background image


Molluscan succession from Holocene tufas in the Czech Karst

(Czech Republic)


Institute of Geology, Academy of Sciences of the Czech Republic, Rozvojová 269, 165 00 Praha 6, Czech Republic;

(Manuscript received December 8, 2005; accepted in revised form March 16, 2006)

Abstract: A detailed molluscan succession from Holocene calcareous tufa deposits at the Kotýz Ridge and in the Císařská
Gorge in the Czech Karst provides the most complete record from central Bohemia. The succession has been recon-
structed from three profiles and chronology provided by AMS radiocarbon dating of charcoal and dating of carbonate
by the U-series method. The early Holocene malacofaunas represented by occurrences of index species Discus ruderatus,
accompanied by other boreo-montane elements (Vertigo substriata, Vertigo alpestris, Perpolita petronella), dominate in
the earliest phases of tufa formation that is dated by the U-series method to 9460 ± 1200 years BP and 5920 ± 1520 years
BP in the Císařská Gorge, or before the interval 5070—4800 years BC at the Kotýz Ridge. The change in composition of
molluscan assemblages follows in the form of total replacement to Holocene Climatic Optimum (Atlantic, Epiatlantic)
malacofaunas, consisting of several indicative elements (e.g. Bulgarica cana,  Truncatellina claustralis), which are
extinct or relatively rare in that region today, and even in the Czech Republic as well. The youngest phases of molluscan
successions are represented by the accession of modern immigrants Xerolenta obvia and Oxychilus cellarius, never
documented before from the Subboreal period in this region.

Key words: Holocene, Czech Republic, radiocarbon age, calcareous tufas, malacofauna.


The faunal response to climatic and vegetational change
during the Postglacial is known in far less detail, although
the information is good for certain groups in certain re-
gions. Land snails are unusual amongst invertebrates hav-
ing a reasonably good fossil record, which could
potentially be used to address several of the basic biogeo-
graphical and paleoenvironmental questions. The most
valuable sequences are obviously those, which are contin-
uous, cover long periods of time and furnish fossil assem-
blages that faithfully reflect the living communities from
which they were derived. Relatively few deposits contain-
ing land snails fulfil all these conditions but calcareous
tufa deposits meet these requirements in large measure
(Preece 1991). The term tufa, sometimes subsumed under
travertine, has been used for a range of secondary calcium
carbonate deposits, but is here restricted to precipitates
formed primarily by the degassing of bicarbonate-rich
groundwaters around small springs. Many such deposits
began to form during the early Holocene and are often
several meters thick covering areas several hectares in ex-
tent. Tufas often yield shells of land snails and freshwater
molluscs in abundance, so that by sampling the deposit at
suitably fine resolution, it is possible to reconstruct the
faunal history of the site in some details.

Over large parts of Europe from the British Isles to the

Mediterranean and from Spain to the Czech Republic, Slo-
vakia and Poland, the rates of calcareous tufa deposition
were high in the early and middle Holocene, but declined
markedly thereafter (Weisrock 1986). Over much of Europe,
it has been postulated that in the late Holocene (since

~2500 years BP) there was a sharp decline in the deposition
of tufa (Goudie et al. 1993). There has been a considerable
debate about the causes of this phenomenon, with some au-
thors stressing the importance of natural climatic changes,
and others asserting that miscellaneous human activities
(e.g. deforestation connected with agricultural activities)
were crucial.

The Czech Karst, formed by Paleozoic limestones and

located in the central part of Bohemia between Praha and
Zdice, is an area with numerous deposits of calcareous
tufa. In this karst area about 70 localities where calcare-
ous tufa precipitates or was deposited in the past are
known (Kovanda 1971; Ložek 1992; Kadlecová & Žák
1998). Several tens of springs in the Czech Karst deposit-
ed calcareous tufa from their waters during the early and
middle Holocene. Tufa deposition continues at a number
of localities in the Czech Karst at present, as well (Kadle-
cová & Žák 1998). In general, the three morphogenetic
types of tufa accumulations can be distinguished (using
terminology of Pedley 1990): 1 – valley-side, with
perched spring accumulations on slopes, small in extent
but locally large in thickness (a very frequent type, but
carbonate precipitation is usually limited for this type
under present-day conditions), 2 – fluvial accumula-
tions in surface streams, usually braided but rarely small
barrage types (with less or more intensive active deposi-
tion under present-day conditions), and 3 – lacustrine
and paludal accumulations (inactive under present-day
conditions, deposited in flat valleys, typically during the
Holocene Climatic Optimum). Most of the calcareous
tufa bodies of the Czech Karst show very similar litho-
logical and biostratigraphic patterns, reflecting Holocene

background image



climatic, hydrological and biotic conditions (Ložek
1992). The largest calcareous tufa body in the Czech
Karst is that at Svatý Jan pod Skalou (literally St. John
under Rock), a 17 m thick valley-side-accumulation, rep-
resenting a significant site for the Holocene stratigraphy
and a unique archive of local climate and nature devel-
opment (Ložek 1967; Kovanda 1971; Žák et al. 2001).

Geographical, geological and hydrogeological


The Czech Karst extends between Praha and Zdice

over an area of ca. 200 km


. With its NE margin, this

karst area reaches to the territory of Prague. It is a warm
and dry area lying at 210—500 m a.s.l. with total annual
precipitation of 500 mm and average annual temperature
of 8—9 ºC (Quitt 1971).

The Czech Karst is located in the central part of the

Prague Synform. This basin contains a continuous suc-
cession of Ordovician to Middle Devonian deposits ac-
companied by volcanic activity (Chlupáč et al. 1998).
The Czech Karst itself, located south-west of Prague, is
composed of upper Silurian shales, limestones and ba-
salt lava flows (diabases) plus volcaniclastic rocks
(tuffs) and Lower to Middle Devonian limestones and
shales. During the Late Paleozoic Variscan Orogeny,
sediments of the Prague Synform were folded into com-
plicated anticlinal and synclinal structures and faulted
(Chlupáč et al. 1998).

Groundwaters of the Paleozoic basin fill emerge in

many places in the Czech Karst: along exposed bound-
aries between limestones and non-carbonate rocks near
the bottoms of karstic valleys, and at fault intersections.
Most of the karst springs are characterized by relatively
stable discharge (between 0.1 and 20 l · s


) and tempera-

ture, somewhat exceeding the average annual tempera-
ture of the area in some cases, which resulted from deeper
karst-water circulation (Kadlecová & Žák 1998; Žák et
al. 2001).

Detailed molluscan successions are reported here from

the tufa sequence at Kotýz Ridge (49º55

’12” N,


’55” E), a tufa cascade deposited from karst water

resurging in a spring named “Kotýz Good Water” lying
about 5 km south of Beroun, and from two tufa cascades in
Císařská Gorge (49º55

’39” N, 14º07’47” E), about 6 km

south-east of Beroun (Fig. 1). “Kotýz Good Water” spring
is located on a transversal fault in Silurian shales and lens-
like bodies of basalt flows. The water probably follows the
fault and cracks from the near Devonian Koněprusy Lime-
stone. The spring is the site of continuing precipitation of
calcareous tufa while an inactive tufa cascade is preserved
in the slope about 15 m above the spring. The Císařská
Gorge is a valley 700 m long and over 100 m deep (Kadle-
cová & Žák 1998). Tufa is intensively precipitated from
waters of a spring in the upper part of the valley. Tufa
forms three cascades, marked by numbers I, II and III in the
upstream direction (Fig. 1). Only Cascades I and II were
subjects of this study.

Material and methods

Malacozoological analyses

Acquisition of samples for paleomalacological and bio-

stratigraphic analyses followed the unified methodology
of Ložek (1964). If permitted by the character of the sedi-
ment, samples of the matrix approximately 4 l in volume
were taken from all macroscopically distinguishable beds
within the studied tufa sections. Molluscs were extracted
from the sediments by a combination of washing and siev-
ing. After careful drying, each sample was disaggregated
in water. Floating snails, which included most of the shells
in any sample, were repeatedly decanted into a 0.5-mm
sieve and then dried in laboratory conditions. After, the
sediment was divided into several arbitrary size fractions
by sieving, to facilitate the picking of shells. These were
systematically removed from the sediment fraction, which
was sprinkled onto a black sorting tray and examined un-
der a binocular microscope at variable magnification ( 6

50). The molluscs were picked using an entomologi-

cal pair of delicate forceps. Individual shells (shell apices
and undamaged shells) were counted, but non-apical frag-
ments of species were also scored to whole individuals fol-
lowing the standard methodology (Ložek 1964).

The results are given in the form of tables of ecological

and biostratigraphical categories (Tables 1, 2), including
the number of taxa and the number of individuals. The
percentage frequency mollusc diagram expresses relative
proportions of the total number of species (MSS mala-

Fig. 1. Schematic map of the geographical position of calcareous
tufas at Kotýz Ridge (No. 1) and in the Císařská Gorge (No. 2).
Stars-like symbols mark tufa accumulations Svatý Jan pod Skalou
and Petránka.

background image



cospectra) and total number of individuals (MSI mala-
cospectra) within the principal ecological groups in that
sample (Fig. 3). Attribution of molluscan species to the
ecological categories in the tables and diagram follows
Ložek (1964) and Alexandrowicz (1987). These ecologi-
cal attributions must be regarded as rather general, for sev-
eral species do not fit neatly into any one specific habitat.
Taxonomic nomenclature follows Juřičková et al. (2001).
All the specimens resulting from this study have been de-
posited in the Institute of Geology AS CR in Prague.

Radiocarbon dating analyses

Dating analyses were performed in the Poznań Radio-

carbon Laboratory, Poland. Small pieces of charcoal (after
treatment with hot solutions of acid, alkali, and acid to re-
move all carbonate carbon and easily soluble organics)
were measured by the AMS (Accelerator Mass Spectrome-
try)  method. The 


C data on organic matter have been

calibrated for variable initial 


C concentration using the

OxCal v3.5 calibration program (Bronk 2001).

Kotýz Ridge

The tufa cascade at Kotýz Ridge (No. 1 in Fig. 1) is pre-

served in the slope 15 m above the present karst spring.
The tufa sequence was exposed by a test hole (Fig. 2). The
section can be subdivided into 16 beds, including the lay-
ers underlaining and overlaining the tufa body. Molluscan
shells are present in all layers, excepting Beds Nos. 15, 14,
and 10 (Table 1). Although the basal complex (Beds
Nos. 16 to 10) is of very low species diversity due to poor
fossilization ability, it occasionally yielded species like
Vertigo substriata,  Perpolita petronella and Vitrea crys-
tallina, characteristic of the early Holocene. This complex
completely lacks typical glacial species, however, the ac-
companying species Vallonia costata and V. pulchella in-
dicate the presence of open habitats of dry to mesic
character. The malacofauna from the overlying Beds
Nos. 9 to 1 markedly differs from that of the basal complex
(Beds Nos. 16 to 10) in the fossil malacofauna they con-
tain. The bed of solid structural tufa (Bed No. 9) already
shows a gradual increase in forest species, but the open
ground species still maintain a markedly higher abun-
dance. It was only in this bed that the early Holocene spe-
cies  Discus ruderatus was rarely encountered. From Bed
No. 8 upwards, a rapid boom of forest species occurs at the
expense of open-country elements. Molluscan species in
this bed document closed woodland conditions, also per-
sisting in the overlying complex (Beds Nos. 7 to 4). Fossil
molluscs from Bed No. 8 and from the interval of Beds
Nos. 7 to 4 indicate environmental conditions of the Ho-
locene Climatic Optimum, namely a massive expansion of
woodland molluscan species such as Bulgarica cana,  Vit-
rea diaphana,  Platyla polita,  Vertigo pusilla etc., and
high abundance of the hygrophilous element Carychium
tridentatum,  whereas the open-country elements and the
majority of catholic elements disappeared or became very

Fig. 2. The studied section of the tufa deposit at Kotýz Ridge, fron-
tal view. Lithology: 1 – dark brown humic Rendzic Leptosol with
sporadic limestone and diabase clasts (4 cm in diameter); 2 – dark
yellowish brown slightly humic soil sediment with rare limestone
clasts (0.5 cm); 2a – very dark greyish brown slightly humic soil
sediment; 3 – dark brown slightly humic soil sediment with abun-
dant tufa encrustations;  4 – pale brown fine-grained tufa with
coarse encrustations in the upper portion of the bed and dark stains
coloured with Fe and Mn oxides; 5 – light yellowish brown tufa;
6 – yellowish brown tufa with coarse encrustations in the upper por-
tion of the bed; 7 – very dark greyish brown tufa with rare, very
dark brown Silurian shale clasts; 8 – very dark brown mouldered
slightly loamy tufa with rare shale clasts; 9 – dark greyish brown
solid tufa with shale clasts; 10 – very dark brown shales (clasts
1 cm in diameter), dark greyish brown loamy matrix; 11 – very
dark brown shales (clasts 2 cm in diameter) with tufa encrustations
(30 %), dark brown loamy matrix; 12 – very dark brown shales
(clasts 2 cm in diameter) with sporadic tufa encrustations, very dark
grey matrix; 13 – very dark brown shales (clasts 2 cm in diameter)
with sporadic tufa encrustations, very dark greyish brown matrix;
14 – very dark brown shales (clasts 3 cm in diameter) with tufa
encrustations (20 %), dark stains coloured with Fe and Mn oxides,
dark yellowish brown matrix; 15 – very dark brown shales (clasts
4 cm in diameter) with sporadic tufa encrustations, very dark grey-
ish brown loamy matrix; 16 – very dark brown shales (clasts 5 cm
in diameter), very dark greyish brown loamy matrix.

background image



background image



rare (e.g. Vallonia  excentrica and Truncatellina cylindri-
ca are completely absent from Beds Nos. 6 to 4).

The abundance of open-country species increases and

that of moist woodland species decreases within Beds
Nos. 2 and 3. After the decline of tufa deposition the cas-
cade was covered by dark brown humic soil sediment with
an increasing number of rock fragments redeposited from
the upper part of the slope. The appearance of modern im-
migrants (e.g. Oxychilus cellarius) is characteristic of this
uppermost bed.

Radiocarbon dates and chronology of tufa deposition

Charcoal from Beds Nos. 8 and 4 was subjected to 



dating (Table 3, Fig. 2). Based on the obtained data and
the occurrence of fossil molluscan assemblages, hori-
zons of different age can be distinguished in the tufa ac-
cumulation. Tufa deposition started in the Boreal
period. Solid tufa with rare shale fragments (Bed No. 9)
contains molluscan fauna indicating a moist, light for-
est environment with decreasing abundance of species
Discus ruderatus, 

Trichia sericea and Perpolita

petronella. The underlying Beds Nos. 16 to 10 can thus
generally be dated to the Late Glacial to Preboreal be-
cause of the sporadic occurrence of tolerant molluscan
species and the prevalence of early Holocene elements.
The interval between Beds Nos. 8 and 4 is characterized
by extensive development of woodland communities;
the abundance of hygrophilous elements dominated by
Carychium tridentatum indicates a humid climate. This
interval can be attributed to the Holocene Climatic Op-
timum – Atlantic and Epiatlantic, as was also con-
firmed by 


C dating to the period of 5070—3090 years

BC (Fig. 2). In Bed No. 3, the accumulation of tufa de-
clines, and the recovered molluscan assemblages from
this bed point to a prominent drying and a massive re-
treat of forests, reflected by a prominent spread of open-
country elements such as Truncatellina cylindrica,

Table 2: Assemblages of Mollusca in selected layers of calcareous
tufa in the Císařská Gorge. Mollusc finds relate to Figs. 4 and 5.
Cascade I: sample 1 – the layer of massive porous brown yellow-
ish calcareous tufa; sample 2 – the layer of limestone scree ce-
mented by brown porous carbonate. Cascades II – sample from
the layer of limestone scree cemented by brown porous carbonate.
Ecological and biostratigraphical characteristics see Table 1.

Table 3: AMS radiocarbon age determinations from charcoal re-
covered from the calcareous tufa at Kotýz Ridge.

Table 1: Assemblages of Mollusca in individual layers of calcareous tufa at Kotýz Ridge. Layer number relate to Fig. 2. Ecological
characteristics: General ecological groups: A – woodland (in general); B – open country; C – woodland/open country; D  – water,
wetlands. Ecological groups: 1 – woodland (sensu stricto); 2 – woodland, partly semi-open to open habitats [W(M) – mesic, W(S) – xeric,
W(H) – damp]; 3 – damp woodland; 4 – xeric open habitats [S – in general, S(W) – partly shaded habitats]; 5 – open habitats in
general (moist meadows to steppes). Woodland/open country: 6 – predominantly dry; 7 – mesic or various (Me – mesic in general,
catholic, Wf – mesic rocks, scree woodland); 8 – predominantly damp; 9 – wetlands, banks; 10 – aquatic habitats (S – swamps,
Q – springs, F – running waters, Pp – periodic). Biostratigraphic characteristics: + – loess species, (+)– local or occasional loess
species, ! – species characteristic of warm phases, (!) – eurythermic species of warm phases, !! – index species of warm phases, G – spe-
cies surviving the Glacial out of the loess zone, (G) – ditto as relics, M – modern immigrants (late Holocene index species). Presence
in layers: 1 – number of individuals, 1? – only approximate determination, (1) – contamination, — – absent.

background image



Pupilla muscorum,  Vertigo pygmaea and common rep-
resentatives of the genus Vallonia. This trend continues
in the overlying strata. Beds Nos. 3 to 1 must be therefore
placed to the late Holocene (Subboreal, Subatlantic and
Subrecent), which is also indicated by the appearance of
modern immigrants such as the steppe mollusc Xerolenta
obvia in Bed No. 2 and the ecologically indifferent Oxy-
chilus cellarius in Bed No. 1.

Císařská Gorge

Tufa is intensively precipitated from waters of a spring in

the upper part of the valley. It forms three cascades, marked
by numbers I, II and III in the upstream direction. Only
Cascades I and II were subjects of this study (No. 2 in Fig. 1).

Limestone scree cemented by carbonate is deposited on

the base of the thickest cascade – Cascade I. Above the
scree, a bed of solid structural tufa was deposited, laterally
interfingering with slope sediments. A young tufa accu-
mulation is presently growing in the erosive channel cut-
ting the cascade (Fig. 4). Cascade II, lying ca. 130 m
upstream, has an analogous lithological character (Fig. 5).

In both cascades, samples corresponding to their radio-

metrically dated bed level (Hlaváč et al. 2003) were select-
ed for malacozoological and malacostratigraphic analysis
(sampling points in Figs. 4 and 5, Table 2). Fossil mol-
luscs from the bed of massive porous brown yellowish cal-
careous tufa from Cascade I (Fig. 4) are markedly
dominated by woodland species. The gastropods Sphyra-
dium doliolum,  Platyla polita,  Vertigo pusilla,  Aegopinel-
la pura,  Discus rotundatus,  Macrogastra ventricosa,
Alinda biplicata,  Clausilia pumila and other species were
found in high amounts. These are accompanied by wood-
land species of medium size (Monachoides incarnatus,
Urticicola umbrosus,  Cepaea hortensis), again highly
abundant. Catholic species of ecologic group C reach
higher abundance than open-country and forest-free ele-
ments, which are considerably less frequent and restricted
only to the species Vallonia costata and the rare Trunca-
tellina cylindrica, probably transported from the ambient
steep slopes. Fossil malacocoenosis clearly indicates a
moist, closed forest covering the valley area, which was
getting lighter towards higher elevations.

A poor molluscan assemblage dominated by Vallonia

costata,  Punctum pygmaeum and Vitrea crystallina was

Fig. 3. Malacospectra MSS and MSI of Kotýz Ridge. Left-side column – layer, middle – chronology (LG – Late Glacial, PR – Prebo-
real, B – Boreal, A – Atlantic, EA – Epiatlantic, SB – Subboreal, SA – Subatlantic, SR – Subrecent), right-side column – total num-
ber of individuals.

background image



Fig. 4. Tufa Cascade I in the Císařská Gorge, frontal view. Lithology: 1 – the youngest calcareous tufa; 2 – massive porous brown yel-
lowish calcareous tufa; 3 – limestone scree cemented by brown porous carbonate, angular clasts with average size of 4 cm; clast size up to
10 cm in the upper part; 4 – grey loose limestone scree; 5 – limestone bedrock; 6 – sampling point on fossil molluscs.

Fig. 5. Tufa Cascade II in the Císařská Gorge, frontal view. Lithology: 1 – the youngest calcareous tufa; 2 – massive porous yellowish
brown calcareous tufa; 3 – limestone scree cemented by brown porous carbonate, angular clasts with average size of 4 cm, clast size up to
10 cm in the upper part; 4 – grey loose limestone scree; 5 – sampling point on fossil molluscs. Detail “A” shows a bed of dated carbonate
precipitated on plant roots. Detail “B” shows the position of carbonate laminae precipitated in a small cavity developed in limestone scree.

recorded in the underlying bed of limestone scree cement-
ed by brown porous carbonate from Cascade I (Fig. 4).
These tolerant species were occasionally accompanied by
the early Holocene element of Discus ruderatus and eco-
logically indifferent Cochlicopa lubrica together with Eu-
conulus fulvus. The rate of woodland species in this bed

was very low being represented only by the species Mona-
choides incarnatus and Platyla polita. A similar snail as-
semblage was also observed in the sample collected from
the bed of limestone scree cemented by brown porous car-
bonate from Cascade II (Fig. 5). The common Discus rud-
eratus and Perpolita petronella are associated with

background image



Vertigo alpestris,  Discus rotundatus,  Aegopinella minor,
Trichia hispida and Ena montana, accompanied by the in-
different species Punctum pygmaeum,  Cochlicopa lubrica
and  Euconulus fulvus. Both of these beds are completely
lacking any of the prominent elements of continuous for-
est characteristic of the Holocene Climatic Optimum or
the youngest Holocene.

Radiometric dates and chronology of tufa deposition

As indicated by molluscan remains in the beds of lime-

stone scree cemented by brown porous carbonate from
Cascades I and II, their fill can be dated to the early Ho-
locene. This is suggested not only by the common occur-
rence of early Holocene species Discus ruderatus,  Perpolita
petronella,  Vitrea crystallina and even Vertigo alpestris,
but also by the notable absence of elements indicative of a
continuous forest of the Climatic Optimum. The early Ho-
locene age of the scree is also confirmed by the radiometric
age of its carbonate matrix – 9460 ± 1200 years BP (Hlaváč
et al. 2003). The encountered assemblages indicate rather
the Boreal period (sample from Cascade I), or possibly a
transition between the Boreal and early phases of the Atlan-
tic (sample from Cascade II) with carbonate matrix dated at
5920 ± 1520 years BP (Hlaváč et al. 2003). The landscape of
this period had the character of open areas alternating with
patches of shady deciduous forests having, however, a dif-
ferent composition than the present ones. Moisture condi-
tions were already more favourable but still did not permit
the development of more continuous woodland complexes,
as documented by the notable absence of demanding wood-
land molluscan species or only a low proportion of hydro-
philous elements.

In contrast with the underlying beds, malacostratigraph-

ic analysis of a sample from the bed of massive porous
brown yellowish calcareous tufa from Cascade I (Fig. 4)
yielded malacofauna indicating the Holocene Climatic
Optimum – Atlantic and Epiatlantic (sensu  Jäger 1969).
The evidence for this is the almost complete absence of
open-country elements as well as by the lack of early Ho-
locene elements or elements from the youngest phases of
the Holocene. In a time succession, the Císařská Gorge
was first filled with talus with angular scree fragments in
the early Holocene (Preboreal, Boreal to early phases of
the Atlantic). The scree was later overlain by accumula-
tions of structural tufa precipitated during the Holocene
Climatic Optimum (Atlantic, Epiatlantic).

Discussion and conclusions

Holocene development of tufa cascades of the Czech

Karst shares many common features conditioned by climat-
ic and hydrologic influence. The formation of cascade, on
the bottoms of karstic valleys, must have postdated the end
of intensive fluvial erosion characteristic of the Glacial/Ho-
locene boundary (see Vandenberghe 1993; Vandenberghe
et al. 1994, a.o.). The tufa is usually underlain by limestone
talus derived from the ambient slopes. Thicknesses of pre-

served talus depend largely on the character of the source
Paleozoic rocks forming valley slopes (e.g. stratification
and tectonic deformation) and on the erosive potential of
the stream at the site the cascade started to develop. Mollus-
can assemblages found in talus beneath the tufa are evi-
dence of the Boreal age of these colluvia.

Talus in the Císařská Gorge is secondarily cemented by

carbonate precipitated from karst waters. Radiometric age
of the cement was determined at 9460 ± 1200 years BP and
5920 ± 1520 years BP (Hlaváč et al. 2003). The tufa cas-
cade at Svatý Jan pod Skalou also started to accumulate in
the Boreal (Žák et al. 2001, 2002). Massive structural tufas
with minimum clastic component were deposited on all
cascades under the climatically favourable condition of
the Atlantic. Accumulations of these structural tufas reach
up to 2—5 m in thickness (Kovanda 1971; Ložek 1992).
Deposition of tufa continued in the late period of the Ho-
locene Climatic Optimum (Epiatlantic). Short-term climat-
ic oscillations, however, occurred in the Epiatlantic, being
characterized by alternation of periods of relative aridity
and humidity. These short-term climatic oscillations are
indicated by horizons of initial carbonate soils and by ta-
lus intercalations in the tufa successions. Climatic oscilla-
tions culminated in a markedly dry and warm period in the
Subboreal (e.g. Jäger 1969). As a result, the sedimentary
record in younger portions of the tufa cascades is litholog-
ically more variable, with alternating beds of friable and
compact tufa, colluvio-fluvial sediments and soil horizons
(Žák et al. 2001). Principal accumulation stages of tufa
cascades in the Czech Karst are terminated in the late Sub-
boreal at 2500 BP (Žák et al. 2002).

The decline of cascade development was due to the

coming dry period, probably manifested in a discharge re-
duction of the karst springs. Reduced water discharges
could have resulted from subrosion, which created new
paths in the basal portions of the tufa bodies. Karst waters
ceased to flow on top of the cascade surfaces. The onset of
this dry period dates to the Late Bronze Age (LBA); it re-
sulted in dramatic changes in vegetation and in molluscan
communities, but also in a decrease in the activity of river
systems. These phenomena have been documented from
many regions of the Czech Republic. A major environ-
mental change dating to the LBA has been documented
from central and northern Bohemia on the basis of multi-
disciplinary study of sediments deposited under sandstone
rock-shelters. Changes in molluscan communities indicate
deforestation associated with more intensive agricultural
use of land (Svoboda et al. 1996; Cílek et al. 1996).

It can be assumed that erosion and destruction of tufa

cascades in the Czech Karst occurred during several flash
precipitation events in the Little Ice Age (LIA) and due to
increased precipitation and severe floods in the 20



tury. Larger volumes of water flowing through the
Císařská Gorge during the LIA are also indicated by the
radiometric age of the carbonate precipitated on plant
roots in Cascade II (Hlaváč et al. 2003). The carbonate was
precipitated on the root at 300 ± 600 years BP, namely at
the time of the undoubted existence of the fracture de-
stroying the cascade (see above). At that time, however,

background image



water must have been still flowing across the upper part of
the cascade.

The molluscan assemblages detected in the tufa bodies

show many common features. Malacospectral analyses of
tufa at Kotýz Ridge (Fig. 2) revealed the proportions of
molluscs of the main ecological groups. As the underlying
complex of the tufa bodies at Kotýz (Beds Nos. 16—11)
contain very poor malacozoological material, the ratio be-
tween the woodland malacofauna component and open-
habitat component or indifferent species is biased to a
considerable degree. In contrast, the overlying beds con-
tain high numbers of molluscan species already and are
fully eligible for statistical evaluation. As shown by the
malacospectra, the overlying beds of the Atlantic-Epiat-
lantic Climatic Optimum marked by intensive tufa accu-
mulation are dominated by woodland species in the
malacocoenoses, while open-country species are scarcely
represented (Svatý Jan pod Skalou, Kotýz) or present in
very low proportions at Petránka (Hlaváč et al. 2003). This
indicates closed woodland formations with prevalence of
moisture demanding molluscan species. After the Epiat-
lantic phase, climatic oscillations culminate with the Sub-
boreal period characterized by a reduced woodland
component and the onset of open-country elements. This
is particularly obvious from the malacospectra of tufas
from Svatý Jan pod Skalou and Kotýz (Hlaváč et al. 2003).
In the Petránka tufa in the Karlické Valley, the Subboreal
period is characterized by the abundant occurrence of gas-
tropod  Vallonia costata, a typical open-country species,
however, the woodland component still maintains a clear
dominance. Molluscan assemblages from the youngest
Holocene phases often share the features of modern com-
munities as shown by molluscan occurrences in the neigh-
bourhood of the tufa bodies (Ložek 1974; Hlaváč 2002).

The Kotýz Ridge and Císařská Gorge assemblages con-

tain a number of species that are absent in the Czech Karst
at present. Indeed, they include the boreo-montane species
Discus ruderatus,  Vertigo substriata,  Vertigo alpestris,
and  Perpolita petronella, that are confined to montane
and submontane zones in the Czech Republic at present,
while during early Holocene they formed the leading in-
dex  Discus ruderatus-fauna (Ložek 1964). Other species,
for instance Bulgarica cana, although occurring during
the Holocene Climatic Optimum at Kotýz Ridge, or in the
Císařská Gorge, and at Svatý Jan pod Skalou (Žák et al.
2001, 2002) is now extinct in the Czech Karst and even
extremely rare in the Czech Republic. These early Ho-
locene land snail assemblages have no exact modern ana-
logues in the Czech Republic, whereas the malacofaunas
of the Climatic Optimum have survived in a few relic for-
est habitats, for instance in the near by Křivoklát Area.

There are relatively few continental sites that have been

investigated in such details as these Kotýz Ridge and
Císařská Gorge sequences and those often lack secure ra-
diocarbon dates. Much work has been undertaken in Cen-
tral Europe, particularly in the Czech Republic and
Slovakia (e.g. Ložek 1964, 1982), Hungary (Füköh et al.
1995) and Poland (e.g. Alexandrowicz 1983). Only the
tufa deposits at Svatý Jan pod Skalou and Švarcava

(Czech Republic) as well as Slovenská dolina Valley (Slo-
vak Republic) provide radiocarbon dates (Šilar & Ložek
1988; Žák et al. 2001). In addition, a considerable number


C-dated Holocene snail successions are available

from archaeological excavations in North-Bohemian sand-
stone rock-shelters (Svoboda 2003) and from calcareous
tufa deposits in southern Poland (Alexandrowicz 2004).
Faunal development shows strong parallels with the suc-
cession described from the Czech Karst. Early Holocene
assemblages are rather different than present-day mollus-
can faunas at these sites. Also middle Holocene ones are
again far richer than the faunas living in those regions to-
day. The species composition described from Western Ger-
many and Luxembourg are, however, quite different. The
species  Spermodea lamellata and Leiostyla anglica char-
acterize middle Holocene assemblages there (Meyrick
2001, 2003), whereas they are unknow in Central Europe.
Some work has recently been undertaken on sequences in
Britain and close to Britain (Preece & Day 1994; Limon-
din-Lozouet 1998; Meyrick & Preece 2001), but these re-
gions are too distant for meaningful comparisons. Further
discussion should be deferred until more mid-European
sites have been investigated. Only then it will be possible
to consider the Czech data in its full European and British


The research has been supported by

the Grant No. IAA300130505 of the Grant Agency of the
Academy of Sciences, Czech Republic. The paper is a part
of the Research Plan of the Institute of Geology, Academy
of Sciences of the Czech Republic (No. AV0Z30130516).


Alexandrowicz S.W. 1983: Malacofauna of Holocene calcareous sed-

iments of the Cracow Upland. Acta Geol. Pol. 33, 117—158.

Alexandrowicz S.W. 1987: Malacological analysis in Quaternary re-

search.  Kwart. AGH, Geologia  13, 1—240 (in Polish, extended
English summary).

Alexandrowicz W.P. 2004: Molluscan assemblages of Late Glacial

and Holocene calcareous tufas in southern Poland. Folia Qua-
ternaria 75, 3—309.

Bronk R.C. 2001: Development of the radiocarbon program OxCal.

Radiocarbon 43, 2A, 355—363.

Chlupáč I., Havlíček V., Kříž J., Kukal Z. &  Štorch P. 1998: Palae-

ozoic of the Barrandian. Czech Geol. Surv., Praha, 1—183.

Cílek V., Jarošová L., Karlík M., Ložek V., Mikuláš R., Svoboda J.

& Škrdla P. 1996: Research of sandstone rockshelters in the
NW part of Protected landscape area Kokořínsko. Part II.
Ochrana přírody 51, 3, 82—85 (in Czech).

Füköh L., Krolopp E. & Sümegi P. 1995: Quaternary malacos-

tratigraphy in Hungary. Malacological Newsletter Supplement
1, 1—219.

Gorka P. & Hercman H. 2002: URANOTHOR v. 2.5. Delphi Code

of calculation program and user guide. MS Arch. Quat. Geol.
Department, Inst. Geol.Sci., PAS, Warsaw.

Goudie A.S., Viles H.A. & Pentecost A. 1993: The late-Holocene

tufa decline in Europe. The Holocene 3, 2, 181—186.

Hlaváč J. 2002: Malacofauna of the Koněprusy Area (Czech

Karst) – woodland, open grounds and secondary habitats.
Český kras, Beroun 27, 4—8  (in Czech).

background image



Hlaváč J., Kadlec J., Žák K. & Hercman H. 2003: Deposition and

destruction of Holocene calcareous tufa cascades in the Czech
Karst (Czech Republic). Geographical Studies 189, 225—253.

Ivanovich M. & Harmon R.S. 1992: Uranium series disequilibrium:

Application to environmental problems. 2


 Ed. Clarendon

Press, Oxford, 1—910.

Jäger K.-D. 1969: Climatic character and oscillations of the subboreal

period in the dry regions of the Central European Highlands.
Proceedings of the 7


 Cong. INQUA, Washington, 38—42.

Juřičková L., Horsák M. & Beran L. 2001: Check-list of the mol-

luscs (Mollusca) of the Czech Republic. Acta Societatis Zoo-
logicae Bohemiae 65, 25—40.

Kadlecová R. & Žák K. 1998: Karst springs of the Bohemian Karst.

Český kras, Beroun 24, 17—34 (in Czech).

Kovanda J. 1971: Quaternary limestones of the Czechoslovakia.

Sbor. Geol. Věd, Ř. Antropozoikum 7, 1—236 (in Czech).

Limondin-Lozouet N. 1998: Successions malacologiques du tardigla-

ciaire Weichsélien: corrélations entre série du Nord de la France
et du sud-est de la Grande-Bretagne. Quaternaire 9, 217—225.

Ložek V. 1964: Quartärmollusken der Tschechoslowakei. Rozpr.

Ústř. Úst. Geol. 31, 1—376.

Ložek V. 1967: Holozäne Binnenwasserkalke und klastische Hang-

sedimente im Böhmischen Karst. In: Kliewe H. (Ed.): Prob-
leme und Befunde der Holozänstratigraphie in Thüringen,
Sachsen und Böhmen. INQUA, Berlin-Prague, 137—178.

Ložek V. 1974: Molluscs of the Bohemian Karst from a nature

protection point of view. Bohemia Centralis 3, 163—174 (in

Ložek V. 1982: Faunengeschichtliche Grundlinien zur spät- und

nacheiszeitlichen Entwicklung der Molluskenbestände in Mit-
teleuropa. Rozpr. ČSAV, Ř. MPV 92, 4, 1—106.

Ložek V. 1992: Net of key sections to landscape development of

the Bohemian Karst. Bohemia Centralis 21, 47—67 (in Czech).

Meyrick R.A. 2001: The development of terrestrial mollusc faunas in

the Rheinland region (Western Germany and Luxembourg)
during the Late Glacial and Holocene. Quat. Sci. Rev. 20, 16—
17, 1667—1675.

Meyrick R.A. 2003: Holocene molluscan faunal history and envi-

ronmental change at Kloster Muhle, Rhein-Pfalz, western Ger-
many.  J. Quat. Sci. 18, 2, 121—132.

Meyrick R.A. & Preece R.C. 2001: Molluscan successions from two

Holocene tufas near Northampton, English Midlands. J. Bio-
geography 28, 77—93.

Pedley H.M. 1990: Classification and environmental-models of

cool fresh-water tufas. Sed. Geol. 68, 1—2, 143—154.

Preece R.C. 1991: Mapping snails in time: the prospect of elucidat-

ing the historical biogeography of the European malacofauna.
Proceedings of the Tenth Internal Malacological Congress,
Tübingen, 1989, 477—479.

Preece R.C. & Day S.P. 1994: Comparison of Post-glacial mollus-

can and vegetational successions from a radiocarbon-dated
tufa sequence in Oxfordshire. J. Biogeography 21, 463—478.

Svoboda J. (Ed.) 2003: The Mesolithic of Northern Bohemia. Dol-

nověstonické studie 9, 1—328 (in Czech, extended English

Svoboda J., Opravil E., Škrdla P., Cílek V. & Ložek V. 1996: Me-

solithic from the regional perspective: New excavations in the
Polomené Mts. Archeogické rozhledy 48, 3—15 (in Czech).

Šilar J. & Ložek V. 1988: Dating of Holocene carbonate sediments

from the Slovenská dolina valley at Valča (District of Martin).
Československý kras 39, 69—76 (in Czech).

Quitt E. 1971: Climatic regions of Czechoslovakia. Stud. Geo-

graphica 16, 1—86 (in Czech).

Vandenberghe J. 1993: Changing fluvial processes under changing

periglacial conditions. Z. Geomorph. N.F. 88, 17—28.

Vanderberghe J., Kasse C., Bohncke S. & Kozarski S. 1994: Cli-

mate-related river activity at the Weichselian-Holocene transi-
tion: a comparative study of the Warta and Maas rivers. Terra
Nova 6, 476—485.

Weisrock A. 1986: Variations climatiques et periodes de sedimenta-

tion carbonatée a l’holocene – l’age des depots. Mediterranée
10, 165—67.

Žák K., Hladíková J., Buzek F., Kadlecová R., Ložek V., Cílek V.,

Kadlec J., Žigová A., Bruthans J. & Š astný M. 2001: Ho-
locene limestones and karst springs at Svatý Jan pod Skalou.
Práce Českého Geol. Úst. 13, 1—136 (in Czech, extended En-
glish summary).

Žák K., Ložek V., Kadlec J., Hladíková J. & Cílek V. 2002: Cli-

mate-induced changes in Holocene calcareous tufa formations,
Bohemian Karst, Czech Republic. Quat. Int. 91, 137—152.