quercus, cf. robur), C14 dated a stedt, 1958). Also, the existence of two moraine generations 3165 ± 115 Y.B.P. and 3130 ± 65 Y.B.P., foundiJ in the Wtirm and two in the Mindel, each probably separated coarse gravel in Trubensee east of Krems (Piffl by a non-glacial interval of considerable length, were reason 1971). ably well demonstrated. What remains unclear, however, is the stratigraphic relation of the non-glacial interlayers within 4. Logs of pine and birch up to 8.5 meters long, C1j the terrace bodies to individual end moraines and terrace sur dated at 9660 ± 115, and 9185 ± 95 Y.B.P., iJ faces. Eberl's substages were never successfully extended Neustifft east of Krems (Piffl, 1971). Logs of simi outside the type area. lar age dated to 9480 Y.B.P. and 9700 Y.B.P. il the gravels of Niederterrasse in Ulm and Lin: Whereas the geomorphologic setting of the Alpine ter (Graul, 1973). races was properly described and stratigraphically exploited, 5. The subhorizontal layer of parabraunerde on tOI its climatic interpretation was incorrect. As we today know of the lower terrace gravel in Murnau, overIaiJ the gravels accumulated not only during glacials but also dur by a Wl1rm moraine (Kaiser, 1963). (Improbabll ing interglacials. The erosional intervals which separate the alternative explanation proposes that the soil plasm terraces were, for the most part, glacial rather than inter leaked through the moraine in Holocene.) Two gen glacial events (Schaefer, 1953). erations of Wtirm gravels separated by till in thl ..;;t M N -0 0.00 r-.... -0 lO ..;;t M N- PTEROCARYA w 0.:0.: 0.: Q.; £:) (f) ex -.J aiai ai ai w W ~~ >-' ~ -"- z 0 • (f) (f) wz I[) :J D- Ol[) 0 B/M 0.7 MY ~ • + + + + PR • II II D GUNZ MINDEL R I S S t W U R M I I I 1 GM MR RW Figure 11. Schematic cross section showing classical Alpine Pleistocene stratigraphy in the Iller-Lech Platte type area in lower Austria and in the vicinity of Zurich. Gravel blank; soils, interglacial clays and silts black; loess stippled. Moraines marked as M 1-2, R 1-2, W 1-2. Numbers 1-14 on top denote sites discussed in the text. 78 same area (Penck, 1922); interglacial peats correla It clearly follows from Figure 11 that the gravels of the tive with the Eemian (Frenzel, 1973) within the low Niederterrasse, which by definition represent Wiirm, must have laying so-called Laufen gravel at Zeifen. accumulated during Late Pleistocene as well as Holocene, in 6. Three localities close to Munich with gastropods glacial as well as the postglacial and interglacial climates. The of deciduous forests in fine grained interlayers of the logical conclusion of our discussion, therefore, is that Wurm, high terrace (Brunnacker and Brunnacker, 1962). as Originally defined, includes Late Pleistocene and Holocene, glacial and postglacial. In a similar way, the older glacial stages 7. Parabraunerde and loess of glacial cycles D, C, and were also deposited during both glacial and interglacial epochs. E, in part accompanied by interglacial snail faunas, Obviously, the climatic concept of classical Alpine glacial overlying local terrace, correlative with the high stages is in serious error. Hochterrasse and merging with its floodloams in Senftenberg and Thallern near Krems. A question could be raised as to whether the use of the 8. The deep weathering horizon capping the thick terms "Wl1rm," "Riss," "Mindel," and "Gtinz" should be accumulation of Pre-Riss gravels of Zink (1940) in restricted only to sediments deposited during the glacial northern Switzerland around Basel and Zurich, phases and thus correlative with the corresponding moraines. overlain by gravels locally labeled as Riss-l (Altriss) This probably was the original climatostratigraphic concept and still younger gravels of Riss-2. In Riss-l, the of Penck. Unfortunately, there is no safe way to distinguish glaciers presumably attained their maximum extent sterile interglacial from glacial gravels within one terrace body. in this part of Switzerland. The Reidmatt forest Furthermore, it is too late for any substantial revision of the bed from the Pre-Riss/Riss-l interval and the Grun system since terraces were extensively used as type units of holz black clays with remains of temperate flora the four glaciations for several decades. presumably corresponding to Riss-l /Riss-2 interval (Woldstedt, 1958). To avoid further confusion, the terms "Wurm," "Riss," 9. The M/R erosional gully which at Zurich cuts 500 "Mindel," and "Gtinz" as well as "Donau" and "Biber" should meters beneath the Mindel terrace and far beneath only be used in connection with either the terraces or with the the present river alluvia filled by rapidly deposited moraines in the Alps, and reference should be made in full to "Rinnenschotter" (Zink, 1940). either "WUrm terrace" or "Wtirm moraine." Different time 10. The red-clayey soil Ferretto on top of the Mindel stratigraphic and climatostratigraphic setting of moraines terrace at numerous locations throughout the Alps and terraces must be taken into account. The terrace sequence (penck and BrUckner, 1909); it is by an order of completely covers time, irrespective of glacials or interglacials, magnitude stronger than any younger soil known in while the moraines and their related outwash gravels are cor the region. rectly interpreted as glacial features but represent only a few millenia of maximum expansion of the piedmont glaciers. 11. At least two generations of Mindel moraines, also two levels of Mindel gravels, in the Iller-Lech Platte, Correlation of the Loess Sequence described by Eberl (1930). with Classical Alpine Glacial Stages 12. A reversely magnetized floodloam on top of the Ferretto in the Giinz terrace at Grabnerstrasse in Four major terraces, CKl, CK2, CK3, and CK4, under Linz (Kukla, 1975; KoC[ and Sibrava, 1976). lie sediments of the eight glacial cycles of the Brunhes epoch 13. Two surfaces of the Gunz terrace recognized by at Cerveny Kopec in Brno (Fig. 6). The oldest one is of Matu Eberl (1930) in the type area. yama age. The upper two are overlain by strongly weathered 14. Lignite lenses in clays capping the Donau gravel at braunlehms. Four terraces-Niederterrasse, Hochterrasse, lung Uhlenberg, west of Augsburg, containing inter erer Deckenschotter, and Alterer Deckenschotter-represent glacial flora with Pterocarya and Carya. These the four classical Alpine glacials-Wurm, Riss, Mindel, and latter elements are abundant in the Waalian inter Gtinz-of Penck and BrUckner (1909). The oldest one, Gunz glacial but not in the following Cromer-type inter terrace in Linz, Austria, is of Matuyama age. The upper two glacials (Filzer and Scheue?lpflug, 1970). terraces, namely the 1 ungere and Altere Deckenschotter, locally bear a strongly weathered soil Ferretto. Both Brno and The authority of Penck was respected to such a degree Linz lie in a similar tectonic position at the border of the that the interglacial layers within the terrace bodies were Bohemian Massif and the forealpine depression. Both belong Interpreted as local anomalies rather than an indication of a to the Danube watershed. Therefore the broad correlation of rartly interglacial age of Alpine terraces at large. Because three Brunhes terraces at Brno with the three Brunhes terraces ossils within gravels are very rare and because only a few of the Danube and of its Alpine tributaries is reasonably well species of vertebrate fauna are specific for interglacials, some justified. We may conclude that with high probability the geologists still believe in the glacial origin of any coarse river Alpine Wurm, as represented by its type unit Niederterrasse, grave!. correlates with CKI and thus covers the younger half of 79 cycle C, the whole of cycle B, and the elapsed section of temperate lusitanian fauna (Madsen, 1928). Today a similar cycle A. Using parallels between the loess system and the assemblage is found along the coast of Portugal in mUch deep-sea record, we further conclude that Wilrm correlates in warmer waters than those of the present North Sea. time with 0 18 oceanic stages 1-6, Riss with CK2, and thus with cycles P, E, D, and the older part of C or with oceanic The Holstein marine transgression is morpholOgically 0 18 correlatives 7-12. Mindel probably correlates with CK3 related to the Elster advance whose glaciofluvial sands it over. and thus with cycles P, G, H, and 1. It is represented in the lies (Fig. 12). The pollen sequences in the lake beds of lIo~ ocean by 0 18 stages 13-18. Gunz correlates with K, J, ~ld a stein age, frequently with a characteristic water fern AZolla part of cycle J and with 0 18 stages 19-22 (Fig. 15). The exact filicuides, point to the presence of forests dominated by COni. stratigraphic position of the breaks between individual terraces fers with low proportion of hardwood trees. Molluscs Palu. of the Alpine system may locally differ from the proposed dina diluviana and Corbicula fluminalis are the abundant scheme, for reasons explained in the discussion of the Brno fossils of the Holstein river beds. Marine molluscs and fora area, but the evidence at hand strongly points to cycles C, P, minifers of Holstein epicontinental seas point to relatively H (or I) and K as episodes of accelerated crustal movements chilly waters, similar to those of the present North Sea. north of the Alps, correlative with the R/W, M/R, G/M, and DIG Alpine interglacials. The Cromer Forest Bed, the type unit of the Cromer interglacial, is exposed in the cliffs of the North Sea in the vicinity of East and West Runton, England (Reid, 1882). THE CLASSICAL NORTH EUROPEAN Oscillating freshwater, brackish and marine layers, contain GLACIAL AND INTERGLACIAL STAGES rich assemblages of Biharian vertebrates and pollen of mixed forests. The Cromer local fauna is of the Biharian wave. It The classical glacial stages in the area formerly covered has numerous extinct elements and is well distinguishable from by the Fennoscandinavian continental ice sheet are named a younger faunal wave accompanying the post-Elster strata. "Weichsel," "Warthe," "Saale," and "Elster" (Table I). They Forests and vegetational history of type Cromer resemble the were recognized from the end-moraine system, crossing Den Postglacial (Turner, 1975). The Lowestoft till overlying the mark, Netherlands, northern Germany, and Poland (Fig. 12), Cromer forest bed is believed to correlate with the Elster and are named after the rivers of this area. Moraines of the advance. In Germany there are several interglacial deposits Weichsel and Warthe have a fresh hummocky constructional containing biharian fauna, broadly correlative with Cromer. topography; moraines of the Saale are dissected and flattened; The annually banded lake deposit in Bilshausen records the and the moraines of the Elster have lost their originalmorphol change from the reversely magnetized late glacial lake beds ogy completely. The former extent of the Elster glacial ad of Matuyama age to normally magnetized full interglacial beds vance was, for the most part, reconstructed from tills and from of Brunhes age (Fromm, unpublished; H. Muller, personal the distribution of erratics. As the type units of each of the communication). four glacial stages are moraines, the resulting system is a morphostratigraphic one. The naming of the moraines is Tllis is, in short, the basic framework of classical Pleisto attributed to Keilhack (1926) who mapped them in north cene stratigraphy of the glaciated parts of northern Europe central Germany. The Warthe moraine was originally grouped The present knowledge on the stratigraphy of principal units with the Weichsel, then regrouped with the Saale by Wold in northwest Europe was summarized in Figure 13. The figure stedt (1954). Finally, it was singled out as a separate glacial shows: stage by Picard (1964). It should be observed that the older moraines lie progressively further south. This was a principal 1. The boreholes made in recent years which revealed condition of their preservation and recognition. the existence of two successive transgressions with temperate assemblages of marine microfaunas for The classical interglacial stages of northern Europe are, merly believed to represent Eem in North Germany in order of increasing age, the Eem, Holstein, and the Cromer. (Wiegank, 1972). They are separated by deposits of They are represented by the deposits of marine transgressions glacial origin. and by peat bogs whose pollen content documents the pres 2. Holocene sediments of Baltic Sea transgressive over ence of temperate hardwood forests in northwestern Europe. the Late Weichsel tills. 3-5. Fresh Weichsel moraines overlying the buried inter The stratigraphic position of the Eem transgression is glacial lake and riverine deposits in the vicinity of established by the fact that its sediments overlie Saale and Berlin. The lower one with mollusc Paludina dilu Warthe tills but are overridden by Weichsel (Fig. 13). In a viana was correlated with Holstein. similar way Holstein deposits overlie the Elster but are buried by Saale. The Eem interglacial is recognized and named 6. Eem marine beds with lusitanian fauna in Denmark. after the marine clays containing shells of the mollusc Tapes 7. Fresh terminal moraines of Warthe. Outside of aureus var. eemiensis accompanied by other molluscs of Warthe on top of the terrace cut into Saale deposits. 80 o 100 KM 1- 2 Wllill 3- 4 ...... 5 ... 0000 6- 7 figure 12. End moraines of the north European glaciations and the extent of the interglacial marine deposits. After Liedtke (1975). 1 - Weichsel, 2 - Warthe, 3 - Saale moraines, 4 - Maximum extent of Saale tills,S -- maximum extent of Elster tills, 6 - Eem marine deposits, 7 - Holstein marine deposits. the fossiliferous travertines of Ehringsdorf, next to 12. Buried channels of Late Elster filled with fluvio Weimar. Though long held as representing the last glacial sands and occasional layer of till. interglacial, they were dated to over 200,000 years 13. Ground moraines of type Elster. and correlate with the lower series of cycle C and with the 0 18 stage 7 (Goddard, et aI., in prepara 14. Biharian vertebrates in Voigstedt, Germany. Here tion). the Elster till and pro-glacial varves cap a thick unit of loess, with sand and gravel at the base. Fossili 8& 10. Two tracks of dissected terminal moraines of Sa ale. ferous silts and sands with Biharian mammals The inner so-called Rehburg moraine track was Trogontherium cuvieri, Dicerorhinus etruscus, etc., possibly overridden by the main Saale. underlie the gravel. These, in turn, are underlain by 9. Marine clays and peats overlying the Saale ground two gravel beds, separated by clay with the inter moraine. The type Eemian. glacial mollusc Mellanopsis acicularis. Layers of two 11. Marine clays and peats of type Holstein overlying interglacials are probably present at this site. the Lauenburger Clay. Two interglacials in super 15. Cromer forest bed in England. Relaton to mainland position with floral succession of Holstein type stratigraphy unknown. were recently described in the region. The older one corresponds to the type Holstein (Ducker, Several were buried until sheets differing in composition 1969). from the embedded rock debris were found in a number of 81 boreholes and excavations. They enabled local subdivisions of to be 10,000 ± 1,000 years. By a similar method, the telll Saale and Elster to be made into three substages each (Cepek, ate deciduous forest of the Holstein interglacial lasted~· 1967). At least three separate interglacials in stratigraphic 13,000 ± 1,000 years, and the so-called Ruhme inter~ position of Cromer were described from the Netherlands of Cromer type in Bilshausen lasted 30 to 32,000 years (}.{f4. (Zagwijn, 1975). Multiple interglacial horizons with Cromer ler, 1974a, 1974b, 1965). The latter, however, was interruPteQ type fauna are also known from Germany. approximately in the middle by a 400-year-Iong expansion pine and birch. During this short episode the temperate s of In Mosbach in the Rhine Valley fossiliferous gravels, cies almost disappeared from the area. The Bilshausen de~ sands, and silts accumulated in the 10km-wide depression closely overlies the B/M boundary. which continuously subsided during the early and a part of middle Pleistocene. The basal gravels yielded young Villa The varved deposits revealed how little time is actuan franchian vertebrates with Archidiskodon meridionalis. The represented by the type units of north-European intergla~ younger main fossiliferous layer overlying the reversely magne Out of the 700,000 years known to have elapsed since the end tized silts (KoC[, et aI., 1973) contains typical Biharian verte of the Matuyama epoch, the Postglacial together with Eetn brates such as Canis lupus mosbachensis, Pantera gombaszo gensis, or Ursus deningeri (Bruning, 1972; Kahlke, 1975). The youngest deposits of the Mosbach basin are correlated with the Upper Middle Terrace of the Rhine. Afterwards the CROMER TYPE ~ subsidence ceased and the basin was gradually uplifted (Wag ner, 1930). 'CROMER' FAUNAS IN VOIGSTEDT i There is no doubt today that more than two intergla cials took place between Elster and Weichsel and that numer ous till sheets were deposited in Europe. But it still remains ELSTER TYPE (;j unclear what is the relation of the tills to the surface moraine tracks and which of them represents the first order advances as ELSTER CHANNELS N opposed to only local intersections of contemporaneous ice lobes. HOLSTEIN = One of the great discoveries of recent decades is the yearly laminated interglacial deposits. The duration of inter SAALE (TYPE ?) glacial environment at such sites has been relatively accurately counted (review by Turner, 1975). The length of the Eemian interglacial in Bispingen, Germany was in such a way found EEM - EEM RIVER SAALE - REHBURG (l) ------> WARTHE Figure 13. Scheme of Pleistocene stratigraphy in north central Europe. Tills and moraines marked by crosses, floodloams and lake beds by stripes, marine deposits in EEM - SKAERUMHEDE (Jl black, sand stippled, gravel marked with open circles, loess blank. Black semicircles: Corbicula fluminalis (II BRANDENBURGER and Paludina diluviana. 1: duplicate Ecmian transgres STADIUM ~ sion according to Wiegank (1972); 2: postglacial peats ~ and marine beds; 3-5: fresh terminal moraines of Weich (") J: ~ (j) sel; 6: Eem beds in Denmark; 7: fresh terminal moraines FRANKFURTER fTI STADIUM r of Warthe, 8 and 10: dissected terminal moraines of "7i Saale; 9: marine deposits in the vicinity of Amersfoort, 01 Netherlands, type Ecmian; 11: Holstein marine beds and Lauenburger clay; 12: Late Elster burried channels and tills; 13: type Elster ground moraines; 14: Biharian POSTGLACIAL III vertebrates in Voigstedt; 15: Cromer forest beds in East Anglia in unknown morphologic relation to European EEM - RUGEN mainland. Modified from Kukla (1977). 82 en w F A UNA z :J N ~ !:: a: >- ~ 10 LOESS, COLUMNAR I' 15 -= • LOESS, SANDY STREAKS 517- GREEN - GREY GLEY - LIKE MOTTLING 25 -= 516- LOESS, COLUMNAR, FEW SNAILS I: I I 30- 002- i I 514- LOESS, SANDY 513- SOIL, DARK GREY, HUMOUS, CARBONATE ROOTCASTS MIT 512- BROWN, DECALCIFIED, KROTOWINAS - ...... 515- I~ 35- LOESS _Mm 518- ~ 519- SOIL, RED-BROWN, DECALCIFIED , 39r STRONG CCo IMPREGNATION LOESS, COLUMNAR 40 - 520- SILT, SAND BANDS, COLLUVIAL MN 5~~= SOIL, RED-BROWN, CLAY SKINS, INSECT BURROWS ...... 522- E8~sSHORIZON 5~~: TWIN SOIL. ~\D-~~~W~6RI~~ANY SKINS SEPARATED 45- ~ SILT, SAND BANDS, COLLUVIAL 526- LOESS, COLUMNAR 527- SOIL, RED - BROWN, DECALCIFIED 528- SOIL, RED - BROWN, CLAY SKINS 31 r CCo HORIZON 50- 30r 29r LOESS, WHITISH 510- SILT, SAND BANDS. COLLUVIAL, GULLIES 16- 509- ~P~f: ~~~~~[j PARTLY DECALCIFIED 55 SOIL, INSECT BURROWS ~W 12- SOIL, RED - BROWN, CLAY PLASMA \ ... BUI- BONATE IMPREGNATION 501- LOESS, WHITISH, FREQUENT SHARDS OF PEARLETTE ASH, NORMAL POLARITY J: O~ w o LOESS, WHiTISH. SHARDS OF ASH (J)z 537- } ow ASH WITH SILT, LAMINATED, COLLUVIAL Q.~ PEARLETTE "0" ASH, c.;OARSE GRAIN. LAMINATED x::::E NORMAL POLARITY . w~ ,gure 14. Buzzard Roost loess section in south central Nebraska, U.S.A. Location in Figure 1. Lithologic symbols same as in Figures 5 and 9. Fauna above 34m from Frankel (1956), below 34m depth section is sterile. Local marklines, not neces sarily time-correlative with those in Europe, are recognized as boundaries between porous eolian loess and decalcified soil (or immediately underlying colluvial silt). Encircled plus sign: paleomagnetic samples with normal polarity. Further details in Schultz and Stout (1961, and other publications). 83 by exceptionally deep erosion which was followed Unfortunately, glacials are represented and named after tb by deposition of channel gravels of the Rhine and moraine tracks which, for the most part, cannot be relat~ Elbe. With high probability, this erosional episode to type interglacial sites. Unknown gaps exist in the phYSica correlates with Alpine M/R. The erosion has been evidence left by glaciers. In practice it is impossible to pro~ explained as a result of intensified crustal move the continuity of a glacial environment between two conseC\l ments north of the Alps (Grahrnann, 1933; Genies tive interglacial deposits. The repetitive nature of glacial ~ er, 1962; Quitzow, 1959). It may correlate with the interglacial deposits, the common lack of distinguis~ Ortenau orogenetic phase as recognized by Witt fossils, and the common truncations of Pleistocene strata 811 mann (I 939) in the upper Rhine basin. further deficiencies of the north-European subdivision. 7. Cromer: Correlation of the Cromer forest bed with In such a situation Pleistocene stratigraphers workinj the loess sequence and with the 0 18 stages is dif in Europe commonly apply a stratigraphic approach that CaJ ficult. Paleomagnetic study of Montfrans (Zagwijn, be labeled a "count from top" (CFT) method. In essence eae) et al., 1971) and of Kukla (unpublished) revealed unit is assigned a minimum possible age by comparing !hi only normal polarity, but the samples are weakly studied section with the accepted climatostratigraphic mode magnetized and the remanence may be entirely due and assuming an uninterrupted deposition. to secondary overprint. Biharian vertebrates indi cate that type Cromer is close in age to the Brunhes/ The CFT method has produced a remarkable number 0 Matuyama paleomagnetic boundary, as are the miscorrelations. A single till, for example, was frequentlJ localities Stranska Skala, Bilshausen and Mosbach. assigned to two different glaciations depending on how com However, Biharian faunal elements are also reported plete the overlying section was. Similar miscorrelations OC from deposits younger than type Elster (Heller and curred with interglacials as well. Frenzel (1973) argues tha Brunnacker, 1966) and from an interglacial con for botanical reasons several interglacial deposits formedJ siderably older than the Brunhes/Matuyama boun believed to represent a single warm interval could not haVl dary (Fink and Kukla, 1977; Lozek, 1972). In actually coexisted. agreement with these observations the alternative position of type Cromer is bracketed by cycles After many years of widespread miscorrelations, thl F and J (Fig. 14). classical north-European terminology cannot be saved. Thl A glaciation older than Cromer is claimed from sooner it is abandoned in interregional correlations the better. Poland (Rozycki, 1969; Mojski and Ruhle, 1965). It is called the "Podlasice" or "Elbe glaciation" and documented by tills and varved lake deposits RELA TION OF EUROPEAN encountered in the subsurface. It is difficult to AND NORTH AMERICAN prove convincingly that the tills are older than the CLASSICAL GLACIAL STAGES type Elster; therefore, their correlation with either 0 18 stage 16 or possibly 20 remains open. The limited scope of the present manuscript does nol permit an attempt to correlate, with the deep-sea sediments, Comments to Correlations in Northern Europe the four classical American glacial stages, i.e. Wisconsin, 00· noian, Kansan and Nebraskan, and the three interglacial stage! The north-European Pleistocene subdivision is based on Sangamon, Yarmouth, and Afton (Table 0. As the Pleistocene a sound climatostratigraphic concept. The principal units gross climatic changes were globally synchronous and as the are the interglacials. Their climate related lower and upper four-fold subdivision of European Pleistocene was shown to be boundaries are clearly fixed in continuous pollen-bearing in conceptual error, the obvious conclusion is that the four sequences by the appearance and disappearance levels of fold subdivision of American Pleistocene is also incorrect. abundant pollen of temperate trees (Turner and West, 1968; Kukla, et aI., 1972; Turner, 1975; Zagwijn, 1975). Type area The American glacial stages are, according to their of the interglacials is the north-western Europe (Fairbridge, original definitions (Geikie, 1894; Chamberlin, 1895; Shirnak. 1972). The north-European interglacials are the only widely 1909), represented by moraines and tills in much the same recognized Pleistocene land-based stratigraphic units whose way as the north-European glacial stages. Although never boundaries occur in continuous sequences and thus comply specified by the authors, the American glacials were under with the principal requirement for a definition of formal stood as times when continental ice was present in the type chronostratigraphic unit (Hedberg, 1972). Since by conven area, which is the Midwest. For the same reasons given in the tion the Pleistocene is divided into glacials and interglacials, discussion of north-European glacial stages, it is highly imPro all strata underlying and overlying interglacials are glacial. bable that the continental ice sheets remained in the Midwest Hence, type sections of the interglacials contain all the boun for continuously longer periods than approximately 10.~ daries of the classical north-European Pleistocene units. years. It is probable that, as in Europe, the end moraines 0 86 Wisconsin and lllinoian, and the outermost tills of Kansan We believe that the detailed study of loess sequences, d Nebraskan, were formed close to the end of a glacial cycle especially those in Nebraska, is a worthwhile objective which an ring the cold peaks which immediately preceded a termina should be further pursued. A large amount of data is already du n It is furthermore improbable that a moraine or till pre available on the loess sequences in the North Platte vicinity. t1~~d in the Mid west would correspond to the exceptionally Figure 14 shows the recently updated stratigraphic column of ~ld 0 18 stage. the Buzzard Roost section, with six major polygenetic soils separated by loess, and covering the interval of the last 0.6 There are morphologic parallels between the Wisconsin m.y. (Schultz and Stout, 1961; Frankel, 1956). The Pearlette nd illinoian drifts, and the Weichsel, Warthe, and Saale "0" ash outcropping at the bottom of the section is radio a oraines. The Wisconsin drifts and the Buffalo Hart moraine metrically dated to about 0.6 m.y. (Boellstorf and Te Punga, ~Leverett, 1899) of late lllinoian age display a fresh and 1977; Naeser, et al., 1971) and is known to be younger than hummocky surface (William and Frye, 1970). The Weichsel the Kansan tills. The Buzzard's Roost section therefore corres and Warthe in Europe have similar morphology. Both groups ponds in time to Illinoian, Yarmouth, and Wisconsin. The were originally viewed as products of a single glaciation, strongly developed soils, weathered to a substantially higher namely Weichsel and Wisconsin respectively. Buffalo Hart degree than the local climax Holocene soil, are more numerous moraine (leverett, 1899) is separated from Wisconsin by a than the classical paleoclimatic model predicts. The number decalcified interglacial weathering horizon (Johnson, 1964). matches, perhaps by coincidence, that of the Pleistocene ter Similarly, Warthe is separated from Weichsel by the Eemian minations of the last 0.6 m.y. interglacial deposits in Schleswig-Holstein. Early- and mid lllinoian moraines show weathered and dissected surfaces Obviously the prospect of west Nebraskan loess sections similar to the type Saale and Saale-Rehburg moraines in linking the glacial stratigraphy in the Midwest with the deep Europe. sea record is promising. More effort should be spent in re search in this area. A major rearrangement of the hydrographic net oc curred between Kansan and lllinoian and between Sa ale and Elster. FINAL REMARKS The type units and the underlying concept of classical Our present results in correlation of deep-sea sediments American interglacials differ substantially from the European with the land-based Pleistocene deposits of Europe is sum ones. Type units of Sangamon, Yarmouth (Leverett, 1898) marized in Figure 15. The diagram covers the last 0.9 m.y., and Afton (Chamberlin, 1895) are sediments and soils whose about half of the Pleistocene. The early Pleistocene sequences only climatic implication is the ice-free surface at the type older than approximately 1 m.y. were not included because localities. It is probable that much of the Midwest was ice the 0 18 stratigraphy of this interval is not yet worked out in free during the formation of the soils, accretion gleys, and the sufficient detail. Tentative correlations of the early Pleisto sandy fills of buried channels originally representing the cene older than 1 m.y. in Europe were published by Zagwijn classical American interglacials. However there is nothing to (1975) and Menke (1975). indicate that the Sangamon, Yarmouth, and Afton deposits were limited to peak warm climates correlative with European Our correlation, still quite fragmentary and inaccurate, interglacials typically marked by the presence of deciduous demonstrates that the paleoclimatic models of European forests in northwest Europe. Obviously the classical American Pleistocene are in serious error. Because the paleoclimatic interglaCials, as originally defmed, represent much longer inter reconstruction was the underlying concept of both the Alpine vals of time than any of the European ones, and, therefore, in and the north-European stratigraphic systems, these cannot terms of the European climatostratigraphic concept, include be correct either. both the glacial and the interglacial climates. The mistaken simplistic subdivision of the Middle and Would an approach, similar to the one used in Europe, Late Pleistocene into three classical interglacials and four be applicahle in correlation of the American glacial stages with glacials is largely the result of a climatic misinterpretation of the deep-sea stratigraphy, using loess, terraces, and lake beds Alpine terraces, of the repetitive nature of incomplete physi Outside the fonnerly glaciated areas as a link? cal evidence in the glaciated areas of northern Europe, and of common persistent violations of basic stratigraphic principles Differences between the two continents partially dis by Pleistocene geologists. Stratigraphic terms were indiscrimi faVor such an approach. There are fewer artificial exposures nately used outside the type areas with meanings different IJlloess and fewer, less-accurate studies of terraces made in than those originally intended. Time content of the chrono cnt'leal areas. On the other hand, the widespread presence stratigraphic units was commonly extended far beyond the of radiometrically dated mid- and early-Pleistocene Pearlette boundaries actually represented in the type deposits. Due to ashes compensates for the deficiencies. the repetitive nature of Pleistocene sediments, this practice 87 leads to wide overlaps of the chronozones composing the precession of the earth in the Quaternary. NatUr system. 244:108-109. e, After decades of misinterpreted usage of the Alpine Bucha, V., J. Horacek, A. KoH and J. Kukla. 1969. PaleOllla and the north-European terminology, it is doubtful whether netische Messungen in Loessen. In periglaZialzon! either can be saved. Few stratigraphers would like to use a loess und Paleolithikum der Tschechoslowakei, J. Deillek system in which the units overlap in time or in which inter and J. Kukla, eds. Tschechoslowakische Akademie der glacials and glacials were labeled by one name. The sooner the Wissenschaften. Geographisches Institut in Brno:123. classical terminology is abandoned in interregional correla 131. tions, the better. Future global subdivision of the Quaternary must be based on continuous sequences such as those in the Capek, A. G. 1967. Stand und Probleme der Quartarstrau. oceans or lakes. graphie im Nordteil der DDR. Berichteder Deutschen Gesellschaft der geologischen Wissenschaften A, Geo. logie, Palaontologie. 12:375404. REFERENCES Chamberlin, T. C. 1896. Jour. Geology. 4:872-876. Arrhenius, G. 1952. Sediment cores from the East Pacific. Swedish deep-sea exped., 1947-1948. 5:1-228. __. 1895. Glacial phenomena of North America In The great ice age. 2nd ed., J. Geikie, ed. New York, Apple. Brandtner, F. 1956. Loess Stratigraphie und palaolithische ton:724-775. Kulturabfolge in Niederoesterreich und den angren zenden Gebieten. Eiszeitalter und Gegenwart. 7:127- Dalrymple, G. B. 1972. Potassium-argon dating of geomat. 175. netic reversals and North American glaciations. In Calibration of hominoid evolution. W. W. Bishop and Boellstorff, J. D. and M. T. Te Punga. 1977. Fission-track ages J. A. Miller, eds. Scottish Academic Press:107-134. and correlation of middle and lower Pleistocene se quences from Nebraska and New Zealand. N.Z. Jour. Demek, J. and J. Kukla. 1969. Periglazialzone, loess und Paleo. of Geol. Geophysics. 20(1):47-58. lithikum der Tschechoslowakei. Tschechoslowakische Akademie der Wissenschaften. Geographisches Institut in Briskin, M. and W. A. Berggren. 1975. Pleistocene stratigraphy Brno. and quantitative paleoceanography of tropical North Atlantic core VI6-205. In Late Neogene epoch boun Dubois, G. 1925. Recherches sur les terrains quaternaires du daries, T. Saito and L. H. Burckle, eds. Am. Mus. Nat. Nord de la France. Mem. Soc. Geol. du Nord 8. Lille, History Micropal. Spec. Pub. 1:167-198. 1925. Broecker, W. S., M. Ewing and B. C. Heezen. 1960. Evidence Ducker, A. 1969. Der Ablauf der Holstein-Warrnzeit in West· for an abrupt change in climate close to 11,000 years holstein. Eiszeitalter und Gegenwart. 20 :46-57. ago. Am. Jour. Sci. 258:429448. Eberl, B. 1930. Die Eiszeitenfolge im nordlichen Alpenvor· Broecker, W. S. and J. Van Donk. 1970. Insolation changes, land. Augsburg. ice volumes, and the 0 18 record in deep-sea cores. Rev. of Geophysics and Space Physics. 8: 169-198. Emiliani, C. 1955. Pleistocene temperatures. Jour. GeoI. 63:538-578. Bruning, H. 1972. Die eiszeitliche Tierwelt im Rhein-Main Gebiet Mosbacher Sande. Naturhistorischer Museums __ . 1958. Paleotemperature analysis of core 280 and fuhrer 4. Mainz. Pleistocene correlations. Jour. Geol. 66:264-275. Brunnacker, K. 1975. The Mid-Pleistocene of the Rhine Basin. __ . 1966. Paleotemperature analysis of the Carribean cores In After the Australopithecines, K. W. Butzer and G. Ll. P6304-8 and P6304-9, and a generalized temperature Isaac, eds. The Hague, Mouton Publishers: 189-224. curve for the past 425,000 years. Jour. of Geol. 74: 109-124. Brunnacker, M. and K. Brunnacker. 1962. Weitere Funde pleistozaner Molluskenfaunen bei Miinchen. Eiszeitalter __. 1972. Quaternary paleotemperatures and the duratioJl und Gegen wart 13: 129-137. of the high temperature intervals. Science. 178 :398401. Bucha, V. 1973. Correlation between field variations and Emiliani, C., T. Mayeda and R. Selli. 1961. Paleotemperature 88 analysis of the Plio-Pleistocene section at Le Castella, Gardner, J. V. and J. D. Hays. 1976. Responses of sea surface Calabria, southern Italy. GeoI. Soc. Am. Bull. 72:679- temperature and circulation to global climatic change 688. during the past 200,000 years in the eastern equatorial Atlantic Ocean. In GeoI. Soc. Am. Memoir 145. R. M. Emiliani , C. and N. J. Shackleton. 1974. The Brunhes epoch: Cline and J. D. Hays, eds.:221-246. palaeotemperatures and geochronology. Science. 183: 511-514. Geikie, J. 1894. The great ice age and its relation to the anti quity of man, 3rd ed. London, Stanford:I-850. Ericson, D. B., W. Ewing, G. Wollin and B. C. Heezen. 1961. Atlantic deep-sea sediment cores. GeoI. Soc. Am. Bull. Geitzenauer, K. R., M. B. Roche and A. McIntyre. 1976. 72:173-286. Modern Pacific coccolith assemblages: derivation and application to Late Pleistocene paleotemperature analy Ericson, D. B. and G. Wollin. 1968. Pleistocene climates and sis. In GeoI. Soc. Am. Memoir 145. R. M. Cline and J. D. chronology in deep-sea sediments. Science. 162: 1227- Hays, eds.:423-448. 1234. Genieser, K. 1962. Neue Daten zur Flussgeschichte der Elbe. Eiszeitalter und Gegenwart. 13:141-156. Fairbridge, R. W. 1972. Climatology of a glacial cycle. Quater. Research. 2 :283-302. Goddard, J., W. S. Broecker and G. J. Kukla. (In preparation.) Filzer, P. and L. Scheuehpflug. 1970. Ein friihpleistozanes Pollenprofil aus dem nordlichen Alpenvorland. Eiszei Gotzinger, G. 1936. Das Lossgebiet urn Gottweig und Krems talter und Gegenwart. 21 :22-32. an der Donau. Fuhrer Ld. Quartar-Exkursionen in Oster reich, Vienna. 1 :1-11. Fink, J. 1954. Die fossilen BOden im osterreichischen loess. Grahmann, R. 1933. Die Geschichte des Elbtales von Leit Quartar. 6 :85-1 07. meritz bis zu seinem Eintritt in das norddeutsche Flach land. Mitteilungen Vereins L Erdkunde Dresden. N.F., . 1955. Das Marchfeld. Beitrage zue Pleistozanforschung. Dresden, 1933.: 133-194. --In Osterreich, Verhandlungen d. Geologischen Bun desanstalt. Sonderheft D. Graul, H. 1973. Foreland of the Alps. Eiszeitalter und Gegen wart. 23/24:268-280. __ . 1969. La stratigraphie des Loess d'Europe. Supple ment au Bull. de L'Assoc. Francaise pour l'Etude du Gross, H. 1957. Die Fortschritte der Radiokarbonmethode Quaternaire :3-12. 1952-56. Eiszeitalter und Gegenwart. 8: 141-181. . 1976. Exkursion durch den Osterreichischen Teil des Hays, J. D., J. Imbrie and N. J. Shackleton. 1976. Variations nordlichen Alpenvorlandes und den Donauraum zwis in the earth's orbit: pacemaker of the ice ages. Science. chen Krems und Wiener Pforte. Mitteilungen der Kom 194:1121-1132. mission fUr Quartarforschung der Osterreichischen Akademie der Wissenschaften. Band 1. Hedberg, H. D., ed. 1972. International Subcommission on Stratigraphic Classification (Report 7a): Introduction to Fink, J. and G. J. Kukla. 1977. Pleistocene climates in central an international guide to stratigraphic classification, Europe: at least 17 interglacials after the Olduvai Event. terminology and usage. Lethaia. 5 :283-295. Quatern. Research. 7 :363-371. Heller, F. and K. Brunnacker. 1966. Halsbandlemmingreste Frankel, L. 1956. Pleistocene geology and paleoecology of aus einer Oberen Mittelterrasse des Rheins bei Nieder parts of Nebraska and adjacent areas. Unpublished PhD. aussem. Eiszeitalter und Gegenwart. 17 :97 -112. thesis. Univ. Nebraska Dept. GeoI.: 1-297. Imbrie, J. and N. Kipp. 1971. A new micropaleontological Frenzel, B. 1964. Zur Pollenanalyse von Lossen. Eiszeitalter method for quantitative paleoclimatology: application und Gegenwart. 15:5-39. to a Late Pleistocene Caribbean core. In The late Ceno zoic glacial ages. K. K. Turekian, ed. New Haven, Yale _____ . 1973. Some remarks on the Pleistocene vegetation. University Press: 71-181. Eiszeitalter und Gegenwart. 23/24 :281-292. Johnson, W. H. 1964. Stratigraphy and petrography of Illi Ganssen, R. 1922. Die Entstehung und Herkunft des Loss, noian and Kansan drift in central Illinois. Illinois GeoI. Mitteilungen v.d. Labor. d. Preuss. GeoI. Landes-Anstalt:4. Survey Circ. 378 :38. 89 Kaiser, K. 1963. Zue Frage der Wtirm Gliederung durch einen __. 1975. Loess stratigraphy of central Europe. In Nte "Mittel-Wtirm-Boden" im nordlichen Alpenvorland bei the Australopithecines, K. W. Butzer and G. Ll. Isaacl Murnau. Eiszeitalter und Gegenwart. 14:208-215. eds. The Hague, Mouton Publishers:99-188. ' Kahlke, H. D. 1975. The macro faunas of continental Europe . 1977. Pleistocene land-sea correlations. I. EuroPe during the Middle Pleistocene: stratigraphic sequence Earth-Sci. Rev. 13:307-374. and problems of intercorrelations. In After the Austra lopithecines, K. W. Butzer and G. Ll. Isaac, eds. The Kukla, G. J. and V. Lozek. 1961. Loess and related del>Osita Hague, Mouton Publishers:309-374. Survey of Czechoslovak Quaternary. Czwartorzed E~ py Srodkowej i Wschodniej. INQUA VIth Intern. Cong Kay, G. F. and M. M. Leighton. 1933. Eldoran epoch of the Institut Geologiczny, Prague. 34:11-28. Pleistocene period. Geol. Soc. Amer. Bull. 44:669-674. Kukla, G. J. and A. KoH. 1972. End of the last interglaCiai Keilhack, K. 1926. Das Quartar in Grundztige der Geologie. in the loess record. Quatern. Resear. 2 :374-383. Vol 2, W. Salomon, ed. Stuttgart. Kukla, G. J. R. K. Matthews, and J. M. Mitchell, Jr. 1972 Kellogg, T. B. 1976. Late Quaternary climatic changes: evi The end of the present interglacial. Quatern. Resear dence from deep-sea cores of Norwegian and Greenland 2:261-269. seas. In Geol. Soc. Am. Memoir 145, R. M. Cline and J. D. Hays, eds.:77-11O. Kurten, B. 1960. Chronology and faunal evolution of the earlier European glaciations. Commentationes Bio Kent, D., N. D. Opdyke and M. Ewing. 1971. Climatic change logicae. 21 :3-62. in the North Pacific using ice rafted detritus as a climatic indicator. Geol. Soc. Am. Bull. 82:2741-2754. Leverett, F. 1898. The weathered zone (Yarmouth) betweeJI the Illinoian and Kansan till sheets. Iowa Acad. Sci Klima, B., J. Kukla, V. Lozek and H. DeVries. 1962. Strati Proc. 6: 171-18l. graphie des Pleistozans und Alter des palaeolithischen Rastplatzes in der Ziegelei von Dolni Vestonice (Unter __. 1899. The Illinois glacial lobe. U.S. Geol. SurveyMon Wisternitz). Anthropozoikum. 11 :93-145. 38:817. Koci, A., W. Schirmer and K. Brunnacker. 1973. Paleo mag Liedtke, H. 1975. Die nordischen Vereisungen in Mittel· netische Daten aus dem mittleren Pleistozan des Rhein europa. Forschungen zur deutschen Landeskunde 204 Main Raumes. Neue Jahrbuch Geol. Palaont. Mh. 9 :545- Bundesforschungsanstalt fUr Landeskunde und Rau· 554. mordnung. Bonn-Bad Godesberg:1-160. v-, v Koci, A. and V. Sibrava. 1976. The Brunhes-Matuyama Boun- Liu, Tung-Sheng and Tsung-Hu Chang. 1961. The "Huangtu" dary at central European localities. Report 3 of Project (loess) of China. Report 6th INQUA Congr. 4:503- 73/1/24. Quaternary Glaciations in the Northern Hemis 524. phere. Prague: 135-160. Lozek, V. 1961. Quartarmollusken der Tschechoslowakei. Ku, T. L. and W. S. Broecker. 1966. Atlantic deep-sea strati Rozpravy Ustredniho ustavu geologickeho 31. Prague. graphy: extension of absolute chronology to 320,000 years. Science. 151 :448. __. 1969. Palliontologische charakteristik der Loss-Serien. In Periglazialzone, Loss und Palliolithikum der Tsche Kukla, G. J. 1961a. Quaternary Sedimentation Cycle-Survey choslowakei. Tschechoslowakische Akademie def Wit of Czechoslovak Quaternary-Czwartorzed Europy srod senschaften. Geographisches Institut, Brno :43-60. kovej i wschodniej. IN QUA VIth Intern. Congo Instytut Geologiczny. Prace 34: 145-154. __ . 1972. Holocene interglacial in central Europe and itS land snails. Quatern. Resear. 2:327-334. __ . 1961 b. Lithologische Leithorizont,e der tschechoslo wakischen Loessprofile. Vestn& Ustredniho ustavu Madsen, V. 1928. Ubersicht tiber die Geologie von Darunark· geologickeho.36:369-372. Danmarks Geologiske Undersogelse. Copenhagen. 5:4. __ . 1970. Correlations between loesses and deep-sea sedi Marosi, P. 1970. Scurta Privire asupra Evolutiei TeorJilof ments. Geologiska Foreningen i Stockholm Forhandlin Loessogenezi. Cluj Univ. Stud. Ser. Geol. Minera1.15:6~' gar, Stockholm. 92:148-180. 73. (English trans. In Loess lithology and geneSls,l .. 90 smalley, ed. Stroudsburg, Pennsylvania, Dowden, Hutch berichte d. preussischen Akadernie d. Wissenschaften. inson and Ross, Inc.:402409. Phys.-math. KI. I nke, B. 1975. Vegetationsgeschichte und florenstratigraphie Picard, K. 1964. Die stratigraphische Stellung der Warthe ~ e Nordwestdeutschlands im Pliozlin und Frilhguartar. Eiszeit in Schleswig-Holstein (Deutschland). In Report Mit einem Bettrag zue Biostratigraphie des Weichsel of the VI Intern. Congr. on Quaternary Warsaw, 1961. Frilhglazials. Geoi. Jb. 26:3-151. Bd 2 Lodz.:191-197. ~lojski, J. and ~. Ruhle ..1965. Geological atl~s of Poland pim,1. 1971. Zur Gliederung des Tullner Ft~des. Ann. Natur stratigraphic and facial problems. In Fasclcule 12. Quat histor, Wien. 75 :293-310. ernary, Warszawa. Prosek, F. and V. Lozek. 1957. Stratigraphische Ubersicht des \!ovius, H.1., Jr. 1960. Radiocarbon dates and Upper Paleo tschechoslowakischen Quartars. Eiszeitalter und Gegen . lithic archeology in central and western Europe. Current wart. 8:37-90. Anthro. 1 :355-391. Quitzow, H. W. 1959. Hebung und Senkung am Mittel-und ~ruler, H. 1965. Eine pollenanalytische N eubearbeitung des Niederrhein wahrend des Jungtertiars und Quartars. interglazial-profils von Bilshausen (Unter-Eichsfeld). Fortschritte in der Geologie von Rheinland und West Geoi. Jb. 83:327-352 . falen.4:389400. . 1974a. Pollenanalytische Untersuchungen und Jahres Reid, C. 1882. The geology of the country around Cromer. -schichtenzahlungen an der eemzeitlichen Kieselgur von Memoirs of the geological survey of England and Wales. Bispingen/Luhe. Geoi. Jb. 21 :149-169. Rona, E. and C. Emiliani. 1969. Absolute dating of Caribbean . 1974b. Pollenanalytische Untersuchungen und Jahres cores P6304-8 and P6304-9. Science. 163 :66-68. --sChichtenzahlungen an der Holstein-zeitlichen Kieselgur von Munster Breloh. Geoi. Jb. 21 :107-140. Rosholt, J. N., C. Emiliani, 1. Geiss, F. F. Koczy and P. J. Wagnersky. 1961. Absolute dating of deep-sea cores by Musil, R. 1959. Osteologicky material z paleolitick~ho s{dliste the Pa 231/ Th/230 method. Jour. of Geoi. 69:162- v Pavlove, Cast II. Anthropozoikum. 8 :83-106. 185. __ . 1969. Stratigraphische Korrelation im Pleistozan Rozycki, S. Z. 1969. Climatostratigraphy and its application auf Grund der Vertebratenentwicklung. In Periglazial, with Pleistocene of middle Poland as example. Geo- Loss und PaIaolithikum der Tschechoslowakei, J. Demek graphia Polonica, Warszawa. 17 :7-39. and J. Kukla, eds. Tschechoslowakische Akademie der Wissenschaften. Geographisches Institut in Brno:61-73. Ruddiman, W. F. and 1. K. Glover. 1972. Ice-rafted volcanic ash: a tracer of North Atlantic paleocirculation. Am. MIilil, R., Valoch and V. Necesany. 1955. Pleistoc~nni sedi Geophys. Union Trans. 53 :423. (Abstract.) menty okollBrna. Anthropozoikum, Prague. 4:107-142. Ruddiman, W. F. and A. McIntyre. 1973. Time transgressive Ninkovitch, D. and N. J. Shackleton. 1975. Distribution, deglacial retreat of polar waters from the North At stratigraphic position and age of ash layer "L" in the lantic. Quatern. Resear. 3:117-130. Panama Basin region. Earth & Planetary Sci. Letters. 27:20-34. __. 1976. Northeast Atlantic paleoclimatic changes over the past 600,000 years. In Geoi. Soc. Am. Memoir 145, Naeser, C. W., G. A. Izett and R. E. Wilcox. 1971. Zircon R. M. Cline and J. D. Hays, eds.:111-146. fission-track ages of Pearlette-like volcanic ash beds in the Great Plains. Geoi. Soc. Amer. 3,7 :657. Schaefer, I. 1953. Die donaueiszeitlichen Ablagerungen an Lech und Wertach. Geologia Bavarica. 19: 13-64. Opdyke, N. D. 1972. Paleomagnetism of deep-sea cores. Revs. of Geophys. and Space Phys. 10:213-249. Schonhals, E. 1951. Uber fossile BOden im nichtvereisten Gebiet. Eiszeitalter und Gegenwart. 1: 109-130. Penck, A. and E. Bruckner. 1909. Die Alpen im Eiszeitalter Bd 1-3. Chr.-Herm. Tauchnitz. Leipzig. Schultz, C. B. and T. M. Stout. 1961. Field conference on the Tertiary and Pleistocene of western Nebraska. Univ. __ . 1922. Die Terrassen des Isartales in den Alpen. Sonder- Nebraska State Mus. Spec. Pubi. 2:1-55. 91 Shackleton, N. 1. 1969. The last interglacial in the marine The Late Cenozoic glacial ages, K. K. Turekian, ed and terrestrial records. Proc. Royal Soc. London. B New Haven, Yale University Press:391424. . 174:135-154. Van Donk, 1. 1976.018 record of the Atlantic Ocean for the Shackleton, N.l. and N. D. Opdyke. 1973. Oxygen isotope entire Pleistocene Epoch. In Geol. Soc. Am. Memo' and paleomagnetic stratigraphy of equatorial Pacific 145. R. M. Cline and 1. D. Hays, eds.:147-164. It cores V28-238: oxygen isotope temperatures and ice volumes on a 105 year-106 year scale. Quatern. Resear. Valoch, K. 1968. Evolution of the Palaeolithic in central and 3:39-55. eastern Europe. Current Anthropology. 9:351-390. ____ . 1976. Oxygen isotope and paleomagnetic stratigraphy Verklitch, M. F. 1969. La stratigraphie des loess d'Ukraine of Pacific core V28-239 Late Pliocene to latest Pleisto In Supplement au Bull. de l'Assoc. Francaise pou; cene. In Geol. Soc. Am. Memoir 145, R. M. Cline and l'Etude du Quaternaire, 1. Fink, ed.:145-150. 1. D. Hays, eds. :449464. Velichko, A. 1969. Les traits essentiels de la stratigraphie de! _. __ . 1977. Oxygen isotope and paleomagnetic evidence for loess de la plaine d'Europe orientale. {n Supplement a~ early northern hemisphere glaciation. Nature. 270: Bull. de l'Assoc. Francaise pour l'Etude du Quater. 216-218. narie, J. Fink, ed.:160-164. Shackleton, N. 1. and R. K. Matthews. 1977. Oxygen isotope Vogel, 1. C. and W. H. Zagwijn. 1967. Groningen radiocarbo~ stratigraphy of dated coral terraces in Barbados. Nature. dates VI. Radiocarbon. 9 :63-1 06. In press. Wagner, W. 1930. Die altesten linksrheinischen Diluvialterril£ Shimek, B. 1909. Aftonian sands and gravel in western Iowa. sen zwischen Oppenheim-Mainz und Bingen. Notizbl Geol. Soc. Amer. Bull. 20:399408. Ver. f. Erdk. u. Hess. Geol. L.-A. 5: 13. Smalley, I. J. 1975. The interaction of great rivers and large Wiegank, K. F. 1972. Okologische Analyse quartarer Foramini deposits of primary loess. In Loess lithology and genesis, feren. Geologie 21. Beiheft 77, Berlin: I-Ill. Benchmark Papers 26, I. J. Smalley, ed. Stroudsburg, Pennsylvania, Dowden, Hutchinson and Ross, Inc. Hal Willman, H. B. and 1. C. Frye. 1970. Pleistocene stratigraph) sted Press:326-370. of Illinois. Illinois State Geol. Survey Bull. 94:204. Smol(kova, L. 1971. Gesetztmassigkeiten der Bodenentwick Wittmann, O. 1939. Die Phasengliederung der diluvialer lung im Quartar. Eiszeitalter und Gegenwart. 22: 156- Rheintalrandtektonik. Bad. geol. Abhandl. 10. Karls 177. ruhe. Soergel, W. 1939. Das diluviale system. Fortschritte d. Geolo Woldstedt, P. 1954. Saaleeiszeit, Warthestadium und Weich gie u. Palaontologie. 7. Heft 39, Berlin. seleiszeit im Norddeutschland. Eiszeitalter und Gegen wart. 415 :3448. Thierstein, H. R., K. R. Geitzenauer, B. Molfino and N.l. Shackleton. 1977. Global synchroneity of Late Quater ___ . 1958. Das Eiszeitalter, Band 2. Stuttgart, Ferdinand nary coccolith datum levels: validation by oxygen iso Enke Verlag. topes. Geology. 5 :400404. Zagwijn, W. H. H. M. Van Montfrans and 1. G. Zandstra. 1971 Turner, C. 1975. The correlation and duration of Middle Subdivision of the "Cromerian" in the Netherlands. Pleistocene interglacial periods in northwest Europe. Pollen analysis, palaeomagnetism and sedimentary pe' In After the Australopithecines, K. W. Butzer and trology. Geologie en Mijnbouw. 50:41-58. G. Ll. Isaac, eds. The Hague, Mouton Publishers:259- 308. ____ . 1975. Variations in climate as shown by pollen analY' sis, especially in the Lower Pleistocene of Europe. In Turner, C. and R. G. West. 1968. The subdivision and zonation Ice Ages: ancient and modern, A. E. Wright and F. of interglacial periods. Eiszeitalter und Gegenwart. 19: Mosely, eds. Geological Jour. Spec. Issue 6. Liverpool, 93-101. Seel house Press:137-152. Van Oer Hammen, T., T. A. Wijmstra and W. H. Zagwijn.1972. Zaruba, Q. 1942. Ungsprofil durch die Moldauterrassen The floral record of the Late Cenozoic of Europe. In zwischen Kamik und Weltrus, Mitteilungen tscheschO' 92 slowakischen Akademie Wissenschaften Prague. 52: 1-38 . • bera, K. 1949. K soucasnemu ryzkumu kvarteru v oblasti Ze Ceskeho Masivu (A propos de l'exploration actuelle des terrains quaternaires dans Ie domaine du Massif Bohe mian). Sbornik Statniho geologickeho ustavu CSR, Prague. 16:731-781. Zeuner, F. E. 1946. Dating the past: an introduction to geo chronology. London, Methuen. link, F. 1940. Zur diluvialen Geschichte des Hochrheines, Mitteillungen Reichsst. f. Bod. Zweigst. Freiburg i. Brnsw 1. 93