Theoretical and Applied Karstology

Material Information

Theoretical and Applied Karstology
Series Title:
Theoretical and Applied Karstology
Emil Racovita Institute of Speleology
Emil Racovita Institute of Speleology
Publication Date:


Subjects / Keywords:
Geology ( local )
serial ( sobekcm )
46 x 25


General Note:
Editor's Page S. Constantin p. 4. In Memoriam Stiuca, E. - Constantin (Costin) Radulescu, 1932-2002. p. 5. TAK Reviews Forti, P. Speleology in the Third Millenium: Achievments and Challenges, pp. 7-26. Men have visited caves for a long time, but speleology started about 20 000 yr BP when the human frequentation of caves was no longer "accidental". The development of caving activities may be subdivided into three periods: prehistory (in which no written report exists), protohistory (in which documentation is available but speleology is not yet that of today) and history (which conventionally started with E. A. Martel). Nowadays caving activities are split into several branches which may be grouped in four categories: explorative, scientific, social and documentary speleology. In the present paper after a short overview on the development of caving activities from prehistory until present day the challenges and goals for the third millennium are outlined, the main problems speleology will have to face being: * exploration of the ice caves in Antarctica and of the volcanic caves in the space; * scientific multidisciplinary investigation of special cave ecosystems; * extensive search for new drinking water supply and new principles in medicine; * environmental protection and sustainable tourism. TAK Articles Onac, B.P., Breban R., Kearns, J., Tamas, T. - Unusual minerals related to phosphate deposits in Cioclovina Cave, Sureanu Mts. (Romania). pp. 27-34. Cioclovina Cave hosted an extensive phosphate deposit estimated at over 50,000 m3. About 30,000 m3 were mined during the first half of the XXth Century and used as fertilizers. The mineralogy of this deposit is remarkable as it consists of several rare and/or unusual cave mineral species, mostly phosphates. Cioclovina Cave is the type locality for ardealite. Over 40 samples were analyzed by means of X-ray diffraction and fluorescence, scanning electron microscopy, optical observations on thin sections, and electron microprobe. Out of 26 minerals presented in this paper, 13 have not been previously documented from this cave (berlinite, burbankite, churchite, chlorellestadite, foggite, paratacamite, collinsite, monetite, fluorapatite, sampleite, romanechite, leucophosphite and todorokite). Furthermore, the first six minerals on this list were for the first time identified within the cave environment. At least one rare mineral species (berlinite or chlorellestadite) may have been produced by spontaneous combustion of bat guano, whereas the other ones formed within the sediment fill indicate reactions between phosphate-rich solutions and limestone bedrock, clays, sandstones or various trace elements. These reactions took place at different pH values producing specific minerals that may have environmental deposition significance. Dumitras, D., Marincea, S., Diaconu, G., Constantina C., Pavel, R. - X-ray powder data on some mineral species from Pestera Curata de la Nandru (Hateg Basin, Romania). pp. 35-42. In the current contribution, we briefly characterize the main mineral species from the fossil bat-guano deposit in Pestera Curatã de la Nandru cave, based on extensive X-ray powder diffraction study. The layered guano deposit inside the cave was well opened by recent archaeological works. Hydroxylapatite and brushite are the most common constituents; associated minerals are quartz, dolomite, calcite and clay minerals (illite and kaolinite). The mean cell parameters of hydroxylapatite, taken as weighed average of nine sets of values obtained by least-squares refinement of X-ray powder data, are a = 9.429(6) Å, c = 6.862(16) Å and V = 528.9(6) Å3. These values account for the stoichiometry, as well, as in the case of brushite [a = 5.808(7) Å, b = 15.183(1) Å, c = 6.241(8) Å, b = 116.38(6) Å3 for a representative sample]. Calcite [a = 4.980(2) Å, c = 17.033(7) Å, V = 365.8(1) Å3] and dolomite [a = 4.807(9) Å, c = 16.062(6) Å, V = 321.4(1) Å3] occur on diagenetic cracks that affect the deposit. Low (alpha) quartz with a = 4.920(3) Å, c = 5.406(4) Å and V = 113.3(1) Å3, probably allogenic, and minor kaolinite and illite are admixed with the phosphates from the guano groundmass. Calic-Ljubojevic, J. Ljubojevic, V. - Karst of the ridge Djevrinska Greda: fluvial influences, caves and groundwater circulation. pp. 43-50 D?evrinska Greda is an elongated and narrow ridge of Upper Jurassic limestone, situated next to the Danube Gorge (Iron Gates) in Eastern Serbia. It is uplifted along the conspicuous D?evrin Fault which extends further to the north, to the Mehedinþi Plateau in Romania, and it is surrounded by non-carbonate rocks to the East (Cretaceous flysch and para-flysch) and to the West (Proterozoic and Paleozoic schists). The relationships between non-carbonate and carbonate lithologies made this karst subject to strong influence of allogenic water input. Due to the small width of the ridge (max. 700 m, 250 m in average), fluvial influence is strong enough to penetrate to the opposite boundary of the limestone. Several separate input-output systems of karst groundwater were determined. Exposed limestone surface does not exceed 5 km2, but 32 caves which altogether have more than 6500 m of passages were explored so far, indicating a significant karst development. The springs are mostly permanent, although relatively weak (up to 10 l/s), and show little seasonal variations. Three springs have elevated temperatures - from 17 to 19 °C. Characteristics of karst springs are the indication of retention capabilities of the karst aquifer as well as of a circulation of the groundwater at great depths. Deep circulation is mostly developed due to the favourable conditions along the regional dislocation - the D?evrin Fault. Jemcov, I., Pavlovic, R., Stevanovic, Z. - Morphotectonic analysis in hydrogeological research of karst terrains. A case-study of SW Kucaj Massif, Eastern Serbia. pp. 51-59. The fracture structures are recognized as being among the principal factors of karst development by analyzing data obtained by remote sensing. Regional fractures were recognized through analysis of satellite imagery - scanograms, while the detailed fault pattern was obtained by stereoscopic analysis of aerial photographs. The method of quantitative geomorphologic analysis was used for the identification of neotectonically active structures. Neotectonic analysis was performed using a morphometric-statistical procedure - the calculation of topographic relief. As a new procedure, supplemental to fault pattern analysis, Digital Elevation Model (DEM) was applied. Comparison of the results obtained by fault pattern analysis using remote sensing, quantitative geomorphologic analysis and digital elevation model with favored directions of karst groundwater flow, revealed a significant control of the position and function of the faults over karstification. General karst groundwater directions and drainage reorientation are mainly controlled by the morphotectonic evolution. Tenu, A., Davidescu, F., Petres R., Coarna, L. - Environmental isotopes studies and the hydrogeological model of South Dobrogea (Romania). pp. 61-72. Due to its natural and anthropic features, South Dobrogea (Romania) is a very interesting area from a hydrogeological point of view. In the region there are two superposed calcareous aquifers - the upper aquifer (Sarmatian) and the lower aquifer (Barremian-Jurassic) - which form the so-called "karstic system" of a strategic importance. Isotopic monitoring (3H, 14C, 13C, D, 18O) was performed over the last 25 years by an integrated research of all natural types of waters, in order to improve the knowledge of hydrogeological parameters and of the regional pattern flow. The main isotopic characteristics of meteoric and surface waters are briefly described in the paper. Within the karstic system, our study focused on the lower aquifer for which an up-to-date regional model was carried out by correlating all hydrodynamic and isotopic information. The main isotopic features can be synthesized as follows: the recharge area is located mainly in the Pre-Balkan Platform (Bulgaria); the main groundwater flow direction is east-northeast, towards Lake Siutghiol and the flow velocities, at regional scale, vary from 100 m/year for the secondary groundwater flow to 500-1,800 m/year for the main one. Taborosi, D. - Biokarst on a tropical carbonate island: Guam, Mariana Islands. pp. 73-91. "Biokarst" refers to erosional and depositional karst features formed by direct biologic action. However, since no distinct karst landform is produced exclusively by biologic action, most features known as biokarst are actually products of both biotic and abiotic processes operating concurrently in intricate interrelationships. Although biokarst landforms are often regarded as features of local interest and limited significance, the effects of living organisms on karst geomorphology are profound on an ecosystem scale, and are widespread, diverse and of fundamental importance. This is especially true in tropical and coastal environments, where the biologic influence on karst is so common and intense that it makes the distinction between "biokarst" and "non-biokarst" thoroughly impracticable. A survey of karst on Guam has confirmed that nearly all of the small- and medium-scale karst features documented on the island are affected by biota, and bear characteristics of what has been termed biokarst. Kohler, H., C. Karfunkel, J. - The Quaternary morphogenesis of the Lagoa Santa tropical karst, Minas Gerais State, SE Brazil. pp. 93-99; +1 folded map. The tropical karst of the Lagoa Santa region, north of Belo Horizonte, in South-Eastern Brazil was studied, in order to describe the Quaternary morphogenesis and to elucidate the morphodynamic evolution of this area. The karst developed over an interfluvial block, at altitudes between 650 and 850 m above sea level. Structural alignments, consequence of the Brasilian geotectonic cycle, were reactivated through the Cenozoic. These alignments, together with climate oscillations through time, control the main trends of the karst scenario and allow the authors to postulate a morphogenetic history from Tertiary through Holocene. Erbajeva, M., Khenzykhenova, F., Alexeeva, N. - Small mammals of the cave sites in the Baikalian region. pp. 101-110. This paper deals with the small mammal assemblages from the cave sites of the Baikalian region. In the region there are more than 300 caves and shelters (Filippov, 1993a, 1993b), some of which contain numerous small mammal fossils. The oldest fauna is the Middle Miocene from Aya cave containing Eurolagus, Heterosminthus and Gobicricetodon. In other caves, more than 30 taxa of insectivores, lagomorphs and rodents were discovered, mainly of Late Pleistocene and Holocene age. Most of the small mammals belong to modern species, which inhabit the region nowadays; however, specimens of Lagurus lagurus and Dicrostonyx sp., whose area of distribution lies outside of the region currently, were also found. Stiuca, E. Arghir, R. - Martes genus representatives in in the Wurmian of Romania. pp. 111-116. Remains of various mustelides (Martes martes, Martes foina, Mustela nivalis, Putorius putorius) have been recovered from the cave Pestera nr. 4 din Scocul Scorotei (Retezat Mountains, Hunedoara county, Romania). This new fossil site of the Southern Carpathians may be assigned to the Late Pleistocene (Middle Würmian). Allometrical and morphological study of the skulls and dentition indicate a climate warming that corresponds to the lower sediment layers (3 and 2), followed by a cooling that corresponds to the upper layer, as well as the existence, during the Würmian of two different types of Martes. Kranjc, A. - The history of karst ressources exploitation: an example of iron industry in Kranjska (Slovenia). pp. 117-123. In Slovenia the karst surface covers 43% of the territory. About 57% is Dinaric and 21% Alpine karst. Dinaric karst consists of Jurassic and Cretaceous limestone, while Alpine karst of Triassic carbonate rocks. Everywhere the iron ore is present. Two types can be distinguished, the form of limonite pieces, and the "ironstone" (iron hydroxides), so-called "broad bean ore". In the 8th Century BC the iron industry prospered in two karst regions: in the Alps and on Dinaric karst. Technology of extracting iron from the ore was simple. The early metallurgists gathered iron ore on the surface and in karst underground. The ore was melted in shaft furnaces with dug-in fireplaces. Because of the ironwork technology, iron industry was scattered all over the country thus no important concentrated pollution appeared. Yet, local impact could have been important in places. The consumption of wood and charcoal was significant including cutting down the forests. In the 10th Century iron industry restarted. Between the 13th and the 15th centuries ironworks began to develop, based on water energy. Extracting and working of iron moved from the plateaux down to the valleys and the ironworks (joint of smelting furnace, fireplaces to heat the iron red-hot, bellows and ironwork hammer) resulted. In 1581, the list of ironworks for Kranjska enumerates 20 smelting furnaces. The direct impact on karst consists of traces of digging on the surface, in dolines, caves and potholes, disused mine shafts and galleries, and "ore pools" scattered across the land. As a direct impact, large quantities of charcoal were consumed and the water was polluted. There were much less forests left than nowadays. The indirect impact consisted mainly in the high concentration of industrial plants and inhabitants. Rapid communication Vremir, M. M. Ridush, B. - Recent paleontological investigations in some caves of the Crimean mountain-range (SE Ukraine). pp. 125-132. Preliminary results of the most recent speleo-palaeontological investigations made in the Crimean high-mountains are presented. Several caves were investigated in the Chatyrdag massif; also new information regarding fossil remains from caves in several other plateaus (Aj-Petri, Yaltinskaya, Karabi) are added. The most interesting results come from the Emine Bair Khozar cave, which significantly contribute to highlight the Late Pleistocene and Holocene vertebrate faunas existing in the remote mountains of the Crimea. Book Reviews Goran, C. - Monografia carstului din Muntii Padurea Craiului. pp. 133-134.
Open Access - Permission by Publisher
Original Version:
Vol. 15 (2002)
General Note:
See Extended description for more information.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-04805 ( USFLDC DOI )
k26.4805 ( USFLDC Handle )
11648 ( karstportal - original NodeID )
1012-9308 ( ISSN )

USFLDC Membership

Karst Information Portal

Postcard Information



This item has the following downloads:

Full Text
Editor's Page S. Constantin p. 4. In Memoriam Stiuca, E.
- Constantin (Costin) Radulescu, 1932-2002. p. 5. TAK Reviews
Forti, P. Speleology in the Third Millenium: Achievments and
Challenges, pp. 7-26. Men have visited caves for a long time,
but speleology started about 20 000 yr BP when the human
frequentation of caves was no longer "accidental". The
development of caving activities may be subdivided into three
periods: prehistory (in which no written report exists),
protohistory (in which documentation is available but
speleology is not yet that of today) and history (which
conventionally started with E. A. Martel). Nowadays caving
activities are split into several branches which may be grouped
in four categories: explorative, scientific, social and
documentary speleology. In the present paper after a short
overview on the development of caving activities from
prehistory until present day the challenges and goals for the
third millennium are outlined, the main problems speleology
will have to face being: exploration of the ice caves in
Antarctica and of the volcanic caves in the space; scientific
multidisciplinary investigation of special cave ecosystems;
extensive search for new drinking water supply and new
principles in medicine; environmental protection and
sustainable tourism. TAK Articles Onac, B.P., Breban R.,
Kearns, J., & Tamas, T. Unusual minerals related to
phosphate deposits in Cioclovina Cave, Sureanu Mts. (Romania).
pp. 27-34. Cioclovina Cave hosted an extensive phosphate
deposit estimated at over 50,000 m3. About 30,000 m3 were mined
during the first half of the XXth Century and used as
fertilizers. The mineralogy of this deposit is remarkable as it
consists of several rare and/or unusual cave mineral species,
mostly phosphates. Cioclovina Cave is the type locality for
ardealite. Over 40 samples were analyzed by means of X-ray
diffraction and fluorescence, scanning electron microscopy,
optical observations on thin sections, and electron microprobe.
Out of 26 minerals presented in this paper, 13 have not been
previously documented from this cave (berlinite, burbankite,
churchite, chlorellestadite, foggite, paratacamite, collinsite,
monetite, fluorapatite, sampleite, romanechite, leucophosphite
and todorokite). Furthermore, the first six minerals on this
list were for the first time identified within the cave
environment. At least one rare mineral species (berlinite or
chlorellestadite) may have been produced by spontaneous
combustion of bat guano, whereas the other ones formed within
the sediment fill indicate reactions between phosphate-rich
solutions and limestone bedrock, clays, sandstones or various
trace elements. These reactions took place at different pH
values producing specific minerals that may have environmental
deposition significance. Dumitras, D., Marincea, S., Diaconu,
G., Constantina C., & Pavel, R. X-ray powder data on some
mineral species from Pestera Curata de la Nandru (Hateg Basin,
Romania). pp. 35-42. In the current contribution, we briefly
characterize the main mineral species from the fossil bat-guano
deposit in Pestera Curat de la Nandru cave, based on extensive
X-ray powder diffraction study. The layered guano deposit
inside the cave was well opened by recent archaeological works.
Hydroxylapatite and brushite are the most common constituents;
associated minerals are quartz, dolomite, calcite and clay
minerals (illite and kaolinite). The mean cell parameters of
hydroxylapatite, taken as weighed average of nine sets of
values obtained by least-squares refinement of X-ray powder
data, are a = 9.429(6) , c = 6.862(16) and V = 528.9(6) 3.
These values account for the stoichiometry, as well, as in the
case of brushite [a = 5.808(7) , b = 15.183(1) , c = 6.241(8)
, b = 116.38(6) 3 for a representative sample]. Calcite [a =
4.980(2) , c = 17.033(7) , V = 365.8(1) 3] and dolomite [a =
4.807(9) , c = 16.062(6) , V = 321.4(1) 3] occur on
diagenetic cracks that affect the deposit. Low (alpha) quartz
with a = 4.920(3) , c = 5.406(4) and V = 113.3(1) 3,
probably allogenic, and minor kaolinite and illite are admixed
with the phosphates from the guano groundmass.
Calic-Ljubojevic, J. & Ljubojevic, V. Karst of the ridge
Djevrinska Greda: fluvial influences, caves and groundwater
circulation. pp. 43-50 D?evrinska Greda is an elongated and
narrow ridge of Upper Jurassic limestone, situated next to the
Danube Gorge (Iron Gates) in Eastern Serbia. It is uplifted
along the conspicuous D?evrin Fault which extends further to
the north, to the Mehedini Plateau in Romania, and it is
surrounded by non-carbonate rocks to the East (Cretaceous
flysch and para-flysch) and to the West (Proterozoic and
Paleozoic schists). The relationships between non-carbonate and
carbonate lithologies made this karst subject to strong
influence of allogenic water input. Due to the small width of
the ridge (max. 700 m, 250 m in average), fluvial influence is
strong enough to penetrate to the opposite boundary of the
limestone. Several separate input-output systems of karst
groundwater were determined. Exposed limestone surface does not
exceed 5 km2, but 32 caves which altogether have more than 6500
m of passages were explored so far, indicating a significant
karst development. The springs are mostly permanent, although
relatively weak (up to 10 l/s), and show little seasonal
variations. Three springs have elevated temperatures from 17
to 19 C. Characteristics of karst springs are the indication
of retention capabilities of the karst aquifer as well as of a
circulation of the groundwater at great depths. Deep
circulation is mostly developed due to the favourable
conditions along the regional dislocation the D?evrin Fault.
Jemcov, I., Pavlovic, R., & Stevanovic, Z. Morphotectonic
analysis in hydrogeological research of karst terrains. A
case-study of SW Kucaj Massif, Eastern Serbia. pp. 51-59. The
fracture structures are recognized as being among the principal
factors of karst development by analyzing data obtained by
remote sensing. Regional fractures were recognized through
analysis of satellite imagery scanograms, while the detailed
fault pattern was obtained by stereoscopic analysis of aerial
photographs. The method of quantitative geomorphologic analysis
was used for the identification of neotectonically active
structures. Neotectonic analysis was performed using a
morphometric-statistical procedure the calculation of
topographic relief. As a new procedure, supplemental to fault
pattern analysis, Digital Elevation Model (DEM) was applied.
Comparison of the results obtained by fault pattern analysis
using remote sensing, quantitative geomorphologic analysis and
digital elevation model with favored directions of karst
groundwater flow, revealed a significant control of the
position and function of the faults over karstification.
General karst groundwater directions and drainage reorientation
are mainly controlled by the morphotectonic evolution. Tenu,
A., Davidescu, F., Petres R., & Coarna, L. Environmental
isotopes studies and the hydrogeological model of South
Dobrogea (Romania). pp. 61-72. Due to its natural and anthropic
features, South Dobrogea (Romania) is a very interesting area
from a hydrogeological point of view. In the region there are
two superposed calcareous aquifers the upper aquifer
(Sarmatian) and the lower aquifer (Barremian-Jurassic) which
form the so-called "karstic system" of a strategic importance.
Isotopic monitoring (3H, 14C, 13C, D, 18O) was performed over
the last 25 years by an integrated research of all natural
types of waters, in order to improve the knowledge of
hydrogeological parameters and of the regional pattern flow.
The main isotopic characteristics of meteoric and surface
waters are briefly described in the paper. Within the karstic
system, our study focused on the lower aquifer for which an
up-to-date regional model was carried out by correlating all
hydrodynamic and isotopic information. The main isotopic
features can be synthesized as follows: the recharge area is
located mainly in the Pre-Balkan Platform (Bulgaria); the main
groundwater flow direction is east-northeast, towards Lake
Siutghiol and the flow velocities, at regional scale, vary from
100 m/year for the secondary groundwater flow to 500-1,800
m/year for the main one. Taborosi, D. Biokarst on a tropical
carbonate island: Guam, Mariana Islands. pp. 73-91. "Biokarst"
refers to erosional and depositional karst features formed by
direct biologic action. However, since no distinct karst
landform is produced exclusively by biologic action, most
features known as biokarst are actually products of both biotic
and abiotic processes operating concurrently in intricate
interrelationships. Although biokarst landforms are often
regarded as features of local interest and limited
significance, the effects of living organisms on karst
geomorphology are profound on an ecosystem scale, and are
widespread, diverse and of fundamental importance. This is
especially true in tropical and coastal environments, where the
biologic influence on karst is so common and intense that it
makes the distinction between "biokarst" and "non-biokarst"
thoroughly impracticable. A survey of karst on Guam has
confirmed that nearly all of the small- and medium-scale karst
features documented on the island are affected by biota, and
bear characteristics of what has been termed biokarst. Kohler,
H., C. & Karfunkel, J. The Quaternary morphogenesis of
the Lagoa Santa tropical karst, Minas Gerais State, SE Brazil.
pp. 93-99; +1 folded map. The tropical karst of the Lagoa Santa
region, north of Belo Horizonte, in South-Eastern Brazil was
studied, in order to describe the Quaternary morphogenesis and
to elucidate the morphodynamic evolution of this area. The
karst developed over an interfluvial block, at altitudes
between 650 and 850 m above sea level. Structural alignments,
consequence of the Brasilian geotectonic cycle, were
reactivated through the Cenozoic. These alignments, together
with climate oscillations through time, control the main trends
of the karst scenario and allow the authors to postulate a
morphogenetic history from Tertiary through Holocene. Erbajeva,
M., Khenzykhenova, F., & Alexeeva, N. Small mammals of
the cave sites in the Baikalian region. pp. 101-110. This paper
deals with the small mammal assemblages from the cave sites of
the Baikalian region. In the region there are more than 300
caves and shelters (Filippov, 1993a, 1993b), some of which
contain numerous small mammal fossils. The oldest fauna is the
Middle Miocene from Aya cave containing Eurolagus,
Heterosminthus and Gobicricetodon. In other caves, more than 30
taxa of insectivores, lagomorphs and rodents were discovered,
mainly of Late Pleistocene and Holocene age. Most of the small
mammals belong to modern species, which inhabit the region
nowadays; however, specimens of Lagurus lagurus and Dicrostonyx
sp., whose area of distribution lies outside of the region
currently, were also found. Stiuca, E. & Arghir, R. -
Martes genus representatives in in the Wurmian of Romania. pp.
111-116. Remains of various mustelides (Martes martes, Martes
foina, Mustela nivalis, Putorius putorius) have been recovered
from the cave Pestera nr. 4 din Scocul Scorotei (Retezat
Mountains, Hunedoara county, Romania). This new fossil site of
the Southern Carpathians may be assigned to the Late
Pleistocene (Middle Wrmian). Allometrical and morphological
study of the skulls and dentition indicate a climate warming
that corresponds to the lower sediment layers (3 and 2),
followed by a cooling that corresponds to the upper layer, as
well as the existence, during the Wrmian of two different
types of Martes. Kranjc, A. The history of karst ressources
exploitation: an example of iron industry in Kranjska
(Slovenia). pp. 117-123. In Slovenia the karst surface covers
43% of the territory. About 57% is Dinaric and 21% Alpine
karst. Dinaric karst consists of Jurassic and Cretaceous
limestone, while Alpine karst of Triassic carbonate rocks.
Everywhere the iron ore is present. Two types can be
distinguished, the form of limonite pieces, and the "ironstone"
(iron hydroxides), so-called "broad bean ore". In the 8th
Century BC the iron industry prospered in two karst regions: in
the Alps and on Dinaric karst. Technology of extracting iron
from the ore was simple. The early metallurgists gathered iron
ore on the surface and in karst underground. The ore was melted
in shaft furnaces with dug-in fireplaces. Because of the
ironwork technology, iron industry was scattered all over the
country thus no important concentrated pollution appeared. Yet,
local impact could have been important in places. The
consumption of wood and charcoal was significant including
cutting down the forests. In the 10th Century iron industry
restarted. Between the 13th and the 15th centuries ironworks
began to develop, based on water energy. Extracting and working
of iron moved from the plateaux down to the valleys and the
ironworks (joint of smelting furnace, fireplaces to heat the
iron red-hot, bellows and ironwork hammer) resulted. In 1581,
the list of ironworks for Kranjska enumerates 20 smelting
furnaces. The direct impact on karst consists of traces of
digging on the surface, in dolines, caves and potholes, disused
mine shafts and galleries, and "ore pools" scattered across the
land. As a direct impact, large quantities of charcoal were
consumed and the water was polluted. There were much less
forests left than nowadays. The indirect impact consisted
mainly in the high concentration of industrial plants and
inhabitants. Rapid communication Vremir, M. M. & Ridush, B.
- Recent paleontological investigations in some caves of the
Crimean mountain-range (SE Ukraine). pp. 125-132. Preliminary
results of the most recent speleo-palaeontological
investigations made in the Crimean high-mountains are
presented. Several caves were investigated in the Chatyrdag
massif; also new information regarding fossil remains from
caves in several other plateaus (Aj-Petri, Yaltinskaya, Karabi)
are added. The most interesting results come from the Emine
Bair Khozar cave, which significantly contribute to highlight
the Late Pleistocene and Holocene vertebrate faunas existing in
the remote mountains of the Crimea. Book Reviews Goran, C. -
Monografia carstului din Muntii Padurea Craiului. pp.


Theoretical and Applied Karstology Volume 15


Theoretical and Applied Karstology EDITORIAL BOARD Managing Editor: Silviu Constantin, "E. Racovilii" Institute of Speleology, Bucharest, Romania. Associated Editors: Bogdan P. Onac, "E. Racovi\A" Institute of Speleology, and University, Cluj-Napoca, Romania. Adrian lurkiewlcz, UN, FAO expert, Romania. Horia Mitrofan, s. for Geodynamics, Romania. Cristlan Lascu, "E. Racovilii" Institute of Speleology, Bucharest, Romania. Editorial Assistant: Manus Vlalcu, "E. Racovilil" Institute of Speleology, Bucharest, Romania. Board of Reviewers: Michel Bakalowlcz, (Monlpellier, France) Andy Baker (Newcastle, UK) Marcian Bleahu (Bucharest, Romania) Pavel Bosak (Prague, Czech Republic) Jose-Maria Calaforra (Almeria, Spain) Gabriel Diaconu (Bucharest, Romania) Wolfgang Dreybrodt (Bremen, Germany) Derek Ford (Hamilton, Ontario, Canada) Paolo Forti (Bologna, Italy) Silvia Frisia (Trento, Italy) Dominique Genty (Saclay, France) Cristian Goran (Bucharest, Romania) Pierre-Yves Jeannin (La Chaux-de-Fonds, Switzerland) Naruhito Kashima (Ehime, Japan) Alexander Klimchouk (Kiev, Ukraine) Stein-Erik Lauritzen (Bergen, Norway) David J. Lowe (Nottingham, UK) Paul Marinos (Athens, Greece) Jacques Martini (Pretoria, South Africa) Andrej Mihevc (Postojna, Slovenia) Petar Milanovic (Belgrade, Yugoslavia) John Mylroie (Missisippi, USA) lancu Ora!}eanu (Bucharest, Romania) Arthur Palmer (Oneonta, USA) loan Povara (Bucharest, Romania) Gheorghe Racovitii (Ciuj-Napoca, Romania) Ezzat Raeisi (Shiraz, Iran) Jean-Noiil Salomon (Marseille, France) Ira Sasowski (Akron, Ohio, USA) Yavor Shopov (Sofia, Bulgaria) Augustin Tenu (Bucharest, Romania) Franco Urbani (Caracas, Venezuela) Yuan Daoxian (Guilin, China) William White (Pennsylvania, USA) Paul Williams (Auckland, New Zealand) Florian Zamfirescu (Bucharest, Romania) Indexation: Georef I EolTURA AcADEMIEI RoMANE vol. 15/2002 ISSN: 1012-9308 Theoretical and Applied Karstology (TAK) is an annual journal dedicated to the karst and caves sciences. The first four issues were published by the "Emil Racovita"lnstitute of Speleology, as a series, between 1983 and 1991. Since 1992, TAK has been edited by the Institute and published by the Romanian Academy's Publishing House. TAK publishes original contributions (review articles, research papers, short 1 notes and book reviews) covering the whole range of karstology and physical speleology: karSt geology and mineralogy, chemistry and physics of karst processes, karst geomorphology, karst hydrology and hydrogeology, speleo-chro-' nology and landscape evolution, speleogenesis, dimate and subterranean enspeleo-paleontology, engineering and environmental problems in karst, karst managment, etc. The Editors welcome the submission of contributions from all over the world. The "Emil Racovita" Speleological Institute encourages the exchange between TAK and specialized journals. Past TAK issues are also available to the Institute. Please address any inquiries concerning publication exchanges, orders, subscriptions, and any article submission to: Theoretital & Applied Karstology Editorial Board lnstitutul de Speologie "Emil Racovi\A" str. Frumoasa 11, C.P. 220-12, R-78114 Romania. tel/fax: +40-21-2128951; +40-21-2113874 email: http:// Cover photographs: Cristian Lascu Language revisions: Rodica Florescu DTP Editor: Silviu Constantin This volume was kindly supported by: Outdoor Project SRL the Romanian Society for Speleology and Karstology Journals of the Romanian Academy Publishing House may be order8d to: Edltura Academlel Romine, Calea 1a nr. 13, P.O.Box 5-42, Rodlpet sA: Piala Ubere 1, P.O.Box 35-57, Romania. fax:+40-21-2226407; tel: +40-21-2224126. Orion Preaalmpex 2000 SRL: P.O.Box n-19, Bucharest5, Romania. \. tel: ... 40-21-3018786; f111t:+40:-21-3350296 . J Galea 13 Septembrie 13, P.O. Box 5-42, 76117 Bucure!lli,


Romanian Academy Edltura Academlei RomAne "Emil RacovltJ" Institute of Speleology Bucurettl, 2002 Theoretical and Applied Karstology Volume 1512002 . .. .. .. .. . ... _,,,, . , .. ,,,,,.;. : ; 43 .. .. .-, ... ;;':.: l:;AcrAin'loana c o . :ar111A :..__.. Envifonmenta l .. isotori .studies : .... L .. .. .. .. : . 6 1 .


4 Editor's Page With its 15th issue, TAK resumes its annual appearance in a continuation of our efforts to increase the frequency of publication and, at the same time, to bring as much valu able karst science as possible on readers' desk, on a regular basis. The contents of this volume partially relies on the papers pre sented at the XVIIIth International Symposium of"Theoreti cal and Applied Karstology" which was held in Bliile Herculane, between May 24-28, this year. I'll use the privi lege of this editor's page to say a few words about this event. There were many pros and cons in our decision to re-orga nize this symposium in 2002. Serious logistical difficulties were foreseen as we could not rely on (almost) any sponsor support; to this we had to add the general lack of enthusiasm of most 'senior-organizers' ofTAK and the transformation of our traditional partner, "Prospectiuni" S.A., into a privately owned company showing now little (if any!) interest for karst resources research ... On the other hand, we have been strongly encouraged in our efforts by professor John Gunn, the presi dent of the IGU Karst Commission, who considered this sym posium very appropriate for the commission's annual meet ing. And, finally, we've got a new partner for the organiza tion, the Romanian Society for Speleology and Karstology with its young and enthusiastic staff. We said then: "OK, let's do it!" It was in January 2002 and the symposium was sched uled for the end of May, that is less then five months to orga nize an international meeting ... not an easy bet, you will agree! After only six weeks of intensive emailing and web contacts things become more encouraging: we got expressions of in terest from more than 50 colleagues from abroad, including the majority of the members of the IGU Karst Commission. Some of our worries faded away and we started to wonder if we wouldn't need a larger meeting room ... Two weeks be fore the symposium we had received over 45 papers and "finn registrations" from about 30 colleagues from abroad and more than 60 Romanians. And then, the cancellation emails started to fill my In box ... Finally, the symposium begun with a little more than 50 par ticipants, among which, 15 colleagues from Brazil, Iran, Ire land, Italy, Moldova, Russia, Slovenia, South Korea, UK, US, and Yugoslavia. Unfortunately, most of the members of the IGU Karst Commission also cancelled their participation on a very short notice, and the annual meeting was no more held. A reduced participation did not necessarily mean a decrease of interest or a lower scientific level: it just made the reunion more informal and increased a bit the time for discussions and debates. However, as an organizer, I cannot hide that see ing the number of "firmly registered" participants being re duced to almost a half during the last ten days before an event, was a very frustrating experience. Theoretical and Applied Karstology, 15 (2002) I wouldn't mention this "inside-TAK" story if I didn't have the feeling that many of those absences were not necessarily a matter of force majeure. I am afraid of a certain trend in the scientific world, which encourage the registration to a venue and the submission of abstract(s), followed by a last-minute preferably after the abstracts' volume was pub lished ... In my opinion, there are at least two factors that have led to this strategy: on the one hand, the generally high registration fees and additional expenses, which discourage students and scientists from poorer countries to even think about participation-sometimes even in spite of some grants. On the other hand, the publication of extended abstracts (which can be reported as "published papers" while keeping the "good stuff" for a "serious journal") certainly encourages this policy of "submit and disappear" ... If the above be true, then less people joining our meetings would mean even higher participation fees which will fur ther discourage people to physically participate in a confer ence. It seems to me like a vicious circle and I wonder if the scientific world doesn't need a serious discussion on this maHer. As for myself, I continue to believe that scientific de bates at a symposium, during the meetings and coffee-breaks, or late-night in hotel rooms and, especially, those during the field-trips are the life-blood of successful science and the best keys to future partnerships. I hope I'm not just a dreamer ... Therefore, I'd like to use this opportunity to thank again our colleagues that joined the last T AK meeting. Some of them also participated in the four-day-long T AK excursions (a nov elty of this edition) and I hope they enjoyed that bit of Roma nia they saw during those days at least as much as I enjoyed tieing in the field with them. A final word that'! must add to this editorial is perhaps the most d i fficultto write. The great absent of this year T A K Sym posium was dr. Costin Rlidulescu, the director of our Institute and also the former director of our journal. He really hoped to be with us in Herculane but, at that time, he was forced to stay in hospital, facing the progressive weakness of his body. Only one month after the symposium, dr. R!idulescu passed away. Beginning with that day, our institute has become poorer ... As a TAK editor during the last ten years, I will certainly miss longtime his discrete, yet very efficient, contribution and supervision. But, knowing him as a modest and sensible per son, I am quite sure that at this point of my phrase he would have gently wave his hand and said "Oh, no ... please stop ... ... we dedicate this TAK issue to the memory of dr. Costin R1idulescu. And we will continue to dedicate our efforts to all of you, colleagues and friends of the underworld. Silviu Constantin


Theoretical and Applied Karstology, 15 (2002) Constantin (Costin) 1932-2002 This year, the "Emil Racovifl" Institute of Speleology has suffered an irreplaceable loss by the unexpected decease of dr. Constantin (Costin) R.adulescu. Born on February 23, 1932 in Bucharest he completed his uni versity studies at the Faculty of Biology, Bucharest University, between 1951 and 1955. During this time his dedication was stimulated by professor Margareta Dumitrescu who encouraged and supported him to pursue the studies on Mammals paleon tology. In 19 56, while preparing his PhD he started systematic excavations in caves from Central Dobrogea such as: "La Adam", Bordeiul de Piatm, Cas ian, Cheia, Gura Dobrogei. He also partici pated in the field campaigns from some caves in the Carpathians: Coltul Surpat, the caves no. I and 2 froin Tlitaru Massif, Cioclovina, etc. In 1958 he becomes a researcher at the Emil Racovifl" Institute, where he will work until the very last moment of his life. After the submission of his PhD thesis entitled "Contributions to t he anatomic comparative study of the modern and fossil artiodactylia from Romania and their importance for the stratigraphy of the Upper Pleistoc-ene" (1963) he started extensive studies on Quaternary mammals. Gradually, the field of his research will extend also to the sediments outside the subterranean environment. Between 1960 and 1963 he started the research of the stratified depos its from the Dacian Basin (the middle course of the Oltet val ley), between the localities ofTetoiu and where re mains ofPiio-Pieistocene large mammals were abundant. The results came rapidly: two new genres were described here: Para dolicopithecus arvernensis gelicus and Mitilanotherium inespec tatum During 1962 and 1975, together with Petre-Mihai Samson and Henriette Alimen he worked in the Bnl$0V Depression where he carried out paleo-biological and paleo-environmental studies. Beginning with 1978 his activity was mostly dedicated to 5 In Memoriam the study of fossil remains from Pliocene and Lower Pleis tocene deposits from the western area of the Dacian Basin (Drartic, Izvoru); here, many associations of micro-mammals with important stratigraphic and evolutionary significances have been described. Under his guidance the first fauna associations of micro mammals from the Alb valley have been described, including five new taxa. It is worth noticing also the discov ery of a new genus of Eocene Embritopodian (Crivadia therium) with two species, as well as that of a species of an Oligocene lndricotheriidian (Benaratherium gabunim). One of the last topics of his researches was related to the anatomy and physiology of Mesozoic multituberculates and their paleoenvironmental significance. Several studies have been elaborated on this issue and a new family, two genera and two species (Barbatodon transyl vanicus and Kogaionon ungurea nul) have been described. Costin R.adulescu published over 150 scientific papers describing mamma lian fauna important both from mor phological and paleoenvironmental viewpoints, that cover a time span of over 60 Ma, from the Upper Creta ceous to the Holocene His exceptional scientific knowledge promoted him as a member of many Romanian and international scientific societies, such as: the Sub-com mission for Loess Stratigraphy and the Sub-commission for the Neo gene/Quaternary limit of INQUA, the International Commis sion for non-marine Tertiary of the lUGS, the Sub-comltlission for Environment Protection of the Romanian UNESCO Com mission. For all his merits, on May 31, 1993, the General Assembly of the Romanian Academy elected him a corresponding member as a supreme recognition of his scientific value. Behind these scientific achievements, Costin Radulescu was a kind, sensitive, warm-hearted and delicate man. If one could figure a person that knows no selfishness, hate and hard feel ings that would probably be the image that we, at the insti tute, will always treasure. Emanoi/ Sliucti


Theoretical and Applied Karstology, 15 (2002), pp. 7-26 TAK Reviews Speleology in the Third Millennium: Achievements and Challenges* Paolo Forti., Italian Institute ofSpe/eology, University of Bologna, Via Zamboni67, 40127 Bologna, Italy. Abstract Men have visited caves for a long time, but speleology started about 20 000 yr BP when the human frequentation of caves was no longer "accidental". The development of caving activities may be subdivided into three periods: prehistory (in which no written report exists), protohistory (in which documentation is available but spelcology is not yet that oftoday) and history (which conventionally started with E. A Martel). Nowadays caving activities arc split into several branches which may be grouped in four categories: explorative, scientific, social and documentary speleo/ogy. In the present paper after a short overview on the development of caving activities from prehistory until present day the challenges and goals for the third millennium arc outlined, the main pro61ems spclcology will have to face being: exploration of the ice caves in Antarctica and of the volcanic caves in the space; scientific multidisciplinary investigation of special cave ecosystems; extensive search for new drinking water supply and new principles in medicine; environmental protection nnd sustainnblc tourism. Key words: cnve history, cave science, cave tourism, cave safeguard. La speleologie au troisieme millenaire: realisations et perspectives Resume L 'homme a visile /es grottes depuis longtemps mais Ia spe/eo/ogie a commence if y a 20000 annees, quand Ia frequentation des grolles a cesse d'elre Kaccidenlel/e I.e developpemenl de /'aclivile speliologique peul eire SOIIS-divise en trois peri odes : Ia pre histoire (doni on n 'a pas de lemoignages ecrils). Ia protohistoire (if y a des sources documenlaire.r mais Ia spe/iologie n 'est pas encore Ia science d'aujourd'hui) el l'histoire {qui commence, convenlionnel/emenl. avec E A. Martel) Aujourd'hui les aclivites speleo/ogiques son/ sousdivisees dans quelques branches qui peuvenl eire groupies dans qualre cate gories : spcleologic explorative, scicntifique, socialc el documcntaire. Dans ce travail, apres un court expose sur /e deve/oppe ment des acliviles speldo depuis Ia prelristoire jusqu 'a present, on discule /es perspectives et les buts actuels pour /e troisieme millenaire. Les principa/es directions de recherche dans Ia spe/eologie sonl: /'exploration des grolles de glace de /'Antarctique el des caviles volcaniques de /'espace; investigations scienlijiques mulli-disciplinaires des ecosyslemes parlicu/iers: recherches extensives pour de nouvelles sources d'eau potable el de nouveaux remedes medicaux ; protection de /'environnemenl et lourisme Mots-ells : hisloire de Ia spe/eoiogie, speleologie scientifique, lourisme spe/eologique protection des cavernes. Introduction The word Speleology comes from the two Greek words ( am:A.alCov and A.oyoa) literally meaning the science of caves (SHAW, 1992) and therefore it should be applied only to scien tific researches performed inside natural cavities. Anyway it is actually nonnally utilized in a broad signifi cance, to define 'any activity' voluntarily perfonned by a person inside a cave (Fig. I), ranging from pure or applied science to documentation, from exploration to social. Surely human frequentation of caves started over one mil lion years BP, but: when did a man become a true "caver''? Opening Conference at the Xllllntcmalional Speleological Congress, Brasilia, July 200 I a corresponding author Email addreu : 0 2002, Editura Acadcmici RomAne. All rights reserved.


8 .. "'.. ;. P. Forti caves:, cnvitics Fig. I. Diagrammatic representation of the four main branches of speleology and of the environments in which they may be active. Representation schematique des quatre branches principales de Ia spe/eologie et des leurs domaines d'aclivite. Human frequentation of caves was purely accidental: most of the time our ancestors utilised just cave entrances for their needs (shelters, burial, etc.) without any acknowledgement of the cave environment itself: in other words, men had no interest to enter a cave just due to its cave nature and there.., fore, in reality, they did not perfonn any caving activity. Probably the first time when men started to consider caves as a peculiar place was only some tens of thousands years ago and the first reason to 'go caving' was to perfonn religious rites, as testified by scores of painted caves spread in France, Italy, Spain, etc. (Fig. 2). In the last 20 000 years caving activities grew in number and types but it is only during the last I 00-I 50 years that "modem speleology" has developed. The aim of this paper is to present a short overview on the main changes speleology underwent from its beginning up to present. The principal achievements of this rather young activity will then be used as a basis for our attempt to answer a question intriguing most of cavers: what will spe/eo/ogy become in the just begun third millennium?


Speleology in the third millennium Fig. 2. Paintings in the Cervo cave (Apulia, Italy). Peintures rupestres dans Ia grotte C,ervo (Apulia, ltalie). The development of caving activities The evolution ofSpeleology may be subdivided in three main periods as follows: Preltlstory A period during which no reports of caving activities have been produced: it started 30-20 000 yr BP and ended around 2000 yr BP. Proto-It/story The period in which caving documentation e xi sts but spe leology was not yet the activity we presently know. It is worldwide accepted that this period ended with the be ginning of the activity of E. A. Martel. History The last and rather brief period during which orga nisat i on and goals remained practically unchanged. The prehistory ofSpeleology As already said in the introduction, men have gone inside caves even deeply since at least one million years ago, to search for shelters and later for burials. But, according to its definition the prehistory of speleology started only since they begun to 'use' the cave for a specific purpose (holy pl a ces) in the Palaeolithic time (30,000-1 0,000 years BP) (SHAw, 1992) or as flint mine, as in Koonalda Caves (Australia) (WRIGHT, 1971, TANKERSLEY et a/. 1997) Later ( < 5,000 yr BP) men started exploring caves to search for specific materials (minerals) they had no chance to find outside: niter was mined in the caves near T i g ri river by Assyrians since 4000 yr BP as food stabili zer ( FORT I 1983), while at the same time the Indians of the region of Mammoth Cave (USA) went deep into the caves to search for mirabilite (purgative) and gypsum (fertilizer) (BROUGHTON, 1972). 9 The proto-history ofSpeleology The proto-history ofspeleology starts only about 3,000 years BP in the Middle East and lasts until Martel gave rise to the modem speleology Anyway even if the whole period covers something less than three millennia it is reasonable to split it into two sub periods: the boundary being during the XVJth Century when the first book comp l etely dedicated to a cave was printed (BUCHNER, 1535) ; a couple of yea rs later the first truly scientific r esear ches on caves were performed From 3,000 years BP to 1535 The very first documented speleol ogical explorat i on of a cave took place in Mesopotamia where, not far from its source, the river Tigr i s flows through a natural rock tunnel. Tiglath Pi Ieser, King of Assyria had his portrait carved at the entrance together with an inscript ion in 3100 BP (OPTiz, 1929) Later, a subs e q uent Assyri a n King, Shalmaneser, in 853 or 852 BC had his men exploring three caves nearby the stream cave. The event is also reproduced in a bronze band of the gate of his r oyal palace in Balawat, now exhibited in the Brit ish Museum Anyway the best, and rather unknown in litera ture, monument of an Assyrian king visiting a cave ( Fig. 3) is just at the entrance ofShapur Cave not far from Persepolis in Iran (FoR T I 1 993). Fig 3. The monolith 8 m t a ll at t h e e n trance o f th e S hapur C av e, Iran, rep r esent i n g th e A ssyr i an K i ng Shapur l aft e r his visit to th e c avity. Le monolith e (8 m haul) a / entree d e I a g ro tte de Sh apur. I ran, rep r e sent ant l e roi assyrie n S hapur /" apres sa visi t e dans Ia cavite.


10 About 2700 yr BP (GILL, 1991) Jews adapted natural shafts and cave passages to bring the water of the Ghion spring within the walls of Jerusalem. Chineses started visiting caves some centuries B.C. and their early exploration of caves are always related to search for health care: stalactites, stalagmites, moonmilk and fossil bones were collected inside caves and then used to prepare medi cines (SHAW, 1992). This medical use of cave material is still popular in Chinese traditional pharmacopoaeia. No detailed records of such explorations are known to survive, except for a description of stalagmites in situ made by Ko Hung about 2300 yr BP {Ko HUNG, 1946). Greek and Roman literature supply several cave descriptions but most of them are clearly fantasies and absolutely unrealistic reports Anyway it is sure that several persons, for different rea sons, visited caves, and left some detailed descriptions. Among those worth mentioning is the Greek Aristotle (384-322 BC), who described several caves with underground riv ers and wrote about the origin of caves and speleothems (SHAw, 1992) and the Latin Titus Lucretius Carus (95-51 BC), who included many interesting references to caves, ranging from hydrogeology to erosion, from earthquake to volcanic caves (on Etna Volcano) and their environment, in his poem titled" De rerum natura"(= about the nature of things) (CIGNA, 1993) P. Forti Moreover the "tassellated pavings" (Fig. 4) found in many part of the Roman Empire, sometimes reproduced the caves world rather accurately (FoRTI, 1998), providing evidence of its frequentation by the Romans. After the decay of the Roman Empire, the interest in caves became completely lost and the major part of the responsibil ity belongs to the Christian Religion which, unlike most other religions, identifies the underground as the "Devil's King dom". Therefore, for more than one thousand years, caves had in terest only for monks, witches, alchemists, robbers, etc ... During this time, the documents related to cave explorations are very few and absolutely unreliable, even if some of them are quite accurate, like that related to Cheddar Hole, England, written by Henry of HUNTINGDON in 1135: . .. Chedder-hole, where there is a cavern which many persons have entered, and have traversed a great dis tance under ground, crossing subterranean streams, without finding any end of the cavern ... The scientific interest for caves and its environment was pre served in countries with non-Christian religions, as testified by Abu Ali Ibn Sina (Avicenna) who tried to explain the de velopment of stalactites between I 021 and 1023 (A VICENNA, 1929). Fig. 4 Portion of a tessellated paving of the "Villa del Casale" (Piazza Armerina, Sicily. III Century) (left), where speleothems are repre sented very clearly: in the sketch on the right speleothems are put into evidence. Partie du pavement de Ia <

Spe/eology in the third millennitun 11 --------S ,_t_, l "' -. ... ,_ __ -Fig. 5. The famous "historical inscriptions" ofPostojna cave grouped in an engraving ofthe XIXth century (HoHENWARD, 1830). Lesfameuses inscriptions historiques de Ia grotte de Postojna groupees dans une du XIX" siec/e (HoHENWARD, 1830). Anyway in all this period most ofthe traces of speleological activities are restricted to dated inscriptions over the cave walls like the world renowned ones in Postojna (Fig. 5). After 1535 The blackout of speleology lasted all the Middle Age and part of the Renaissance. Only in the XVIth Century the inter est in caves of travellers and later of scientists started again, as testified by the first printed book fully dedicated to a cave trip (BucHNER, 1535): the single copy still existing ofthis 4-pages pamphlet on the exploration of Breitenwinner H6le in Germany is currently conserved in the State Library in Ulm. In 1590 the first description and speleogentic theory for the volcanic caves ofMt. Etna (Sicily, Italy) was reported (FlLODEI DE HoMODEIS, 1590) and eighty years later a first detailed de scription of the behavior of the air currents inside the cave was given by Johannes HERBJNJUS (1670). But the book which is generally considered as the first trea tise ofSpeleology, is "Mundus Subterranneus" by Athanas ius KIRCHER (1674). Just at the same time when Kircher wrote his famous book another important scientist was active in speleology: Johann Wei chard V ALVASOR, who travelled all over the classic karst area and systematically visited and described its caves. His book in four volumes "Die Ehre des Hertzogsthums Krain" ( 1689) has to be considered the first documentation of a true cave file and it is presently among the most rare books on speleology ever printed in the world. Some of the first scientific investigations led to the elabora tion of curious and incorrect theories, even if well accepted at that time. In the second half of the XV lith Century a scien tist put forth the idea that speleothems are true rock plants (BEAUMONT, 1676), the development of which is very similar to that of normal trees. The theory was later perfected in 1704 by J.P. TouRNEFORD who wrote: ... TJ:at certain rocks nourish themselves in the same way as plants. Perhaps they reproduce also in the same way ..... that there are seeds which gradually swell up and develop the regular structure which is perhaps hidden beneath their surface ... Thus the congelations grow up from seeds."


12 These are the reasons why in the XVIIIth Century some of the most common types ofspeieothems were often represented just as part of a tree: with stalactites as roots, stalagmites and columns as trunks, helictites as leaves or flowers (Fig. 6). The XVIIIth Century and the start of cave tourism Cave tourism started with King Tiglath Pileser in I 100 BC, and several visits to caves are variously reported since that time up to the XVIIth Century. True organized cave tourism developed only in the XVIIlth Century when some caves become well known all over the world and therefore a tourist organisation grew around them. The most renamed and worth of mention are: Postojna (Slovenia) Kungur (Russia) (Fig. 7) Antipatros (Greece) Staffa (Scotland) These caves started to attract hundreds of visitors per year and many other cavities were therefore open for tourism also outside Europe: Mammoth Cave (Kentucky) was officially opened as a show-cave in 1816 but it has been shown as a tourist attraction some tens of years before (GURNEE, 19.93). Most of these early show caves are still important nowadays, being visited by several hundreds ofthousand of visitors per year. The success of cave tourism was also testified by the fact that, at the end of the XVIIIth and at the beginning of the XIXth Century, caving books were rather popular and be came the tourist guides of the most renowned caves (LANG, 1806, HOHENWART, 1830, BULLIT, 1845). The XIXth Century and the first true scientific researches in caves In the second half of the XVIIIth Century, the fast evolution of research in many scientific fields inevitably improved the interest for caves and their environment. Fundamental b.ooks were printed in karst hydrogeology (V ALLISNERI, 1717), cave morphology (CAPPELLERI, 1767) and mineralogy (ZIMMERMANN, 1790). But it was during the XIXth Century that practically all the studies, which can be performed inside caves, started (CoNFIGLIACHI, 1819). As a logical consequence of the devel opment of several different "branches" in Speleology a first scientific treatise on cave and karst was printed by SCHMIDL (1854): all the aspects of caving, from biology to mineralogy, from cave exploration to mapping, etc are shortly described in this book. P. Forti Fig. 6. Tournefort's theory forced some artists to represent speleothems as roots (stalactites), trees (stalagmites), flowers (coralloids), grass (helictites and crystals) (Antiparos Cave, Greece). La theorie de Tournefort a inspire de nombreux artistes a representer les speleothemes com me des racines (stalactites), des arbres (stalagmites), des fleurs (coraloi'des), de l'herbe (helictites et cristaux) (Grolle d'Antiparos, Grece). Martel and the start of modern speleology It is generally accepted that "modern speleology" started at the end of the XIXth Century thanks to the activity of the French caver Eduard Alfred Martel (1859-1938) Even if other cavers, mainly from the "Classical Karst", were surely active before him (SHAw, 1993, SEz. SPEL. C1rrA DI CASTELLO, 1993, LAURETI, 2001), the reason why Martel has been correctly chosen as a symbol of the birth of"new spele ology" is due to his popularisation of this activity. In his life


Spe/eology the third millennium l.t ,1r.M ;/.lor (" 'It'" "" J',IJ,t,t l .'. ) '1'''"":,,,.,, 'nht'l0 ,,, l o'llf' ./.m.rl. ,1', f.',('lt.H'IUrJ .. < r t,rn {"" tfr 'I'IU ,/1"/,7 .. /U';, ,/, 0 '' .u,..,,w}',nvo ; ;;, ,. .. & .... IS A-Mt'N AtllardJ.. I J' .'1 m!JI' o/t' .. ,.,. ol'MVIf.r I rltoY't" .,.,.14/f..k ('t'J .K .,,.,;n, f'.l r Jtq jH.., H ./rJ , ;,,.,,., 11 ,,rrt .. : Rn"1' t'U' .r!' /'l'r.l,iuv /r krTY t J:,.,,.,r ,r.., AoYAI"I"' 1'/ ;u,"hrHII'VII/lift/ IY-' ; .: "'""" ,. ,, .... ,,. lit,JN I r,&v V G"Jrr 1 )I .. .u-.-.,;t',l lttAtttor.fvl n.r Zilll tr ,.,.; J v,. .u.,.k ,/ .f.,/r-rruut, 13 Fig. 7 Map of the Kungur gypsum cave open to tourism in the Ural Mts Russia, in a copper engraving of the middle of the XVIII Century. La carte de Ia grotte en gypse de Kungur ouverte pour le public dans les Monts Dural (Russie), dans une gravure en cuivre du XVIII' siecle. Fig 8. Title page of"Les Abimes" by E. A. Martel. Page de couverture de Les Abimes par E. A. Martel. he explored caves all around the world and printed over 700 articles and over 20 books on caving topics (CHABERT & . COURVAL, 1971 ), the most famous being" Les A bimes" ( 1894) (Fig. 8), thus making speleological activities well known worldwide Thanks to him, important journals dedicated to Speleology started to be printed, the most important of which was : "Spelunca" (I 895) and a year later "Memoires de Ia Societe de Spe/eologie" At the same time when Martel started going caving the first speleological societies were founded both on local and on national basis: the very first was the HOhlenKlub of Alppenzel in Switzerland which was founded in I 860 (BAcHLER I 905), but lapsed three years later. tllli'


14 Most of these associations were founded by cavers working in the Classical Karst area: in 1873 the "Abteilung fur Grottenforschunt' in Trieste, in 1879 Vereinfiir Hohlenkunde in Wien. But several Caving Clubs were also founded by dif ferent European nations as the Yorshire Ramblers' Club in 1892 (BELIEOUSE, 1899) and the Kyndwr Club in 1899 (BAKER, 1903) in England. The oldest still active amongst these first speleological groups is presently the "Circo/o Speleo/ogico /drologico Friu/ano" founded in Udine (Italy) in 1 898 (ANON. 1898). The complexity reached by caving activities at the end of the XIXth Century required the establishment of true caving or ganizations. Therefore between the end of the XIXth and in the first years of the XXth Century the first National Speleo logical Societies were organized in order to improve the co operation between different caving clubs of the same coun tries: the first two were the Societe de Speleologie, created by Martel, in France in 1885 and the Societa Speleologica founded in Italy in 1903 (ALZONA et a/. 1903). In the first decades ofthe XXth Century many ofthe european countries with well-developed speleological activities com pleted the national structure oftheir caving associations, which were often subdivided into three different levels: national, regional and local. Another important evolution was represented by the organi sation of regional and/or national speleological files and na tional speleological libraries in which all relevant infonna tion about the explored caves of a given area were collected and made accessible to the public at large. a. Q) Q) '0 E 0 0 \0 1\ Ul .Q) > ro 0 400 350 300 250 200 150 100 50 0 number of known caves deeper than 1000 m o 500m P Forti The increased necessity of exchanging infonnation between cavers led to the organisation of Speleological Congresses. Italy was the first country in which a National Speleological Congress was held in 1933 (CAl, 1933}, followed by France in 1938 (Sot. SPEL. DE FRANCE, 1939). Finally, the First /nter nptiona/ Congress ofSpe/eology was organised in Paris 1953 with a participation of over 200 participants coming from 27 different countries of five continents. Since then, an increasing number of speleological meetings, symposia, conferences, congresses were organised each year on a local, regional, national, and international basis: pres ently far over 200 of such reunions are scheduled yearly and their attendance ranges from a few tens to over 2000 cavers. Speleology at the End of the Second Millennium Speleology underwent a fast development in the second half of the XXth Century. Speleological activities started in many new countries and the number of caving clubs in the world rapidly exceeded the number of I 000. The explorations The explorations were pushed deeper and deeper into the earth (Fig. 9): the first -1000 has been achieved only in 1956 in side the Gouffie Berger (CouRBON et a/., 1989) but in 2000 the caves deeper than 1 km were over 65 in J 6 countries of 4 continents (ANON., 2000}. 70 -60 50 40 30 20 10 0 a. Q) Q) '0 E 8 0 ,.. 1\ Ul Q) > ro 0 Fig. 9 Diagrammatic representation of the evolution in time of the exploration of caves deeper than 500 and I 000 meters. Representation schematique de /'evolu tion en temps des explorations des grottes avec des profondeurs de plus de 500 et, rr!spectivemenl, /000 m. 1940 1950 1960 1970 1980 1990 2000


Speleology in tire third millennium Speleological materials and techniques have greatly improved in the last tens of years of the Second Millennium: the intro duction of the SRT about 40 years ago (DoBRJILA, 1973) dis closed the underground world to practically all people inter ested to it, and the last diving technology allowed for fantas tic explorations very deep into the saturated zone (PROSSER & GREY, 1992). Finally the cheap and easy travelling around the world allows the organization of hard speleological explorations in countries where caving activity was never performed before {JUDSON, 1973; BROOK & W AL1l1AM, 1978; BADINO et al., 1999). Simply, at the end of the Second Millennium no limits seem to exist for Explorative Speleology. The Scientific Research The most important speleogenetic mechanisms were under stood (KLIMCUOUK. et al., 2000), while in the last few tens of years caves proved to be the most powerful tools to perform research in a wide number of disciplines (FORTI, 2000; 2002), the most important of which are listed in Table I. 15 Since the last few tens of years, monographic volumes on most of these topics have been printed: "Morfologia Carstica" (BLEAHU, 1974), "The Science of Speleology" (FoRD & CULLJNGFORD, 1976), "Morphogenetlcs oj Karst Regions" (JAKUCS; 1977), "Karst hydrology and physical speleology" (BOGLI, 1980), "Paleokarst" (JAMES & CHOQUETTE, 1988), "Encyclopaedia Biospeleologica" (JUBERlliiE & DECU, 1994), "Cave Minerals of the Worlcl' (HilL & FoRTI 1997). Nevertheless most of the results in many of the scientific branches of speleology are scattered in hundreds of general and/or specialised journals which are rarely at easy disposal of the cavers. The reason for this situation is that important worldwide scientific speleological journals are still very rare ("International Journal ofSpeleology", "Cave and Karst Sci ence", "Theoretical and Appiied KarstologJi", "Memoires de Biospeleologie"); moreover, they have only a small circula tion rarely exceeding 1000 copies, the single exception being the "Journal of Cave and Karst Studies", the former "NSS Bulletin", with over 5000 copies. Table I Main pure and applied sciences interested in cave environment (after Fom, 2000, modified). Le.v princlpa/es sciences pure.v el app/icaliottr liees au milieu souterrain (d 'apres Fmm, 2000, modifie) . Discipline Archaeology Biology Physics Engineering Medicine Geology Geomorphology Geochemistry Geophysics Hydrogeology Mineralogy Palaeontology Sedimentology Stratigraphy Structural geology Volcanology Fields of interest remains, graffiti, rock-paintings adaptation strategies, microbiology, chcmoautothropic environments meteorology, climat

16 Therefore it is still very difficult to get update.d infonnation in several branches of scientific speleology. Since 1970, the International Union of Speleology decided to print yearly a general bibliographic bulletin .. Speleological Abstracts" ; in which all the publications related to any speleological field are reported; from 2002 this bulletin will be also available through the net. But to know ofthe existence of a paper will be of little use if the possibility to obtain copies of it cannot be guaranteed. The scarcity of available issues formany of the caving jour nals and for most of the caving books makes it impossible to have them available at least in the National Speleological Li braries. For this reason the UIS set up a network of Docu mentation Centres (Table 2), covering all the main speleo logical areas of the world: upon request, they are ready to supply photocopies from the journals and/or books present in their library. Table 2 The UIS Documentation Centres (UIS, 2000). Les centres de documentation de I'U/S (U/S, 2000) Country Argentina Austria Belgium France Germany Japan Great Britain Italy Poland Portugal Romania Slovenia Spain USA Switzerland Venezuela Institution Library "Dr Emilio Maury" Documentationszentrum des Instituted fllr Centre Documentation UBS/SSW Documentation Federation de Speleologie Bibliothek des Verbandes des Deutschen und Karstforscher Natural Science Museum British Cave Research Association Library Centro Italiano di Documentazione Speleologica "Franco Anclli" Library of"Kras i speleologia" Biblioteca Sociedadc portuguesa de espeleologi a lnstitutul de Speologie ,Emil Racovit!i" lnstitut za raziskovanje kraza Centre de Documentaci6 espeleologica National Speleological Society Library Bibliotheque centrale Societe Suisse de Speleologie Biblioteca Sociedad Venezolana de Espleleologia P. Forti The Economic and Strategic Interests in Caves In the last century, the social interest in cave environment grew dramatically both from the economic and strategic point of views. The economic importance of caves derives from several activities that can be perfonned inside them; the most impor tant of which are related to: Tourism Health care Agriculture Industry After the Second World War, tourism related to show caves and natural parks in karst areas has grown and it is still growing Address Grupo Espeleologico Argentino, Heredia 426 (C I 427NF) Buenos Aires c/o Naturhist. Museum Burgring 7-AI 014 Wien Maison de Ia Speleologie, Rue Belvaux 93 B-4030 Liege 28 rue Delandine, F-69002 Lyon c/o G. HotTman, Am Untersten Hammer 9, D-58644 lserlhon c/o Dr Ueno, Hyakunin-cho 3-23-I, Shinjuku, Tokio 160 c/o Roy Paulson, Holt House, Holt lane, Lea,Mattlock, Derbyshire DE4 5GQ lstituto Italiano di Speleologia, Via Zamboni 67, I40 127 Bologna Laboratory of Karst Environment, ul. Bedzinska 60, PL -4 I -200 Sosnowiec Rua Saraiva de Carvalho 233, P1350 Lisboa str. Frumoas!i; I I, R-78 I 14 Bucharest Knjiznica, Titov trg 2, SV 66230 Postojna Ap.C 32 I I 0, E-08080 Barcelona 6 I, Cave A venue, Huntsville, Alabama 35810 Bibliotheque de Ia Ville CH-2300 La Chaux-de-Fonds Apartado 4 7.334 Caracas I 04 II


Spe/eology in the third millennium Table 3. Evaluation of the economical importance of show caves in the world (after CJGNA eta/., 2000, modified). Evaluation de /'importance economique des grottes touristiques dans /e monde (d'apres CrGNA eta/., 2000 modifie). Total number of show caves Number of"important" show caves (over I 00,000 visitors/year) Total number of visitors/ year Money spent yearly for visiting show caves () People directly employed in show caves People, whose salary comes indirectly from show caves -800 -100 170,000,000 -I' 700,000,000 -200,000-300,000 -I 00,000,000 rapidly and currently represents a very important income in the budged of several countries. In Table 3 a rough evalua tion of the present-day main parameters in cave tourism are presented but it must be emphasised that these figures must be at least doubled if Natural Parks with karst interest will also be considered. Fig. 10. Portion of the title page of the first guide of Giusti thermal cave, Italy, with an engraving of patients standing inside the cave (TtJRCHE111, 1873). Partie de Ia page de couverture du premier guide de Ia grotte thermale de Giusti, ltalie, montrant les patients a /'interieur de Ia grotte (/'URCHh.7TI, 1873). 17 The second activity of economic .importance perfonned in caves is that related to health care: in the antiquity thermal have been used as Thermae (VERDE, ioOO), but it was from the first half of the XXth Century onwards that thermal caves started to become important from the economic point of view (Fig. I 0). In the second half of the last century the cold caves also started to be widely utilised for health care (speleotherapy) mainly in the countries of Eastern Europe (SANDRI, 1997): currently speleotherapy is nonnally used against several diseases like allergenic asthma, arthrosis, etc. (AA.VV., 1997). Finally, caves are widely used for mushrooms growing and for some special cheese production. Since the beginning of the human history caves have been, and still are, widely utilized as strategic places during wars. In the beginning they were mainly utilised as shelters for people and or strategic materials But with the development of the war technics, caves were widely used also as storage places of strategic materials like fuel and ammunition, command headquarters, harbors and airports (Fig. II). Finally, starting with the Cuban revolution {NuNEZ JIMENEZ, 1987), they become a privileged battlefield, 'as testified also by the recent war in Afganistan.


18 P. Forti Fig II. An airplane taking off from the airport built up by German aviation in the Bedeilhac cave (Pyrenees) in 1944 (BERNADAC, 1975). Avian deco/lant de /'aeroport construit par I 'aviation allemande dans fa grotte de Bedeilhac (Pyrenees) en 1944 (BHRNAIJAC 1975) The National and International Speleological Organisations Most of the National Speleological Organisations and the to tality of International ones were founded after the Second World War. Speleological Associations are presently active in over 100 countries of the world and in most of them national and/or regional structures have been established. The First Interna tional Congress ofSpeleology was held in Paris in I953, while the U.I.S. ("Union Internationale de Speleologie") was founded in Postojna in 1965 during the IVth International Congress {ANON., I 973). Two Regional Speleological Federations exists: the FEALC (Federacion Espleologica de America Latina y del Caribe) founded in I981 during the VIIIth International Speleologi cal Congress in Bowlin g Green (URBANI, 1982), and FSCE (Speleological Federation of the European Community), es tablished in 1990 during the XVIth Italian Congress of Spe leology in Udine (C!RCOLO SPEL. IDROL. FRIULANO, I 993). The UIS, which is recognized by the UNESCO as a non governmental organisation, has actually 57 country members (BosAK, 2002, personal communication) and its commissions cover most of the speleological activities around the world. The UJS leadership is currently recognized worldwide being supported by the whole speleological community, although its real power is still scarce due to the difficulties to be really operative in a professional manner. as it lacks a permanent structure. Th e Ch a ll e ng e s of th e Third Millennium The speleological challenges for the Third Millennium are many, and the most intriguing and/or important of them will be shortly outlined. Explorations Exploration shall be pushed forth to reach targets that would have been unbelievable until just a few years ago. Karst sys tems surely exceeding 2 km in depth exists in different areas of our planet, therefore the depth of2000 min a single cave will soon be reached and even surpassed: it is only a matter of time.


Speleology in the third millennium But length or depth records will not be the main challenge for caving explorers. There will be other boundaries to be sur passed in the future. The systematic exploration of the frightening but fascinating world of the ice caves of Antarctica (BADJNO & MENEGHEL, 200 I) will become reality in the next few tens of years. Also the exploration targets will change dramatically in the next few centuries. lt is currently well understood that our planet is not the only one in which caves developed: volcanic caves have already been detected on several planets of our solar system (GREELEY, 1977; 1991; LICITRA, 1999). These lava tubes (Fig. 12) may sometimes reach incredible lengths ex ceeding by far the longest caves on Earth. All these caves inside and outside the solar system are waiting to be explored and mapped by spaceman cavers .... And this is not science fiction ... It is quite sure that this explorations will happen in the near future not because ofthe scientific importance of such caves but due to the practical one: in fact they may be regarded as ideal places for the installation of the first settlements for space colonization. Pure and Applied Scientific Research The importance of caves in pure and applied scientific re search will largely improve in the Third Millennium. The most important fields in pure research where it is reason able to expect a noticeable increase of interest are: high-resolution paleoenvironmental reconstruction; microbiology low-enthalpy reactions special ecosystems Caves are amongst the most durable geomorphic features and represent perfect sedimentological traps for physical and chemical deposits that may be kept untouched therein even over a very long span of time. In the last decade caves, and especially the hosted speleothems, yielded the best and pow erful tools to reconstruct paleo-environments and paleo-cli mates of the Late Quaternary, sometimes allowing a resolu tien of up to one year or lower (FoRD, 1997; SHoPOv, 1997, ANTONIOU et a/. 200 l ). Moreover, speleothems proved to be extremely useful as natural recorders of strong earthquakes of the past, thus allowing a better definition of the seismic hazard (FORTI, 1999; QuJNJF, 1998). It is therefore reasonable to forecast for the Third Millen nium a very fast increase of all those paleo-environmental and paleo-seismic analyses. In the last few years, microbiology proved to be fundamental in caves: plenty of bacteria and other micro-organisms pas sively and/or actively interfere with the cave evolution, like those involved in the sulphur cycle (Fig. 13) (FoRTI, 2001). 19 Fig. 12. Moon: collapses along a lava tube close to the border of the Oceanus Procellarum. Effondrements le long d'un tube de lave sur Ia lune proche de Ia bordure de /'Oceanus Procellarum. The complex biochemical reactions involved in the develop ment of different cave deposits, though still not completely understood, clearly bear an interest and importance far ex ceeding the simple speleogenetic one (CoNTOs, 2001; NoRTHUP el a/. 1997; SASOWSKY & PALMER, 1994). Normally they are low-enthalpy reactions and unraveling them is fundamental in order to improve our understanding of the natural mechanisms by which even ore bodies of economic interest may be formed and mobilized (FoRTI, 1989; l996a). The study of biologically driven reactions is also fundamen tal to enhance our knowledge oyer peculiar environments like the chemoautotrophic ones, which characterise not only caves (SARsu et a/. 1996) but also some deep sea environments and it is obvious that such studies may be performed much more easily in caves (FORTI et a/., 2002). Caves' microbiology is presently not known despite its great scientific interest. Thus it is reasonable to expect an increase in the co-operation between cave biologists and cave geochemists in the near future in order to obtain a fast im provement the study of these phenomena. The results of the applied research in caves will be even more important: in fact some of the most important challenges of


20 Hydrogen sulfide Hz$ (Desulfovlbrlo & Deaulfotomaculum) bacterial reduction (Dissimilatory sulfate reduction) Bacterial oxidation (Thlobaclllua & Sulfolobus) Organic sulfur compounds the whole humanity will probably be solved within karst en vironments. Among them two are worthwhile to be quoted: extensive search for new drinking water resources; new active principles in medicine. The most relevant problem for future generations will surely be the drinking water supply. The pollution and the increase in demand are rapidly depleting all currently known sources of water. Karst aquifers started to be heavily utilized as a civil water supply in the last decades of the Second Millennium, but it is highly expected that in the near future the water hosted in karst areas will become if not the single, surely the most important among the relatively low-cost sources of drinking water (Fig. 14). Caves should become extremely important sites for medical research. In fact even if it is still questionable the fact that caves could have been the "nursery" for new terrible diseases like Ebola (Fig. 15) and/or AIDS (HALLIDAY, 1999), it has been definitely demonstrated that caves and their peculiar eco systems may host specific and, in some cases, new viruses, bacteria and/or other micro-organisms. The research started on this specific topic in the last couple of years in the United States (BIGELOW, 1998) and it is just at the very beginning. The first achieved results suggest that they wiii be very fruitful: several hundreds of new micro-organisms and viruses have been observed in caves and some tens have been selected as potential "active vehicles" and they wilt be tested in the next tens of years. Sulfate so: other 20% P Forti Fig. 13. The sulphur cycle (after FoRTI, 200 I). Le cycle du soufre (d'apres FORTI, 2001). 1975 2025 80% Fig. 14. Variation in civil water supply between 1975 and 2025 (FAO unpublished report). Tendance d'evolution des sources d'eau potable entre 1975 et 2025 (d'apres FAO, rapport non-publie).


Speleology in the third millennium Fig. 15. The author visiting Kitum cave (Mt. Elgon, Kenya), supposed to be the lair of Ebola virus. L 'auteur, visitant/a grotte de Kitum (Mt. Elgon, Kenya) supposee d'etre Ia taniere du virus d'Ebola. But the most important problem that pure and applied scien tific research will have to resolve even before improving its presence inside caves is the preservation of cave environment. No real attention has been focused on the degradation of cave environment induced by scientific research: but often rese archers do destructive sampling far over the real necessity of their studies. This happened because no real control over the be havior of scientists in a cave environment has been exercised until present. It is time to begin to enforce such a control... Therefore the Third Millennium shall l ead to a systematic decrease of destructive sampling in each of the speleological scientific fields. Environmental Protection Like the scientific ones, all other (social, economic, etc.) ac tivities inside caves will experience a fast increase during the 21 Third Millennium. Among them, tourism and other related activities shall expand greatly reaching new countries and ar eas, thus involving perhaps billions of peoples. An obvious consequence of the increased human activities in cave and karst environment is the risk of a fast degradation of the related ecosystems. Therefore the Third Millennium must be mainly dedicated to a strict control over different caving activities to ensure their sustainable development in time and the whole speleological community must cooperate in particular by means of: developing active protection of karst areas and caves; improving eco-compatible tourism; self limitation of some activities inside caves; strict control of the scientific activities therein. The simplest and the easiest way to ensure protection from environmental risks is surely the creation of cave and/or karst parks and reserves. But ruling the territory is normally a matter of Local and/or National Governments and therefore at local and national level the speleological associations must improve their voice at the politicallevel. In the same time, at the international level the UIS must co operate with UNESCO in order to improve the number of karst and cave sites inserted among the World Heritages. Anyway it is absolutely unrealistic to think of protecting more than 1-2% of the total karst patrimony of our planet in this way. Development of Environmental Sustainable Tourism Despite the opinion of most cavers and speleological asso ciations it must be stressed that tourism does not automati cally mean damaging of the cave environment. On the con trary, sometimes tourism may be the unique chance to preserve some very fragile and precious environ ment (FORTI, 1996b ). The present-day knowledge about cave environment, microdimate and ecosystem is complete enough to allow: the transformation of virtually any cave into a show cave, maintaining unaltered its natural value; the choice of the best materials and ways to do it; the control of the tourists presence in such a way that none of the natural parameters of the cave will change in a permanent manner.


22 buildings prklng road system receptivity spec:IHc goologlcel lnveaUgatlona lighting conMrvetloll P. Forti t.....,.ture rel"""t points ...,. ....... profit Fig 16. Environmental impact assessment for the development and management of a show cave (after CIONA eta/., 2000, modified). Evaluation de /'i mpact du deve/oppem ent et administration d'une grotte touristique sur le milieu (d'apres CIGNA eta/., 2000, modifle)


Speleology in the third millennium This is only a matter of controls before, during and after the tourist development (CIGNA eta/., 2000; CIGNA & BURRI, 200 I): in other words, a matter of correct management of a show cave (Fig. 16). In the Third Millennium all caves opened to tourism must be monitored and must have a Technical Commission with the right to change the tourist flows on the basis of the achieved data; cave managers should understand that monitoring is not a simple cost, but instead a very productive investment. The UIS and ISCA (International Show Cave Association) co-operate in order to transform this statement into a common feeling in the near future. Anyway, from now on, no cave without a monitoring network and a Technical Commis sion be supported by any cover wherever in the world Sustainable Criteria for Cave Exploration Caves are very fragile environments and perhaps the single ones that are still partially unaltered in our world, consequently speleological explorations affect them greatly. Until now most of the ecological concern was directed to ward show caves, but in reality cave tourism affects far less than 0.1% of the whole subterrannean patrimony, while all other economic and/or strategic activities affect no more than another 0.1 %. It is obvious that speleological explorations affect a relevant fraction of I 00%. The first exploration always causes a loss of naturality far more than any subsequent activities and it may damage the cave environment far more than a well organised tourist transformation. Further damages are brought about by any subsequent caving trips (BADINO, 2000; 200 I). It is the time for cavers to become aware of it and try to mini mize their impact on the underground world by ruling self limitation criteria in cave explorations. As already verified in many of the most famous caving areas of our planet: A limit .for the frequentation of any cave St exist. The guideline already available for show caves may be a reasonable starting point for the definition of the ac cess criteria to a wild cave. A limit must exist for cave modification during the exploration. It is really unacceptable to have no limits (as currently) to enlarge entrances and/or passages, to destroy speleo thems and physical infilling, to empty sumps, without any due care to cave microclimate, sediments, etc ... 23 The impact on caves by improving the techniques and materials must be kept to a minimum. Most of the cave damages during exploration or caving trips arise from the use of wrong technics and/or materials: all cavers must always be updated on them and those highly impacting discharged. It is the time for the speleological community to discuss these topics on any level and to rule out consequently. Final Remarks The fast development of speleology in the last few decades clearly suggests that caving activity in the Third Millennium will surely experience an increase in any one of its fields. Extreme explorations will be performed not only on our planet; cave and karst tourism will rapidly expand reaching new coun tries on the five continents: several hundreds, if not thou sands, of new show caves will be implemented to accomplish the request of billions oftourists. Cave environments will become more and more important for exciting research in plenty of scientific fields, while the deep karsts will show their importance in terms of renewable resources (first of all for the drinking water supply) greatly enhanced. Anyway, as a direct consequence of its development, spele ology will also have to face several hard risks for caves and karst areas: first of all the possible unacceptable environmen tal alteration as a direct consequence of an excessive anthropic pressure over their fragile environment. A definite limit for frequentation has to be fixed not only for show caves, but also for any of the natural cavities. Moreover the most impacting exploration activities and tools must be banned from caves: as in the case of carbide lighting, which may cause, and in reality causes, relevant damages to the cav ern environment (burning and/or disfiguring speleothems, polluting the water, supplying heat to the environment) and may now be easily replaced by electric LED (BRUMMEL, 1999; CIIAII.LOIJX, 1999). Finally the unjustified scientific oversampling, a risk that has been underestimated mitil now if not completely neglected, must be strictly avoided. This problem already exists but the increase in Scientific research in the near future will surely bring inside caves researchers who will not be completely aware of the problem or, even worse, be indifferent to cave protection. To ensure the preservation of caves and karst for the future gen erations it is necessary that the speleological community will cause a voluntarily limit to the activities to be per formed in caves and this should be the main role of the Interna tional Union of Speleology in the Third Millennium.


24 Something in this direction has already been done but much more still has to come: the UIS has the knowledge, the possi bility, the authority and the right not only to control but also to rule these processes and all the national speleological as sociation must co-operate with the UIS for the preservation of the cavern environment worldwide. People and ideas are fundamental to reach this goal, but not enough: in fact the UIS needs much more money and t

Speleology i11 the third millenniwn CIRCOLO SrDI.I!OLOOICO IDROLOOICO FRIULANO (1993) Atti del XVI Congrcsso Nnzionalc di Spclcologia. Le Grotte d'Jtalia, S.4, 15, p.13. CoNFIOLIACHI, P. ( 1819) Monografla del Prot eo Angulno di Laurenti. Pavia, 120 p. Comos, A. (200 I) Biominerali,falion in Caves. PhD Thesis, Univer sity of Sidney, 221 p. CouRBON, P., CHADERT, C., BosreD, P., & LINDSLEY, K. (1989) Atlas of the great caves of the world. Cave Books, St. Louis, 370 p. DooRILLA, J. C., MARBACH, G, & PEIONE, B., ( 1973) Techniques de Ia Speteologie Alpine. Paris, 100 p. FILODEI DE HoMC>DI!IS, A. (I Aetnae Topographia. Perugia, 18 p. FoRD, D. C. ( 1997) Dating and Palco-Environmental Studies of Spelcotbcms. In: Cave Minerals of the world. (HILL, C. A. & FoRTI, P. Eds.) NSS, Huntsville, pp. 271-284. FoRD, T. D. & CuLLINOFORD, C. H. D. (1976) The Science ofSpele ology. London, 594 p. FoRTI, P. (1983) I mincrali di grotta. Speleo, 9, pp. 9-24. Fom, P. (1989) 1l1c role of sulfide-sulfate reactions in spclcogcncsis. Xth Int. Spel Congr., Budapest, l, pp. 71-73. FoRTI, P. (1993) Breve nota a margine di un vclocc viaggio in Iran (22-30 Ottobrc). Sottoterra, 95, pp. 21-23. Fom, P. ( 1996a) Thermal Karst Systems. Acta Car.mlogica, XXV, pp. 99-117. FoRTI, P. (1996b) Turisticizzazionc c tutela dcll'ambicntc ipogco: due aspctti non contrastanti. Alii Int. Symp. "Grotte turistiche e monitoraggio ambient ale", Fabrosa Soprano 1995, pp. 49-56. FoRTI, P ( 1998) La storia delle concrczioni si allunga di 400 anni. Speleologia, 38, p. 67. Fom, P. ( 1999) Evidcncias tect6nicas y slsmicas a partir del estudio de espclcotemas: conocimiento actual y desarrollo futuro. Inti. Symp. Cueva de Nerja, pp. 19-33. Fom, P. (2000) La speleologia: un strumento di indaginc scientifica Geologia deii'Ambiente, 3, pp. 2-9. Fom, P. (2001) Biogenic Speleothems: an overview. Abstract of Papers, 15"' Int. Symp of Biospeleology. /nJervales, Brazil, pp. 3-6. FoRTI, P. (2002) Motivi di interesse geologico delle grotte. Convegno Massa-Carrara, Settembre 200 I, i11 press. Fom, P., GAJ.DENZI, S., & SARDU, S. (2002) The hypogcnic caves: a powerful tool for the study of seeps and their environmental effects. ConlinenJal Shelf Research. (in press). GILL, D. ( 1991) Subterranean waterworks of biblical Jemsalcm: ad aptation of a karst system. Science, 2545037, pp.l467-1471. GREELEY, R. (1977) Lava tubes on other planets. Alii Scminario "Grotte Laviche", Catania 1975, pp.181-191. GREELEY, R. (1991) Lava tubes in the Solar System. Proc. 6"' Int. Symp. On Vulcanospeleology. Hilo, Hawaii, pp. 223-230. GuRNEE, R. ( 1993) A briefhistory of cave studies in the United States before 1887 ( 16'h to the 19'k century). Proc. Int. Symp. "Protostoria della Speleologia", Cillo di Castello 13-15 Settembre 1991, pp. 293-308. 15 HALLIDAY, W. R. (1999) Overview of the 8Intemational Sympo sium on Vulcanospcleology. Inti. J. Speleol., 278, 114, pp. 9-10. HeRDINIUS, J. ( 1670) Dissertationes de admirandis mundi cataractis supra et subterraneis. Copenhagen, 267 p. HILL, C. A. & Fom, P. ( 1997) Cave Minerals of the World" NSS, Huntsville, 464 p. HoHENWART, F. G, VON (1830) Adelsberger und Kronprinz Ferdi nands-Grotte. Wien, 16+10+14 p. HUNTIONTON, H., OF ( 1853) The chronicle ... (FoRRilSTI!R, T. Trans. & Ed.) London. KIRCHER, A. ( 1674) Mundus Subterrane11s. Amsterdam, 372 +S 16 p. KUMCHOUK, A. B., FORD, D. c .. PALMER, A. N., & I>REYBRODT, w. (Eds) (2000) Speleogenesis, evolution of karst aquifors. NSS, Huntsville, 528 p. Ko HUNG (1946) Pao p'u Tzu (Book ofthc preservation-of-solidarity master). (FI!IFI!L, E. Trans.) Mon11menJa Serica, II. JAKUCS, L. (1977) Morphogenetics of Karst Regions. Bristol, 284 p. JAMES, N. P. & CnOQUETTt:, P. W. (Eds.) (1988) Paleokarst. Springer, 416p. JunER'IlflE, C. & Docu, V. (Eds.) ( 1994) Encyclopaedia biospeologica. I, Moulis-Bucarest, 834 p. JUDSON, D. (1973) Ghar Paraz1. London, 216 p. LANG (1806) Gal/erie der Enterirdischen Schopfimgs-Wunderund des menschlichen kunstsleisses 1mter der erde. Leipzig, 416 p. LAURETI, L. (200 I) Storia della Spclcologia SSI Quadrni didattici, II, 40 p. LICITRA, G (1999) Grotte vulcaniche nel mondo .... ed oltre. Centro Etneo "Dentro il Vulcano", pp. 47-56. MARTEL, E. A. (1894) Les Abimes. Paris, Dclagravc, 578 p. NoRTHUP, D. E., REYSENBACH, A. L., & PAce, N. R. (1997) Microor ganisms and speleothems. In: Cave Minerals of the World (H1u., C. A. & Fom P., Eds.), NSS, Huntsville, pp. 261-266. NuNEZ JIMENEZ, A. ( 1987) Geografla y Espeleologia en Revolucion. La Habana, 190 p. Omz, D. (1929) AssyrerkOnigc als HOhlcnforscher. Mitt. Hohl.-11. Karstforsch., 2, pp. 58-61. PROSSER, J. & GREY, H. V. (1992) NSS Cave Diving Manual. NSS, Huntsville, 378 p. QUINIF, Y. (Ed.) ( 1998) Karst & Tectonics. Speleochronos, hors scrie, 199p. SANDRI, B. ( 1997) Development of spelcotherapy in Europe since the constitution ofspelcotherapy commission within the Union lntcmationale de Spclcologic. P1oc. Inti. Conf. "Protection and medical utilisatio11 of karst environment", Hanska Bystrica, pp. 62-64. SARBU, S. Nf .. KANE, T. C., & KINKI.E, B. K. (1996) A chemoau totrophically based groundwater ecosystem. Science, 272, pp. 1953-1955.


26 SAsowsKv, I. D. & PALMER, M. V. (1994) BreakJhroughs in Karst Geomicrobiology and Redox Geochemistry. Symposium stract, Karst Waters Institute, 112 p. SCHMID!., A. (1854) Die Grotte unHohlenvonAdelsber, Lueg, Planina and Laos. SEZIONF. SPELEOLOOICA ,CmA 01 CASTELLO" ( 1993) Protostoria della Speleologia. Cilia di Castello, 352 p. SHAW, T. D. (1992) History ofCave Science. Sydney Spel. Soc., 338 p. SHoi'Ov, Y. Y. (1997) LumincscenceofCave Minerals. In: Cave Min erals oftlte world. (HILL C. A. & Fom, P., Eds), NSS, Huntswille, pp. 244-248. Soc 1m SPELEOLOOIQUE oF. FRANCE ( 1939) Actes du I er Congres Na tional de Speleologie. Nimes, 76 p. TANKERSLEY, K. B., MUNSON, c. A., WATSON, P. J., SHAFTER, N. R., & FRUSHOUR, S. S. ( 1997) Archaeology and Speleothems. In: Cave Minerals oft he World (HILL C. A. & Fom, P., Eds), NSS, Hunts ville, pp. 266-270. P. Forti ToURNI!I'ORD, J .P. (1704) Description du labyrinthe du Candie,avec quelques observations sur l'accroissement et sur Ia generation des pierres. Mem. R. Acad. Paris, pp. 406-424. TURCHIITrl, 0. ( 1873) Guida dei bagni a vapore naturale della Grotto di Monsummano. Loescher, Firenze, 94 p. U.I.S. (2000) BBS-Speleological Abstract. Varese, p. 2. URRANI, F. P. ( 1982) Speleological Association of Central, South America and Caribbean. UIS Bulletin, 112, p.l3. V ALUSNERI, A. ( 1726) Lezione accademica intorno a I 'nriglne delle fontane. Venice, 408 p. V ALVASOR, J. W. ( 1689) Die Ehre des Hertzogthums Krain Lanbach., 696+838+732+676 p. VERDI!, G (2000) II termalesimo di Sciacca dalla preistoria a[ XX secolo. Sarcuto, Agrigcnto, 280 p. WRIGIIT, R. V. ( 1971) The Archaeology oft he Gallus Site, Koonalda Cave. Aust. Inst. Aboriginal Stud., Canberra, 133 p. ZIMMERMAN, M. (1790) Voyage a Ia nitriere naturelle qui se trouve a Molfetta. Venezia, 44 p. received: 27 February 2002 revised: 26 July 2002 accepted: 30 July 2002


YAK Articles Abst r act Theoretical and Applied Karstology, I S (2002), pp. 27-34 Unusual minerals related to phosphate deposits in Cioclovina Cave, Mts. (Romania) Bogdan P. Onac 1 h, Radu Breban 3 Joe Kearns 4 & Tudor Timat 1 2 w 1 University of Cluj, Dept. of Mineralogy, Cluj-Napoca, Romania. 1 "Emil RacoviJll" Institute ofSpeleology, Cluj Branch, Clinicilor 3400 Cluj-Napoca, Romania. J Speleoclub "Proteus", C.P. 633, 2750 Hunedoara, Romania. 4 Materials Research Laboratory, The Pennsylvania State University, University Park PA, 16802, USA. Cioclovina Cave hosted an extensive phosphate deposit estimated at over 50,000 ml. About 30,000 ml were mined the first half of the XXth Century and used as fertilizers. The mineralogy of this deposit is remarkable as it consists of several rare and/or unusual cave mineral species, mostly phosphates. Cioclovina Cave is the type locality for ardealite. Over 40 samples w ere analyzed by means of X-ray diffraction and fluorescence, scanning electron microscopy, optical observations on thin sections, and electron microprobe. Out of 26 minerals pre sented i n this paper 13 have n o t been previously documented from this ca v e (berlinite, burbankite, churchitc, chlorellestadite, foggite, pnratacamite, collinsite, fluorapatite, sampleitc, romanechitc, leucophosphitc and todorokite). Furthermore, the first six minerals on this list were for the first time identified within the cave environment. At least one rare mineral species (berlinite or chlorellestadite) may have been produced by spontaneous combustion of bat guano, whereas the other ones formed within the sediment fill indicate reactions between phosphate-rich solutions and limestone bedrock, clays, sandstones or various trace clements. These reactions took place at different pH values producing specific minerals that may have environmental deposition significance. Keywords: cave minerals, phosphates, guano combustion, Cioclovina Cave, Romania. Quelques mlneraux partlculiers formes en relation avec les dep6ts de phosphates de Ia Grone de C/oclovina (Monts $ureanu, Roumanie) Resume La grotte de Cioclovina abrite un depot mas., if de phosphates, doni /e volume .:1 "ere e.vtime a plus de 50.000 mJ. Approximativement30. 000 m, ont ete exploites dans Ia premiere moitie du XX siecle et utilises dans /'agriculture. La minera/ogie de ce depot est remarquab/e dufait qu'/1 contient quelques especes minera/e.v rares etlo"U particulieres, Ia p/upart des phosphates. La grotte de Cioc/ovina est/a localite-type de /'ardealite. Plus de 40 echantillons ont ete analyses par diffraction de RX etjluorescence, au microscope e/ectronique a ba/ayage, par des observalion.t optiques et microprobe e/ectronique. Des 26 mineraux presente v dans le travail, I 3 son/ n o uveaux pour celle grolle (/e berlin it e. /e burbankite, /e churchite /e ch/orel/estadite, /efoggite, /e paratacamite le collinsite, le monetite, lefluorapatile, /e samplefte, /e romancchite, /e leucophosphile et/e todorokite). En outre, les six pre miers mineraux onl ete identifies pour Ia premierefoi., dans /e milieu souterrain. Au moins une des especes minera/es rares (le berlinile ou le chlorellestadlte) pourrait etreformee par Ia combustion spontanee du guano, landis que /cs autres mineraux presents dans /e sediment suggerenl des reactions entre /es solutions riches en phosphate.,, d'une part, et/e calcaire des parois /'ar gile, les gres o u divers elements rares, d'autre part. Ces reactions ont eu lieu en diverses conditions de pH et ont produ it des mineraux specifiques, a signi}icalion environnementa/e. MotS-clis : mlnP.raux de grotte, phosphates, combustion du guano, Grotte de Cioclo!,ina, Roumanie. corresponding author. E mail bonnc@biogc.ubbcluj ro C 2002, Editura Acadcmici Rom4ne. All rights reserved.


28 Introduction The phosphate deposit located within the Cioclovina Uscatli Cave1 (=Dry Cioclovina, hereafter called Cioclovina for sim plicity) contains a fascinating assemblage of minerals that in cludes, apart from many rare pho.sphates, several carbonates, silicates, sulfates and hydroxide species. Most of the minerals identified within the phosphate deposit form nodules, crusts, bands of earthy-masses interbedded with sand, gravels, clays and concentric layers around weathered limestone blocks, vol canic, and metamorphic chunks. Only a few of the minerals form euhedral or subhedral crystals; all the others appear as earthy-masses. Extensive investigations on the mineralogy ofthis cave occurred at the beginning of the XXth Century when over 30,000 m3 of phosphate sediments were mined out and used as fertilizers. Cioclovina Cave is the type locality for ordealite, .XHP04 ) 4Hp, originally reported by HAlLA (1931) and fully described by SGIADLER (1932). Not until 1997 was the mineralogical in terest in Cioclovina Cave rekindled when one of the authors (RB) began his graduate thesis. A few years later, CoNSTANTINESCU eta/. ( 1999) pub I ished a note on the presence of crandallite, incorporating some new data on the phosphate association. A recent extended ab stract and a forthcoming paper include new mineralogical data on the bat guano deposit from Cioclovina and first mention the presence of tinsleyite KAI1(P04MOH)Hp and carbon ate-hydroxylapatite, Ca5(P04,C03MOH), (DUMITRAS & MARINCEA, 2000; MARINCEA eta/., 2002). The aim of this preliminary report is to extend the previous mineralogical work carried out on Cioclovina Cave phosphate deposit, moreover, to present additional data and some comments on the most interesting mineral species. It must be stressed that the complex mineralogy of this cavern cannot be entirely presented in a single research paper. Geological and speleological setting. Sample collection The study area is situated in the upper part of the Luncanilor Valley, on the west-southwest side of the Mountains (Fig. I, inset). The basic stratigraphy around Cioclovina Cave consists of a thick carbonate sequence of Upper Jurassic (Stromberg-type facies) and Lower Cretaceous ( Urgonian fa cies) ages. Underlying the carbonates are the gneisses : of the Unit (Getic Nappe), and Permian to Lower Ju rassic (Liasic) detrital deposits (BALINTONI & BucuR, 200 I ; SntLA, 1981) (Fig. I) Cioclovina Cave (code 2063/8, GoRAN, 1982) develops in the Malm-Neocomian limestones and consists of a long fossil pas sage interrupted in the middle by a descending gallery and a 1 Also known as Ciclovina Cave. B. P. Onac el al. 10 m deep pit that connects the upper level with the stream passage. Climbing a 30 m-Iong chimney, one can continue in the inactive passage. This level continues as well decorated passages rooms (Fig. 2). The cave has been known since the late XIXth Century when scientists visited it to search for cave bear fossils and to investigate the extensive phosphate deposit. Much of the 15 to 20m-thick phosphate deposit was mined out of the cave between 1912 and 1941. Nevertheless, in many parts of the cave one can still notice layers of phos phate sediments (5 to 6 m-thick) covered by flowstone in their upper part (Fig 3). Although the cave has a natural entrance, it is seldom used. A mining gallery dug during the course of guano-phosphate ex ploitation is the preferred The relative humidity is between 75 and 95% throughout the sampled part ofthe cave, while the temperature remains con stant year-round in the range of 8-9 C. Variations, however, of both parameters occur in the sector where the man-made tunnel penetrates the cave Over 40 samples were collected from various settings along the main passage, between the natural entrance and the Bivouac Room (Fig 2). These samples consist of crusts, nodules, aggregates of tiny crystals, and earthy-masses. In many of the occurrences the phosphate deposit actually represents phos phatized clay and residual or fluviatile silica-rich sediments. At other locations phosphate minerals precipitate at the locus of reaction with limestone, resulting in the metasomatic re placements taking the form of banded coatings. Except for the green-bluish nodules admixed with the pale-blue tiny crystals (sample #1323), all the samples were hand-specimen size pieces. To avoid hydration/dehydration or any other post sampling process the samples were kept in airand water-tight plastic bags. Analytical methods The mitt'eralogy of phosphate deposits was studied using a combination of optical microscopy, X-ray diffraction (XRD), X-ray fluorescence (XRF), scanning electron microscopy (SEM) equipped with energy dispersive spectrometer (EDS), and electron microprobe analysis. Polished thin sections were prepared from all hardrock samples, for studies using a NIKON Optiphot 2-POL petrographic mi croscope. The remaining material and the soft samples were crushed and ground. Powder splits were taken for the rest of the analyses. XRD data were obtained using Scintag Pad V and Philips PW1800 diffractometers, both operating at 45 kV and 40 rnA using Ni-filtered Cu Ka radiation. Silicon was added as an internal standard. The step-scan data were continuously col lected over the range of 5 to 85 29, using a step interval of 0.025 29.


Unusual minerals in Cioclovina Cave ClUj ClociOYina c .... ... A N 29 Fig I. Location of the study area and its geology. I schists and gneisSes (Precambrian); 2. Cioclovina Red Beds (P1); 3. conglomerates and sandstones {11); 4. limy sandstones (12); 5. cherty limestones (J,); 6. reef limestones (K1); 7 Quaternary loose sediments; 8. sandstones & marls (K2)( after S11LLA, 1981 ) Situation de Ia zone etudiee et carte geologique. / schistes et gneiss (Precambrien) ; 2 Couches rouges de Cioclov ina (P1); 3. conglomerats et gres (J1 ) 4 gres calcaires (J1); 5. calcaires a cherts (JJ); 6 calcaires recifaux (K1 ) ; 7. sediments Quaierna i res ; 8 gres et marnes (K1 ) "(d' apres SnuJ, /98/). ,f_ -_j .60 3-Fig. 2. Cross section through the C i oclovina Cave showing the toea-7 tion of sampling points. .. 8 D Coupe transversale de Ia grotte de Natural entrance 'f sampling points The bulk chemistry of most of the samples was detennined by standard fused-button methods using a Spectro Analytical X Lab 2000 XRF instrument. The elemental composition of se lected samples was obtained using a JEOL JXA-6800 Superprobe operating at 20 kV accelerating voltage and 5 nA beam current. For SEM observations and EDS spectra analysis, freshly frac tured rock fragments, powder s plits and crystal aggregates were gold-coated and inspected with a Hitachi 3500N electron mi croscope. 0 300m C i oclovina avec Ia localisation des points d e c hantillonnage C C. 5 Proteus Hunedoara 1999 Results and discussions Various authors (ScHADLER, 1929 1932; HAlLA, 1931 ; CoNSTANTINEScu eta/., 1999 ; & MARINCEA, 2000) have reported the presence of phosphate minerals such as ardealite brushite, taranakite, crandallite, tinsleyi te, and other non-phosphate cave minerals (e.g., calcite, gypsum, aragonite) in Cioclovina Cave. Therefore, only the more unusual or new cave-type minerals will be discussed here. A more detailed presentation of the fonnation of phosphate mineral assemblage in Cioclovina Cave, along with cave


30 Fig. 3 Photograph showing the phosphate sediment infillings. Photographie du depot de remplissage phosphatique. environmental significance ofthis deposit will be the topic of forthcoming papers. The cave minerals identified in Ciclovina Cave are listed in Table I and were assigned to six chemical classes. Although the title of this paper suggests only those minerals related to the phosphate deposits are examined, we also included a short presentation of the rare carbonates and two manganese ox ides. Carbonates Ciclovina Cave contains a variety of speleothem types and subtypes (see HILL & FoRTI, 1997 for details) composed of calcite: soda straw, stalactites, stalagmites, columns, rimstone dams, flowstone, calcite rafts and corraloids. The dry condi tions in some parts of the cave allow fast C02 evaporation allowing aragonite to precipitate from supersaturated solutions. Burbankite, (Ca,NaMSr,Ba,CeMCOJ5 is a rare anhydrous carbonate that was identified by means of XRD and energy dispersive spectrometry (EDS) in a lacustrine-like sediment sequence near the Bivouac Room. It appears as a thin crust composed of sub-millimeter yellow grayish anhedral crystals. B. P 01Ulc et al. Burbankite was found in association with colorless or milky white needle-like brushite and gypsum crystals. The origin of burbankite is considered to be as follows: sodium is initially leached from the surrounding silicates, then transported into the cave by the underground stream. The alkali becomes bal anced in solution by OH" ions through hydrolysis and after reacting with carbon dioxide yields alkali carbonates solutions that precipitated burbankite under dry and poor or no drainage conditions. Oxydes and hydroxides Both romanechite, (Ba,H10)(Mn4+,Mn3+)5010 and todorokite, (Mnl+,Ca,Mg)Mn3 4+0;H10, have been found as milli meter-thick black coatipgs on cave walls and on, or between clasts in the fossil alluvial sequence in different parts of the cave. Identification relies on infrared spectroscopy (IR) and chemical analysis. The manganese in romanechite is not com pletely oxidized, as a result both Mn4+ and Mn3+ occur in the mineral structure. This admixture of romanechite and todorokite formed when reduced manganese (Mn2+ ) travels in seeping water until an oxidizing environment (produced by guano) is encountered (WHITE, 1997). Halides The only representative of this class is the paratacamite, Cu1(0H)3CI. Presently, only its orthorhombic polymorph (atacamite) was identified in caves from Australia and South Africa (HILL & FoRTI, 1997). Paratacamite appears to be the thermodynamically stable phase at ambient temperature (SHARKEY & LEWIN 1971; POLLARD et a/ 1989). Although it has been reported as the end product of various oxidation and hydrolysis processes in oxidized zones of base metal ores in arid climates, it is certainly not confined only to such environ ments. Paratacamite was found near the Bivouac Room as finely crys talline greenish nodules growing on and within brushite-rich sediments. In addition, it has been identified admixed with pale-blue sampleite on the surface of gypsum flowers. The XRD patterns ofparatacamite are sharp (the strongest diffrac tion lines are 5.461A (100), 2 752A (55), and 2 266A (45)) and in good agreement with the data recorded in ICDD file 25-1427 ofparatacamite. The EDS spectrum obtained on dif ferent rhombohedral crystals of paratacamite is equally sharp, Cl and both Ka, and 4lines ofCu being present. The mean analytical results (and ranges) for the electron-microprobe analyses of paratacamite nodules are: CuO 74.85% (73 .5-76.1), Cl16.15% (15.4-16 9), (Hp)calc 13.6%, sum 104.6%, less 0 = C13.7, total100.9 wt.%. The deposition of paratacamite is attributed to the chemical reaction betWeen chlorine and copper. The chlorine probably derived from both bat guano and percolating meteoric waters. The copper was supplied by the rocks above, which contain disseminated sulfides.


minerals in Cioclovina Cave Table I Minerals identified in Cioclovina Cave-all minerals in bold are first time mentioned in this eave, whereas those in bold & italic were never documented from a cave environment worldwide (chemical formula according to MANOARINO, 1999) Mineraux identifies dans Ia grolle de Cioclovina lous les mineraux dontle nome est en caractere.v gras son/ mentionnes pour Ia p"miere {ois dans celle grol/e; ceux dontle nom est ecril en caracleres gras el cursifs n onljamais ete menlionnes nulle pari au monde dans le m i lieu soulerrain (formules chimiques d'apres MANDARTNO, /999). Phosphates Chemical class Minerals Calcite Carbonates Aragonite Burbanklle I Sulfates Gypsum fllite Silicates Kaolinite Quartz Goethite Ollides and hydrollides Romanechite Todorokite Halides -ParataCllmite Phosphates Ardealitc Berlinite Brushite Carbonate-hydroxylapatite Cll/ore/lestadite v' Churchite V Collinsite Crandalli tc Fluorapatite Foggite v Hydroxylapatite Leucophosphite Monetite Sampleite l Tarnnakitc & v Chemiul Formula CaC03 CaC03 (Na.Cah(Sr,Ba,Ce),(C03), a Pe)+O(OH) Ca2(S04)(HP0 4}4H2 0 '-'" AIPO v CaHP0.2H 2 0 v C:l.s(PO COlh(OH) v Ca}(Sio ro so.h(CI,F) YP042H20 v ..,.... C:IAI3(P04h(OH)$H 2 0 v v CaAI(P0.)(0HnH10 v v KFez)+(PO.h(OH)2HzO v CaHP04 v' NaCaCulcro.).a ,/ 31 Our study identified 15 phosphate minerals, four having never been recorded in the cave environment (Table I) To now, berlinite was found or synthesized only under hydro thennal conditions at temperatures exceeding 186 oc {MURAOKA & KJHARA, 1997; WISE & LOH, 1976). Below this tem perature, variscite (AIP0;2Hp) is the thennodynam i cally stable mineral fonn. Therefore we believe variscite underwent a dehydration process due to relatively high temperature con ditions caused by in-situ guano combustion that allowed its transfonnation into berlinite. The stability of the newly fonned mineral under cave conditions needs further investigations to be fully understood. Berlinile, AI PO 4 is a rare aluminum phosphate that was found as sub-millimeter crystals along cracks in well-cemented clay or impregnating the body of this clay. X-ray diffractometry showed the crystals to be entirely composed ofberlinite. The strongest diffraction lines are 3.367 A (1 00), 4.277 A (20), and 1.83A (18) being very close to those obtained from WestanA (Sweden) and Katumba (Rwanda) specimens (GAu.AGHER & GERARDS, 1963; STRUNZ, 1940). The X-ray diffraction data for the sample are given in Table 2. By using the UnitCe/1 Program (HoLLAND & REDFERN, 1997), the lattice parameters of the berlinite were calculated (Table 2); the values compare well to those given on ICDD card no. 10-423 for synthetic berlinite. The XRD of pale pinkish crusts deposited just below the limestone chunks that are buried in heavily compacted and transfonned alluvial sediments near the Bivouac Room w e r e revealed to be composed of dllorel/estadite, Ca5(Si0 4,PO 4,SO )3(CI,F). This unusual mineral is isostructural with the minerals of the apatite group (RousE & DUNN, 1982). It may have fonned due to partial replacement of phosphate by silicate and sulfate according to


32 Table 2. Indexed powder pattern ofberlinite and itsunit cell parameters. Indexation des raies de diffraction X du berlinite et les parametres reticula ires. d(A) hkl 4.277 20 100 3.989 4 101 3.619 2 003 3.367 100 102 2.471 7 110 2.303 7 104 2.255 8 112 2.140 13 200 1.991 5 202 1.830 18 114 1.682 5 204 1.666 I 210 1.615 2 106 1.551 10 212 1.461 2 116 1.425 300 1.391 5 214 1.381 7 302 1.294 3 215 1.262 2 304 1.234 4 220 1.205 2 222 a= 4.94(4)A, c = I 0.87( I )A, V= 230.1 (2) N B. P. Onac et al. the ellestadite scheme (McCoNNELL, 1937; 1938) with the compensation of vaience (2PO/-SiO/" +SOt) or during bat guano combustion when brushite, gypsum and silica might have been locally melted. Members of the ellestadite group were described elsewhere in the world and also derived from fire or thermal treatment of industrial waste (SEJKORA et a/., 1999; NEUBAUER & POLI.MANN, 1995). Discriminating between the two alternative genetic processes was, however, not possible at this stage of our investigations. Collected from several locations around the Bivouac Room were aggregates and nodules composed of transparent needle like crystals of brushite. Among these nodules matted white veins were observed. XRD revealed the material to be com posed of churchite, YP04H10, a rare-earth phosphate that contains minor amounts of La, Ce or Er. Churchite crystals under SEM appear lath-like or flattened on (010) (Fig. 4). As with the other phosphates, churchite derives its phosphorus from the abundant guano deposits whereas yttrium was prob ably leached from weathered schists. Cioclovina Cave represents the third location ever to host collinsitc, Ca1(Mg,Fe1+)(P04 ) 1H10. This rare cave mineral formed within a 4 to 5 em polychrome, thick-sandwiched hy droxylapatite crust. Collinsite appears along the boundaries between different colored lamina, exfoliating as millimeter thin walled balloons lining dissolution cavities. The mineral, iden tified by means ofXRD, shows diffraction lines similar to those reported by ONAC ef a!. (2001). Collinsite is believed to have precipitated from bat guano in damp, near-neutral pH envi ronment. Fig. 4. SEM image of churchite crystals. Image MEB des cristawc de clwrchite.


Unusual minerals in Cioclovina Cave Fluorapatite, Ca5{P0.)3 was identified by ineans of XRD and XRF in 3-5 mm thick multi-layered crusts in a narrow passage beyond the Bivouac Room. These crusts appear as patches varying in color from yellow-green to orange, to brown, and black. Under petrographic microscope, fluorapatite appears as colorless agate-like structure, being uniaxe (-). Depending on the degree of saturation of the seeping water and on pH (between 6 and 6.8), fluorapatite seems to represent an intermediate step in the development of hydroxylapatite (HUGHES et a/., 1 989; HtLL & FORTI, 1997). X-ray powder diffraction scans on black, earthy-mass aggregates collected from below brown-reddish crandalliterich clays showed (apart from a few quartz lines) poorly resolved reflections that fit the characteristic lines of foggite, CaAI(PO J(OHkH20. Under SEM, foggite appears as bunched platy aggregates. Unfortunately, the fragility of the sample made it difficult to obtain an undamaged mount and hence charge-free SEM images. Certainly, a qualitative EDS or electron-microprobe analyses would confirm the presence of foggite in this cave. Considering, however, the location of our sample, one can consider foggite as a product of the partial decomposition of crandallite. Leucophosphite, KFC'*(PO .MOH)H20, forms thin pale yellowish-brown crusts no more than I mm thick within white taranakite veins (up to 15 mm thick) in a section below the Bivouac Room. This section has top clays intermixed with fine alluvial sediments. Leucophosphite has been identified by means of XRD. Unit-cell parameters based on ten leastsquare refinement of3l XRD reflections were found to be a= 9.736.0426 A, b = 9.6464.0457 A, c= 9.7034 062 A, and 13 = 1 02.72.11 close to the ones recorded in the ICDD file 9-446. Tinsleyite, KAI2(P04)z(OH)4Hp, the isostructural aluminum variety ofleucophosphite, has been previously documented from another location in this cave by MARINCEA et a/. (2002). Monetite, CaHP04 was found inseveral locations throughout the cave along the artificial trail that cut the phosphate deposit. It occurs as white to white-yellowish friable material directly overlying the upper part of the limestone chunks that are buried in the phosphatic soil. All diffraction patterns exactly matched the standard monetite lines from ICDD file 9-80 and 9-80a. Cell dimensions were refined from the corrected d values using the UnitCe/1 Program. The cell parameters are: a = 6.9047 006 A, b= 8.576.008 A, c=6.649I.006 A, a =93.977.087, 13 =9l.559.074,andy= 127.703.066. Monetite is a dry and low pH environment indicator. Dehydration of brushite represents the most common pathway that leads to monetite. Sampleite, NaCaCu5 2(PO .J.CI-5820, as mentioned above, was identified along with paratacamite growing on gypsum flowers within or on the upper part of brushite-rich sediments. The mineral is translucent and has a pale blue color and silky luster. It was studied by XRD. So far, no other investigation methods were applied to this mineral. Genetically, sampleite was deposited from the same solutions that precipitate paratacamite. 33 Conclusions The diverse and interesting mineralogy of Cioclovina Cave is due to an unusual set of circumstances. First of all, it hosted an enormous amount of phosphate sediment as well as thousands of spelaeus (cave bear) remains within it. Combustion of gu&to has caused local transformations that may be respon. sible for the presence of at least two minerals formed herein. Nevertheless, unlike most limestone caves, Cioclovina is lo cated adjacent to metamorphic and siliceous rock terrains. The sinking streams coming from these terrains carried significant quantities of allochtonous sediments that the cave pas sages during periodic flooding. The presence of various sulfides and other trace elements (Cu, Ce, Y) within the sediments and limestone bedrock contribute to an even more di verse mineralogy. Phosphate-rich solutions imbued the sediment column under various pH conditions causing clay, com pact alluvial sediments, and limestone to be phosphatized to various degrees. This is how, many ofthe twenty-six minerals listed in Table I were precipitated solely in this highly peculiar phosphate environment ofCioclovina Cave. Acknowledgments We thank Dr. Jacques Martini for useful discussions on various aspects of the paper. We also acknowledge the support of the European Community Access to Research lnfrastmcture action of the Improving Human Potential Programme, con tract HPRICT-1999-00008 awarded to Prof. B. J. Wood (EU Geochemical Facility, University of Bristol) that enabled us to perform the XRF and part of the XRD and electron micro probe analysis. References BALINTONt, I. & BucuR, I. I. (200 I) Short outlook on the structure of the South Carpathians. In: Field trip guide 4t1t Regional Meet ing of /FAA (BucuR, I. 1., FtLtPESCU, S ., & SASARAN, E., Eds ), Prcsa Univcrsitarli Clujcanft, Cluj, pp I. CoNSTAtmNESCU; E. MARINCEA, S & CRActUN, C. ( 1999) Crandallite in the phosphate association from Cioclovina cave, Mts., Romania). In: Mineralogy in the system of the earth sci ences (ANASTASIU, N. & IUNCA G Eels.), Imperial College Press, London, pp. 1-5. DuMnMs, D. & MARtNCEA, S. (2000) Phosphates in the bat guano deposit from the "Dry" Cioclovina Cave, Mountains, Romania. Rom. J Mineral Dep 19, J pp. 43-45. GALLAGHER, M .1. & GERARos, J. F. (1963) Berlinite from Rwanda Mineralogical Magazine, 33, pp. GORAN, C .J1982) Catalogul sistematica/ pe1terilor din Romania. FRTA, cess, 496 p HALLA, F. ( 1931) lsomorphc Beziehungen und Doppe1salzbildung zwischen Gips und Brush it. Z. Krist., 80, pp. 349-352. HILL, C A. & FoRTI, P ., Eds (1997) Cave minerals ofthe world(2.. eel.), NSS, Huntsville, Alabama, 463 p


34 HoLLAND, T. J. B. & REDFERN, S. A. T. ( 1997) Unit cell refinement from powder diffraction data: the usc of regression diagnostics. Mineral. Mag., 61, pp. 65-77. HuGHI!S, J.M., CAMERON, M., & CROWLEY, K. D. (1989) Structural variations in natural F, OH, and Cl apatites. The American Min eralogist, 74, pp. 870-876. MANDARINO, J. A. ( 1999) Fleischer's Glossary of Mineral Species 1999. The Mineralogical Record Inc., Tucson, Arizona, 225 p. MARINCEA, S., DuMITRAS, D., & GIBERT, R. (2002) Tinsleyitc in the "dry" Cioclovina Cave (Sureanu Mountains, Romania): the sec ond occurrence. Eur. J. Mineral., 14, 1, pp. 157-164. McCoNNELL, D. (1937) The substitution of Si04 -and S04-groups for P04-groups in the structure of apatite, cllestadite, the endmember. The American Mineralogist,ll, pp. 997-986. McCoNNELL, D. ( 1938) A structural investigation of the isomorphism of the apatite group. The American Minera/ogist,l3, pp. 1-19. MuRAOKA, Y. & KIHARA, K. ( 1997) The temperature dependence of the crystal structure of berlinitc, a quartz-type form of AIP04 Phys. Chem. Minerals,l4, pp. 243-253. ONAC, B. P., MYLROIE, J. E., & WHrrE, W. B. (2001) Mineralogy of cave deposits on San Salvador Island, Bahamas. Carbonates and Evaporites, 16, /, pp. 8-16. PoLLARD, A. M., THOMAS, R. G., & WILLIAMS, P. A. ( 1989) Synthesis and stabilities of the basic copper(II) chlorides atacamite, paratacamite and botallackite. Mineralogical Magazine, 53, pp. 557-563. B. P. Onac et a/. RousE, R. C. & Dt1NN, P. J. ( 1982) A contribution to the crystal istry ofellestaditc and the silicate sulfate apatites. The American Mineralogist, 67, pp. 90-96. ScHADLER, J. ( 1929) Mincralogische-petrographischc carakteristik der Phosphat .ablagerung in dcr Cioclovina-1-ll!hle bei Pui. Pub/ica lii/e Muzeului Hunedoara, 5. Sc.HADLER, J. (1932) Ardealit, ein neucs Mineral + 4Hp. Central-hi. Miner. Abt. A., pp. 40-41. SEJKORA, J., HouZAR, S., & SRI!IN, V. (1999) Chlorine-rich hydroxy lapatite from Zastavky near Bmo. Acta Musei Moraviae, 84, pp. 49-59. SHARKEY, J. B. & LEWIN, S. Z. (1971) Conditions governing the for mation of atacamite and paratacamite. The American Mineralo gist, 56, pp. 179-92. Snu..J., A (1981) Geologie de Ia region de Hateg-Cioclovina-Pui-Bilnita (CarpatesMeridionales). An. Inst. Geo/. Geofiz.,66, pp. 91-179. STRUNZ, H. ( 1941) Isotypie von Berlinit mit Quartz. Z Kristallogr., 103, pp. 228-229. WHITE, W. B. ( 1997) Thermodynamic equilibrium, kinetics, activa tion barriers, and reaction mechanisms for chemical reactions in karst terrains. Environmental Geology, 30, 112, pp. 46--58. WisE, W. S. & LoH, S. E. ( 1976) Equilibria and origin of minerals in the system Alp3-AIP04-H10. The American Mineralogist, 61, pp. 409-413. received: 23 January 2002 revised: 30 July 2002 accepted: 30 July 2002


Abstract Theoretical and Applied Karstology, 15 (2002), pp. 35-41 X-ray powder data on some mineral species from Petera Curata de Ia Nandru (Hateg Basin, Romania) I Delia G. 1 *, Marincea 1 Gabriel Diaconu2 Ciprian Constantina 3 & Romeo Pavel4 Dept. of Mineralogy, Geologica/Institute of Romania, str. I, 78344, Bucharest, Romania. 1 "Emil Racovilii" Institute ofSpeleology, str. Frumoasa II, 78114 Bucharest, Romania 3 Institute ofGemo/ogy, University, str. Avram lancu II, 3400, Cluj-Napoca, Romania. 4 Museum of the Dacian and Roman Civilization, str. I Decembrie 39, Deva, 2700, Romania. In the current contribution, we briefly the main mineral species fromlhe fossil bat-guano deposit in CuratA de Ia Nandru cave, based on extensive X-ray powder diffraction study. The layered guano deposit inside the cave was well opened by recent archaeologi cal works. Hydroxylapatite and brushite are the most common constituents; associated minerals are quartz, dolomite, calcite and clay minerals (illite and kaolinite). The mean cell parameters of hydroxylapatite, taken as weighed average of nine sets of values obtained by 3 least-squares refinement of X-ray powder data, are a= 9.429(6) A, c = 6.862(16) A and V = 528.9(6) A These values account for the stoichiometry, as well, as in the case of brushite [a = 5.808(7) A, b = 15.183( I) A, c = 6.241 (8) A, p = 116.38(6) A3 for a representative sample]. Calcite [a= 4.980(2) A, c = 17.033(7) A, V= 365.8(1) A3 ] and dolomite [a= 4.807(9) A, c = 16.062(6) A, V= 321.4(1) A3 ] occur on diagenetic cracks that affect the deposit Low (alpha) quartz with a= 4.920(3) A, c = 5.406(4) A and V = 113.3( I) A3 probably allogenic, and minor kaolinite and illite are admixed with the phosphates from the guano groundmass. Key words: cave minerals, bat guano, X-ray powder data, hydroxylapatite, brushite, calcite, dolomite, quartz, CuratA de Ia Nandru. Donnees obtenues par diffration de rayons X en poudres sur quelques especes minerales de Ia grotte crPe$tera Curata de Ia Nandru (Bassin de Hateg, Roumanie) Resume Dans cette etude, no us proposons d 'offrir une caracterisation sommaire des principales especes minerales trouvees dans /e depot {ossile de guano de chauve-souris de Ia grotte dite Pe:;tera Curatii de Ia Nandru, basee sur /'utilisation extensive de Ia diffraction de rayons X en poudres. Le depot de guano intercate en couches dans les sediments de cette grotte est rendu accessible par des travaux archeologiques recents. Les composantes les plus communes sont /'hydroxylapatite et Ia brushite; les mineraux associes sont le quartz, Ia dolomite, Ia calcite et des mine raux argileux (illite et kaolinite). Les parametres reticulaires de /'hydroxylapatite, calcules sur 9 sets de donnees a/finees extraites des diflractogrammes de rayons X en poudres, sont a = 9, 429(6) A, c = 6,862(16) A et V = 528,9(6} A1 Ces valeurs indiquent Ia stoechiometrie du mineral aussi, comme dans le cas de Ia brushite [a= 5,808(7) A. b = 15,183(1) A, c = 6,241(8) A, P = 116,38(6) A1 pour un echantillon representatif]. La calcite [a= 4,980(2) A, c = 17,033(7) A, V = 365,8(1) A1 ] et Ia dolomite [a= 4,807(9) A, c = 16,062(6} A, V = 32 1,4(1) A1 ] ont ete identifiees sur des fissures diagenetiques affect ant le depot. Du quartz alpha avec a = 4, 920(3} A, c = 5, 406(4) A et V = j 13, 3(1) A1 probablement allogenique, de /'illite et de Ia kaolinite sont melanges avec les phosphates dans Ia masse de guano. Mots-ells: mine raux de grottes, guano de chauve-souris, donnees de diffraction de rayons X en poudres, hydroxylapatite, brushite, calcite, dolomite, quartz, Pe:;tera Curatii de Ia Nandru. corresponding author. E-mail address: 2002, Editura Academiei Romane. All rights reserved.


36 Introduction Pqtera Curatii de Ia Nandru (=the Clean Cave from Nandru) on named the Nandru Cave for simplicity), has a cer tain notoriety because of the archaeological works that were carried out inside. Two levels of habitation, ofMousterian age, were described (CARCIUMARU, 1980). No mineralogical inves tigations of the cave were carried out so far. The opportunity offered by the archaeological works and the accessibility of new material prompted us to initiate a brief mineralogical investigation of the fossil bat guano deposit from this cave. X-ray powder diffraction was used as principal ana lytical tool because of the frequent presence of admixtures and of the very fine-grained nature of the crystals. /Jiserici V. D. G. eta/. Geological setting Geographically, the cave is located in the Valley, in the east-northeastern part of the village Mic, about 6 km north from I-iunedoara, the major town in the area. Geologi cally, the cave is located in a calcareous ridge built of Lower Carboniferous limestones of the Hateg Basin et a/., 1980) (Fig. I). The cave is very small, being composed of a principal chamber c. 9 m-long, continued by a short passage 6 min length. A simplified sketch of the cave is given in Fig ure 2. Abundant alluvial deposits, consisting of silteous "terra rosa" with important clay participation may be observed on the floor of the cave. Fig. I Simplified geological map showing the location of the Nandru Cave (redrawn from eta/., 1980). Key: I. Lower Carboniferous (carbonaceous orthoquartzites, limestones, dolostones); 2. Cenomanian (ammonitesand gastropods bearing sandstones); 3. Lower Miocene (gravels, sands, calcareous clays); 4. Lower Sarmatian (Ostrea-bearing limestones, eincrites); 5. Upper Sarmatian (sands, gravels, sandstones, clays); 6. Holocene deluvial deposits; 7. Holocene alluvia (sands, gravels, sandy clays); 8. fault; 9. geological limit; I 0. Curatl'i de Ia Nandru. Scale I :50 000. Carte geologique simplifiee avec Ia localisation de Ia grotte de Nandm (d'apres eta/., 1980). Ugende: 1. Carboni.fore inferieur (orthoquartzites carbonates, calcaires, dolomites); 2. Cenomanien (calcaires greseu.x a ammonites et gasteropodes); 3. Miocene inferieur (graviers, sables, argiles calcaires); 4. Sarmatien inferieur (calcaires a Ostrea, cinerites); 5. Sarmatien superieur (sables, graviers, gres, argiles); 6. Depots de/uviau.x Ho/ocenes; 7 Depots al/uviau.x Holocenes (sables, graviers, argiles sableu.x); 8. faille; 9. limite geologique; I 0. Curatli de Ia Nandru. Echelle I: 50.000.


X-ray powder dot on mineral species from Peftera Curata de Ia Nandru 37 0 3m Fig. 2. Sketch of the Nandru Cave (redrawn from CARCIUMARU, 1980). Symbols in the legend represent : 1 archaeological works; 2. location of the samples. Esquisse de Ia Grote de Nandru (d'apres CiRCJUMARU, 1980). Les symboles representent: l.fouilles; 2. point d'echantillonnage. A pit approximately 3 m-Iong, 2.5 m-wide and 4 m-deep, dug during the archaeological investigations, is located in the cen tral part of the main chamber. This pit opened a bat-guano layer about I m-thick. Textural particularities show that the guano was subject to a weak diagenesis; the deposit is granu lar and only slightly indurate. Our preliminary investigations showed, however, that the ammonium-bearing mineral spe cies, characteristic for this kind of deposits, are lacking. Some of the samples used for this investigation were collected from the dumps and walls of the archaeological works and particularly from the pit dump located at the center of the main chamber The phosphate-rich zones consist of90-95 %hydroxylapatite + brushite, with minor amounts of quartz, dolomite, calcite and clay minerals (illite and kaolinite). Analytical methods X-ray powder diffraction (XRD) analysis was conducted us ing a Siemens D-5000 Kristalloflex automated diffractometer equipped with a graphite monochromator (Cu Ko. radiation, A. = 1.54056 A). The apparatus Was operated at a voltage of 40 kV, with a beam current of 30 rnA. The data were collected from 5 to 90 29, at 0.02 per second. Records were made with slit widths of0.1, 0.1, I mm and 0.6 mm (for divergence, scat ter, antidiffraction and receiving slits respectively) Synthetic fluorite (a= 5.4626 A at 20 C) was used as iin internal stan dard, and intensities were measured at peak maxima. Back ground was subtracted from raw intensity data interactively, using the computer program "Diffrac AT" in all the cases. The cell parameters were refined with the CELREF program (APPI.EMAN & EvANs, 1973), as revised for microcomputer use by BEN9IT (1987). Freshly exposed surfaces of many aggregates were observed using a JEOL J .S.M.-840 scanning electron microscope set at 15 kV acceleration voltage and I 0 nA beam current. Scanning electron microscopyenergy dispersion spectroscopy (SEM EDS) analyses were perfomied using the same apparatus, equipped with a Tracor Northern TN-2000 system. Prior to the analysis, the samples were twice covered with gold. Fluorescence tests were performed using a portable Vetter ul traviolet lamp, with 254 and 366 nm filters. Description of the main mineral species Hydroxylapatite The color of the mineral varies from beige cream to light green, depending on the frequency of fluid and mineral inclusions (iron-bearing, as shown by SEM-EDS) and on the nature of the allocromatic pigment. In all cases, the mineral is transpar ent to translucent. The luster varies from vitreous (for bunches of crystals) to earthy (for masses) and somewhat pearly. The mineral does not effervesce when etched with HCI; no fluorescence was observed either under short-or long-wave ultra violet radiation. The SEM study shows that crystals of hydroxylapatite from Nandru are always disposed as compact radiating masses mac roscopically perceived as nodules or small crusts. The indi vidual crystals are usually tabular, with diameters of less than 15 J.lm and thickness of up to I J.lm (Fig. 3). The crystals are highly fractured, and weathering products (brushite, but also some gel like, iron-bearing phases) may be observed within the fractures. In all cases the specimens showed reasonable crystallinity. However, the lower crystallinity of some of the samples pro duces a broadening of the XRD peaks, and reflections such as (213) and (321) are difficult to resolve A problem occurred in the choice of the basic symmetry ofthe mineral, for a reasonable indexing of the observed diffiaction patterns The structure of hydroxylapatite is based on P04 Ca(I)09 anctca(2)0,0H polyhedra, the hydroxyl groups being located in columns parallel to the c axis. The hydroxyl groups are too large to fit in the Ca triangles and are consequently displaced along c by 0.35 A from the normal position in


38 Fig. 3. SEM photomicrographs of typical h yd ro xy lapatite aggregates. Images MEB des agregats typiques d 'hydro:xy / ap at it e Table I Crystallographic pa r amete r s of h y droxylapatite sampl es f ro m Nandr.u Cave. Parametres Clysta/lographiques des echa n ti llons d 'hydroxylapatite de Ia grotte de Nandru. a V{A3! n N NlA 9.432(3) 6.863(3) 528 8 ( 3) 3 53 NIB 9.426(3) 6.875(2) 529 1 (3) 3 42 NIC 9.433(1) 6.822(2) 530.1 ( 1) 10 32 N3A 9.435(4) 6.86 1 (5) 5 28.9(5) 3 35 NSA 9.435(2) 6.866(2) 529.3(2) 4 56 N6A 9.430(3) 6.862(3) 528 5 ( 3) 3 40 N 6 B 9.413(2) 6 879{2) 527. 9(2) 8 36 N6C 9.428( 1 ) 6.870( 1 ) 528 8 (1) 7 3 6 N6E 9.4 28(3 ) 528. 7 (4) 3 25 I = number of refinement cycles 2 = number of retlections used for r e fin e ment (28 = I 0-8 5). D. G. eta/. fluorapatite (KAY et a/., 1964, BRUNET et a/ 1999). This re sults in a monoclinic superstructure mentioned by ELLIOT et a/. (1973) and in the lowering of the symmetry from hexago nal (P6im) to monoclinic (P21/b) The difference between the unit-cell parameters a and b is however too small to be easily observed (according to HuGHES et a!., 1989, a is with only 0.002 A greater than b) and the Pangle closely approximates 120 C Consequently, within the resolution of the diffraction patterns perfonned during this study the diffraction peaks correspond ing to a hexagonal unit-cell were taken into consideration Note that in all the diffractometric records the single peaks remain unsplit and the cell parameters were successfully and reliable refined based on hexagonal cells In spite of the poor crystal linity of some of the samples, which resulted in reduced inten sities of the main reflections and broadening of some others, indexing was always possible. The XRD patterns were indexed in analogy with PDF 86-1201. Sets of 25 to 56 reflections were used to refine the cell dimensions, which are summa rized in Table 1 The full set ofX-ray powder data is available from the corresponding author upon request. The mean cell parameters resulted as weighed average of the values in Table I are a = 9.429(6) A, c = 6.862(16) A and V = 528.9(6) A3 (standard deviations, calculated as Std = a n-J are given in brackets). These values are consistent with the data obtained for the synthetic hydroxylapatite by BIGI e t a/. (1996) or BRUNET eta/. (1999): a = 9.421(2) A c=6.882(2) A and V = 529.0(4) A3 The differences between the cell parameters in Table 1 may reflect chemical variations [i e ., the (F CI)-forOH substitutions] and within the l i mits of the error, are quite nonnal even for pure (synthetic) hydroxylapatite (e g ., SMITH & LEHR, 1966). Brushite Brushite occurs as a snow-wh i te powdery coating on hydroxylapatite or as nodular earthy masses (several mm to 0.5 em in diameter) in the bat guano groundmass which is composed principally by hydroxylapat i te. Brushite from Nandru does not fluoresce under either short w ave (254 nm) or long-w a v e (366 nm) ultraviolet light. The SEM study shows that the mineral occurs as irregular lin ing of some hydroxylapatite bunches of crystals, partial fill ings of micro-veins or cracks affecting the hydroxylapatite mass, or, more frequently, as ov e rgrowths of hydroxylapatite aggregates, in which the textural relationships clearly show that brushite postdates hydro x ylapatite The irregular aggre g a tes of brushite are very fine grained As may be observed in F i g. 4, the individual cryst a ls are platy on (0 l 0), and may reach up to 1 0 J.lm across, no more than 5 J.lm wide and 1 J.lm thick In habit they closely resemble syn t hetic brushite illustrated by RINAUDO eta!. (1994) or S JVAKUMAR eta/. (1998), but no crys tallographic fonns may be distinguished.


X-ray powder dat on mineral species from Curata de Ia Nandru 39 Calculated and measured X-ray powder diffraction data for this mineral are given in Table 2. The lines were indexed in analogy with PDF 72-0713, which is reproduced for comparison in the table. The cell parameters of the brushite samples from Nandru are close to those of synthetic brushite (a= 5.812(2) A, p = 15.180(3) A, c = 6.239(2) A, b = 116.43(3) A3 according to B EEVERS, 1958 or CURRY & JONES, 1971) indicating that their chemistry is reasonably close to the ideal composition. Quartz Quartz was identified in both phosphateand carbonate-bearing samples from the guano deposit. Because of its small crystal siz e and dispersed nature, this mineral may be completely camouflaged in the phosphate matrix and therefore difficult to separate The textural relationships alone failed to offer valu able criteria to discriminate between the authigenic and the allogenic nature of quartz. The lack of any associated AI-and K bearing phosphate (e.g., taranakite, leucophosphite), prompted us, however, to consider the allogenic nature of this mineral species, because reactions involving clay minerals and H3P04 normally produce authigenic quartz in excess (see MARINCEA eta/., 2002) The X-ray powder patterns ofboth the handpicked separates and ofthe residua obtained after the selective dissolution ofhydroxylapatite by HCIIeaching indicate in all cases the presence of! ow (alpha) quartz. The cell parameters obtained by l e ast squares refinement of 25 XRD reflections obtained for a representative sample(N5A)area = 4 920(3)A,c =5.406(4) A and V = 113.3(1) A3 They are relatively close to those determined by WILL et a/. Fig. 4. SEM photomicrograph of a typical brushite aggregate. Image MEB d'un agregat typique de brushite. (1988) for the stoichiometric@quartz [a= 4.91239(4) A c = 5.40385(7) A and V= 112.933 A3]. Dolomite Dolomite was found sporadically in t}le bat-guano deposit from Nandru The mineral occurs as massive, white-pinkish deposits on some diagenetic cracks affecting the guano mass This textural particularity, as well as the availability of Mg2+ de rived from the Sarmatian cinerites in the area eta/., 1980), suggest an authigenic, late diagenetic nature of this carbonate, which was probably deposited by direct precipita tion from a Mg-rich "moonmilk". A mechanism of dissolution recrystallization type may also be taken into consideration. The cell parameters refined for a representative sample (N4) from 20 XRD reflections in the 29 range between 10 and 88, are a= 4.807(9) A, c = 16.062(6) A and V = 321.4(1) A3 Note that the a cell parameter closely match that found by REED E R ( 1983) for a nearby stoichiometric dolomite from Lake Arthur, United States [a = 4 8069(2) A], whereas the value of c is g reater than the value obtained by the quoted author for the same sample [c = 16.0034(6) A], suggesting limited (Mn,Fe )-for-Mg substitutions. Calcite Calcite is scarce in the guano deposit from Nandru C ave, being common only in the sediments in vicinity. In the phosphate-bearing samples, the mineral occurs as milky-white semitranspar ent, filling oflate diagenetic veins. The cell parameters refined for a representative sample (N6) are a =4.980(2) A, c= 17.033(7) A and V =365.8(1 )AJ They do not differ significantly from those


40 D. G. DumilrOf et a/. Table 2. X-ray powder diffraction data for selected samples of brushite from Nandru Cave1 Donnees des diffractions de rayons X en poudres des echantillonr de brush ite de Ia Grotte de Nandru. Crt. Sample N2A SampleNJA PDF 72-0713 (hkl) no. dmc.,.(A) d cal. (A) In .. (A) d.,.dA) Inn d "'"" CA) deal J...Aj2l 1110 I 7.563I 7.59I5 IOO 7.5249 7.5823 IOO 7.590p 7.5899 100 (020) 2 4.230I 4.2359 43 4.2192 4.2332 64 4.2371 4.2372 79 (121) 3 3.7899 3.7957 7 3.7847 3.79I2 .9 3.7950 3.7942 5 (040) 4 3.0429 3.0459 42 3.0384 3.043I 57 3.0460 3.0461 54 (14 I) 5 3.0429 3.0408 42 3.0384 3.0375 57 3.0460 3.0404* 54 (I I 2) 6 2.9223 2.9257 I 2.9277 2.9234 5 2.9244 2.9240 35 (121) 7 2.8501 2.85I7 4 2.8444 2.85I7 4 2.8534 2.8533 6 (21 I) 8 2.6678 2.6684 I ---2.6675 2.6675 2 (051) 9 2.6209 2.6227 18 2.6I85 2.6203 25 2.62I5 2.6223 32 (150) IO 2.6209 2.6232 I8 2.6I85 2.6I98 25 2.62I5 2.62I3 32. (022) II 2.5997 2.6017 I3 2.5953 2.60I6 II 2.6022 2.6020 20 (200) I2 2.2659 2.2672 2 -2.2670 2.2670 3 (161) 13 2.1696 2.1705 7 2.1685 2.1680 12 2.1701 2.1702 12 (15 2) I4 2.1448 2.1460 8 2.1422 2.1451 14 2.1461 2.1460 II (240) 15 2.0980 2.0985 2 2.0989 2.0974 8 2.0990 2.0990 4 (251) 16 2.0822 2.0836 4 ---2.0822 2.0820 6 (112) I7 2.0I92 2.0196 I 2.0151 2.0188 5 2.0191 2.0189 2 (211) I8 1.9728 1.9736 I 1.9788 1.9719* 7 1.973S 1.9739 3 (213) 19 1.8979 1.8979 I 1.8950 1.8956 5 I.8975 1.8975 2 (080) 20 I.8755 1.8760 6 1.8750 1.8737 IO I.8752 1.8751 IO (062) 21 I.8565 I.8566 4 1.8545 1.8567 6 I.8575 1.8574 5 (321) 22 1.8147 1.8140 II 1.8116 1.8128 20 1.8139 1.8139 12 (260) 23 1.7969 1.7970 4 1.7949 1.7955 9 1.7972 1.7972 6 (26 2) 24 1.7788 1.7789 2 ---I. 7787 1.7787 2 (181) 25 1.6 I 54 1.6151 I 1.616I 1.6131 3 1.6146 1.6146 I (091) 26 I.5632 I.5702 I 1.5680 I.5682 3 1.5696 1.5696 2 (082) 27 I.5520 1.55I8 2 1.5514 I.5502 c. 1.5516 1.55I6 5 (20 4) 28 I .4352 1.4346 I 1.4392 1.4330* 3 1.4336 1.4335 3 (I43) 29 1.3914 1.3912 I --I.3921 1.392I I -(413) 30 I.3696 I.3694 2 1.3690 1.3679 7 I.3690 I.3690 3 (1.10.1) 3I 1.3696 1.367I 2 1.3628 I.3653 3 I.3664 I.3664 2 (I5 4) 32 I.3346 I.3400 I 1.3372 1.3384 3 I.3397 I.3397 2 (O.I 1.1) 33 I.3346 1.3341 I I.3305 1.3326 3 1.3339 1.3339 I ( 1.1 1.0) 34 I.3346 1.334I I 1.3305 1.3326 3 I.3339 1.3337 I (0. I0.2) 35 1.30I4 I.3009 I 1.30I6 1.3008 2 I.JOI I 1.30IO 2 (400) 36 I.2280 1.2281 I -I.2278 1.2279 I -(1.12. I) 37 1.2169 1.2I67 I --1.2 I62 1.2I61 2 (282) 38 1.1529 1.153 I I 1.1521 1.1524 5 1.1523 I. I526 2 (352) 39 I. I529 1.1527 I 1.1521 1.1513 5 1.1523 1.1523 2 (0.12.2) 40 1.1036 1.1027 I ---1.1033 1.1033 I (543) a= 5.808(7) A a= 5.808(2) A a= 5.812(2) A b= I5.I83(1)A b =15.165(4) A b =: 15. I80(4) A c = 6.241(8) A r 6.233(2) A c"" 6.239(3) A = 116.38(6)0 = 116 39(2)0 I6 .44(1)0 1 Monochromatized Cu K., radiation, A.= 1.54056 A, 28 = Number of refining cycles: 8; 6; 3. 2 As calculated by us, using the CELREF program (APPLEMAN & EVANS, 1973). Asterisk in d.,.1 column refers to reflections rejected during the refinement.


X-ray powder dat on mineral species from Pe1tera Curata de Ia Nandru 41 detennined by EFFENBERGER eta/. ( 1981) for the synthetic cal cite [a=4.9896(2) A and c= 17.0610(11) A], which accounts for the stoichiometry. Discussion and conclusions The (hydroxylapatite+ brushite) mineral association is one of the most common in the fossil bat-guano deposits from caves (see Hn1..& FoRTI, 1997, and referred works). Both minerals were fonned by interaction between the bat guano and the phosphate rich solutions derived from it and the calcareous basement or the "moonmilk" deposits. This suggested origin appears to be sup ported by the experimental work ofEwoT eta/. (1959) that showed that calcium phosphate precipitates from saturated phosphoric as brushite up to pH 6.93 and as hydroxy lapatite at higher pH levels. Variations in the acidity of the guano-derived solutions may be caused by the breakdown of calcium carbonate as a result of its interaction with the phos phoric acid in these solutions. One will note that in the bat guano deposit from Nandru, hydroxylapatite always overcoats References APPLEMAN, D. E & EvANS, H T., JR. ( 1973) Indexing and least-squares refinement of powder diffraction data U.S. Geo/. Surv., Comput. Contrib., 20, (NTIS Doc PB-216). BEEVERS, C. A. ( 1958) The crystal structure of dicalcium phosphate dihydrate, Acta Cryst., 11, pp. 273-277 BENOIT, P. H. (I987) Adaptation to microcomputer oftheApplemanEvans program for indexing and least-squares refinement of powder-diffraction data for unit-cell d imensions. Am. Mineral., 12, pp. 1018-1019. B1m, A., FAUN!, G., FoRES11, E., GAZZANo, M, RwAMmm, A., & RovER!, N. (1996) Rietveld structure refinements of calcium hydroxylapatite containing magnesium. Acta Cryst., B 52, pp.87-92 BRUNIIT, F., ALl.AN, D R REDFERN, A. T.S-., ANGEl. R.J ., MILETICH, R., REICHMANN, H.J. SERGENT, J., & HANFLAND, M. ( 1999) Compressibility and thermal expansivity of synthetic apatites, with X= OH, F and Cl. Eur. J. Mineral., 11, pp. 1023-1035. CARCIUMARU, M. ( 1980) Mediul geograjic in Pleistocenul superior culturi/e pa/eolitice din Romania. Ed. Academiei R.S.R, 268 p CuRRY, N.A. & JoNES, D.W. (1971) Crystal structure ofbrushite, calcium hydrogen orthophosphate dihydrate: a neutron-diffraction investigation. J. Chem Soc A, pp. 3725-3729. EFFENBERGER, H., MEREITER, K., & ZEMANN, J. ( 1981) Crystal structure refinements of magnesite, calcite, rhodochrosite, siderite, smithsonite and dolomite, with discussion of some aspects of the stereochemistry of calcite-type carbonates. Z Kristallogr., 156, pp. 233-243 Et.LIOT, J .S., SHARP, R.F., & LEwis, L (1959) The effect of the molar Ca!P ratio upon the crystallization ofbrushite and apatite. J. Phys. Chem.,63,pp. 725-726 . EwoT, J. C., MACKIE, P E., & YoUNG, R. A. (1973) Monoclinic hydroxylapatite. Science, 180, pp. 1055-1057. the carbonaceous ground, whereas brushite occurs distally, as crusts on hydroxylapatite. Calcite occurs in direct contact with brushite only in the fillings of the secondary produced by the diagenetic desiccation of the guano deposit The persistence of unreacted illite and kaolinite inside the guano deposit shows, however, that the phosphoric solutions were not enough aggressive to decompose the clay minerals with fonnation of taranakite or leucophos phite + quartz, and places the lower pH limit at about 4.58 (the "pure brushite'1 limit found by EwoT et al., 1959). Acknowledgements This research was partially funded by the Rh6ne-Aipes region, through a TEMPRA fellowshipawarded to S.M in 1999-2000. XRD patterns were recorded in the SPIN laboratory of Ecole Nationale Superieure des Mines de Saint-Etienne, whose staff is gently acknowledged. The study was also supported by a grant from the Romanian Ministry of Education and Research (program CERES, contract MEC-C 113/2001 ). H1t.1., C. & FoRTI, P ( 1997) Cave minerals of the world Second Edi tion. Natl. Speleological Society, Hunstville, Alabama, 240 p. HuGHES, J. M CAMERON, M., & CROWLEY, K. D (1989) Structural variations in natural F, OH and Cl apatites. Am Mineral., 74, pp. 870-876 KAY, M I., YoUNG, R A., & PosNER, A S (I 964) Crystal structure of hydroxylapatite Nature, 204 pp. 1050-1052 MARINCEA, . D .; & GmERT, R. (2002) Tinsleyite in the "Dry" Cioclovina Cave (Sureanu Mountains, Romania): The second world occurrence. Eur. J. Mineral., 14, pp. 157-164. M., G., KRAUTNER, H. G., KRAUTNER, F., TICLEANU, N STANCU, J., PorEScu, A., & PoPESCU, G. (1980) Geological map of Romania, I :50000. Explanatory note. Hunedoara sheet. Inst. Geol. Geof, Bucharest. REEDER, R J . ( 1983) Crystal chemistry of the rhombohedral carbonates. Rev. Mineral ., II, pp. 1-47 RINAUDO, c LANFRANCO, A M., & FRANCHINI-ANGELA, M. (1994): crystallizations from calcium phosphate solutions in the presence of SO/-. J. Cryst. Growth, 142, pp. 184-192. SIVAKUMAR, G.R.,. GIRUA, E.K., NARAYANA KALKURA, s., & SUBRAMANIAN, C. ( 1998): Crystallization and characterization of calcium phosphates : brushite and monetite. Cryst Res. Techno/., 33, pp. 197-205. SMITH, 1. P. & LEHR, J. R (1966)An X-ray investigation of carbonate apatites. J. Agr. Food Chem., 14, pp. 342-349 Wlt.L, G., BEt.-rorro, M., PARISH, W., & HART, M. (1988) Crystal struc tures of quartz and magnesium gerrnanate by profile analysis of synchrotron-radiation high-resolution powder data. J. Appl. Cryst., 21, pp. 182-191. received: 4 tfpril 2002 revised: /0 November 2002 accepted: 19 November 2002


Abstract Theoretical and Applied Karstology, 15 (2002), pp. Karst of the ridge Greda: fluvial influences, caves, and groundwater circulation Jelena Calic-Ljubojevic1 2* & Vladimir Ljubojevic2 I Geographic Institute ".Jovan Cvijic" of the Serbian Academy of Sciences and Arts, fJureJakSica 9, 11000 B_elgrade, Serbia, Yugoslavia. Student Speleologic ami Alpinistic Club (ASAK), Studentski trg 16, I 1000 Belgrade, Serbia, Yugoslavia. Dfcvrinska Greda is an elongated and narrow ridge of Upper Jurassic limestone, situated next to the Danube Gorge (Iron Gates) in Eastern Serbia. It is uplifted along the conspicuous D'.cvrin Faull which extends further to the north, to the Mehcdinti l>latcau in Romania, and it is surrounded by non-carbonate rocks to the East (Cretaceous nysch and para-nysch) and to the West (Proterozoic and Paleozoic schists). The relationships between non-carbonate and carbonate lithologies made this karst sobject to strong innucncc of allogenic water input. Due to the small width of the ridge (max. 700 m, 250m in average), nuvial innucncc is strong enough to penetrate to the opposite boundary of the limestone. Severn! separate input-output systems of karst groundwutcr were determined. limestone surface docs not exceed 5 km1 but 32 caves which altogether have more than 6500 m of passages were explored so far, indicating a significant karst development. The springs arc mostly permanent, although relatively weak (up to I 0 1/s), and show little seasonal variations. Three springs have elevated temperatures-from 17 to 19 C. Characteristics of karst springs are the indication of retention capabilities of the karst aquifer as well as of a circulation of the groundwater at great depths. Deep circulation is mostly developed due to the favourable conditions along the regional dislocation -the Dfevrin Fault. Key words: contact knrst, tluviokarst, warm karst springs, Carpatho-Balkanides Resume Le karst de Ia crete Dievrinska Greda: influences fluviati/es, grottes et circulation souterraine Dievrinska Greda est une crete etroite et allongee formee surles calcaires du Jurassique superieur dans Ia Serbie de /'Est pres du Danube, dans Ia zone des Portes de Fer. La crete est soulewJe le long de Ia faille de Dievrin (prolongee vers /e nord dans /e Plateau de Mehedin(i, en Roumanie) en direction N-S et entouree par des roches non-carbonatees a /'est (le flysch et para-flysch du Cretace) eta /'ouest (des schistes cristallins proterozol'ques et pateozoiques) Le rapport entre les formations carbonatees et non-carbonatees a determine une importante influence de l'eau all ogene sur ce karst Com me Ia largeur de Ia crete est reduite (700 m au maximum, 250m en moyenne), /'influence fluviatile est suffisamment import ante pour que les rivieres penetrentles calcaires Quelques systemes de drainage distincls on/ ele mis en evidence Qui que Ia Sllrface d 'aflleurement des calcaires ne depasse pas 5 km1 32 grolles avec une longueur cumulee de plus de 6500 m on/ ete exp/oreesjusqu 'a present./ait qui prouve 1111 developpement significatif du karst. La plupart des sources sont permanentes, me me si leurs debits son/foibles (

44 Location, Geological Surroundings and Evolution The limestone ridge Dfevrinska Greda is situated in Eastern Serbia and belongs to the Southern Carpathians Further towards the south, there is the continuation to the westernmost part of Mt. Balkan, so one may also consider that Dfevrinska Greda, as well as other mountains and ridges of Eastern Serbia, belongs to the Carpatho-Balkan mountain range. Dfevrinska Greda stretches in a N-S direction, over a length of 20 km, along the regional dislocation of Dfcvrin Fault. With its northernmost point, Dfevrinska Greda reaches close to the Danube Gorge (the Iron Gates) and stretches further to the north, towards the Mehedinti plateau in Romania (Fig I). One of the Romanian most famous cave systems, Topolnita Epuran, is situated in the portion of limestones which corre sponds to those of the ridge Dzevrinska Greda The ridge is situated within the system of great nappes of the Southern Carpathians. It is formed of limestones of Upper Jurassic age. On the west, Dfevrin Fault makes its geological contact with the Proterozoic and Paleozoic schists (gneisses, micaschists, gneiss-micaschists, amphibolitcs). On the east ern side of the ridge, the limestones continue further to the east, but are covered with Cretaceous flysch-like sediments. The schists belong to the so-called Getic Nappe (and one sub Getic Nappe), while the Cretaceous formations are partly autochtonous (lower beds, age Albian to Senonian), and partly belong to the Severin Nappe (Lower Cretaceous). All these geological units are present also on the other side of the Danube, in Romania One may assume that the overthrust took place after the end of the Cretaceous, and that the faulting is a younger process since the nappes are disturbed by the faults as well (Fig. 2) On the official geological map (BoGDANOVIC et a/, 1973), the conspicuous limestone peak ofDzevrin is mapped as the north ernmost point of the ridge (on the right bank of the Danube). However, recent detailed sedimentological surveys showed that the limestone of the peak Dfevrin does not belong to the main ridge ofDzevrinska Greda, but represents a separate olistholitic block (GRUBIC, 1992). This fact does not have dir ect conse quences on the karst studies of Dzevrinska Greda but it is worth mentioning for a correct interpretation of large scale structural and stratigraphic relations in the area Surface Morphology and Karstic Fluvial Interface General inclination of the topographic surface is from west to east, and the river courses are mostly oriented in this way. The only exceptions are several weak sinking streams which are using short surf.1ces where flysch is inc! ined towards the I imestone, ROMANIA SERBIA ..... 0 . ... J. Calic-Ljubojevic & V. Ljubojevic 0 . [Z] 1. L:...:.:...:..: I,_, -l 3 5km Pig. I Simplified geological map of the limestone ridge Dzevrinska Grcda (south ofthe Danube, Serbia) and Mehedinti Plateau (north of the Danube, Romania). According to the national Geological Surveys, adapted. I : Limestone ridges (Upper Jurassic/ Lower Cretaceous); 2 : Cretaceous sediments (Autochtonous and Severin Nappe); 3 : Proterozoic and Paleozoic schists (Getic and sub-Gctic nappes). Carte geologique simplifiee de Ia crete calcaire Dzevrinska Greda (.l'lld du Danube. Serbie) et d11 plateau de Mehedin{i (nord du Danube Roumanie) D apres les cartes des services geologiques nationaux, adaptees I : cretes calcaires (Jurassique superieur/ Cretace inferieur) ; 2 : sediments cretaces (Autochtone et Nappe de Severin); 3 : schistes proterozoi'ques et pa/eozol'ques (Nappes Getiques el


Kar.vt of the ridge Dievrinska Greda and three short gorges at the southern part of the ridge, which flow to the river systeni inclined towards the south. Four river systems of W-E orientation traverse the limestone ridge, and only the trunk streams managed to entrench into limestone and fonn through-gorges (Fig. 3). Severin napp (Sinaia beds) autochtonc ___... Pt "' ....... rJ "' "" ...., ] / ;:; ,;-., ...... ...... ...;. "' Gctic nappe I ..... J Severin nappe .AP"f'i:.\=1.!:._ (s 1 t ) t 0 I .......... ,:----. .,. . .. . -Y':O J 11\.11.1 1\:l s f CJ ( autochtone ridge Dzevrinska Grcda Fig. 2 Sketch of the general geological sequence of napping and faulting in the aren of the Danube Gorge, and particularly D2evrinska Grcda. View to the north. I:.Squisse de I 'evolution geologique des charriages et dislocations dans Ia zone du Defile du Danube et particulierement dans celle de Dievrinska Greda. Vue vers le nord. 45 Regarding the wider regional morphology, there are still dis putes about the origin of the vast levelled surface with .an av erage height of 400 m a.s.l., which is dissecte d by the above mentioned streams. Is it a remnant of the Dacian Sea (abra sion or is its origin tluvio-denudational? There was no detailed research on this issue and for now, each option is hard to be completely argued. After the regression of the sea, the tectonic activity was too !ntense to allow the fonnation of such a large and flat fluvial levelled surfaces (a series oflower ones extends further to the east). On the other hand, it is as well hard to prove that the possible marine terrace (probably of Pliocene age) could have been preserved until nowadays. The total surface of exposed limestones of Drevrinska Greda is relatively small -less than 5 km2 The width of the ridge vary from several tens of metres to a maximum of 700 m (in the region of Drenjar Hills), with an average of about 250 m. Apart from the mai n ridge, limestones are also present as outcrops within the autochtonous Cretaceous fonnation on the east. The presence of impervious terrains on both sides ofthe karst ridge gives to this area an attribute of contact karst and/or fluviokarst, with prominent allogenic influence Two main types of contact are tectonic (on the west) and sedimentary (on the east) 'Reactions' of surface waters when they reach karst are various. The rivers arc either 'stronger' than karstthey entrench and fonn through-gorges; or there is the typical karst case-streams sink and res urge after a certain portion of subterrannean flow (either partially traversable (caves) or not). This depends on several factors: first, tectonic pattern and previous development of karst conduits, which probably made some portions of limestone more 'attractive' to be attacked by surface waters ; second, as this is the case of positive feedback effect, greater discharge of a river strengthens it to pass the I imestone barrier. Some oft he gorges may also have been fonned by cave roof collapse, but there are no obvious evidences for this. Areas of fluvial influence from western and eastern side are overlapping or the influence from the western side is so strong that it reaches the opposite boundary of karst. Dzevrinska Greda fits into the model of stripe karst, elabo rated by LAURITZEN (200 I). It is a double-sided karst with a large allogenic perimeter relative to the area; more precisely, its length to average width ratio (y) is 80. Drevrinska Greda has a contact which is by its characteristics closest to stripe karst contact type I: sub vertical unconfined; while caves are mostly of morpho type D i.e. extensive, linear drainage routes (see Figures 6 and 8). Tectonic conditions The dominant structural element of the area is Dz e vrin Fault. It has a corl'lplex kinematics (Fig 4); one component of the movement is vertical (nonnal), with a sub-vertical fault plane (eastern block uplifted, as seen also in Fig. 2), while another component is of strike-slip character, which makes this fault


46 .. ... / ; __ I I \ \ Dunube 'i, ) Kasajna cave ( ( I I : ' \ ---' '. (. / ) '-. ( --( ... .. \ ponors springs through gorges \ ' J. Calic-Ljubojevit & V. Ljubojevit Fig. 3. Position of Dlevrinska Greda in relation to the surfac e drainage network, and some of the typical contact karst features. La position de Dfevrinska Greda par rapport au reseau de drainage de surface et quelques elements typiques du karst de contact.


Karst f!ftlze ridge D:!evrinska Greda + + a b 47 I I ----. I '""'" I \I .,_ ... ,.. .. \1 f': + -o-. '... {,.,. , ) \ c Fig. 4. Sketch pfthe kinematics of the Dzevrin fault (on the left) and Schmidt's net projections of fracture planes along the limestone ridge (a: the whole ridge; b: the Drcnjar I-I ills area; c: limestone outcrops to the east of the main ridge) Esquisse de Ia cinemalique de Ia faiiie de Dzevrin (a gauche) el projeclions sur le reseau de Schmid/ des plans defraclure le long de Ia crete calcaire (a: pour louie Ia crele; b: pour Ia zone des Co/lines de Drenjar; c: aj)leuremenls calcaires a I 'esl de Ia crete principale). also dextral transcurrent (the eastern block moved towards the south). The opinion among Serbian geologists is that the vertical movements have stopped and that, currently, only the transcurrent movement is active (GRUBIC, 1992). The fault plane dips towards the west, but very prominent, even more numerous, are as well those fracture planes that dip towards the north, perpendicular to the main fault (see net projections in Fig. 4). However, caves are mostly developed paralelly to the fault strike (N-S), using tension joints of the vertical movement. Some passages are formed along tension joints of the trans current movement (striking approximately SW-NE). Escarpments are visible in many places, but the central zones ofthe faultcrushed and broken (in the sense of CAR & SEBELA, 200 I) are missing from the surface, because they were eroded or are partially masked by screes; no tectonic breccia nor mylonite can be seen. Only the fiss4red zone remains visible (Fig. 5). Water Input to Karst, Groundwater Directions, and Warm Springs Three types of water input to karst are present: allogenic streams give concentrated input, precipitation gives a dispersed inpttt, while input from surrounding aquifers is in small quantities at numerous places, mostly from the western fissured aquifer in the schists. Sinking streams are in most cases seasonal (with small catchment areas and greatly depending on rainfall), and have discharges of 0-20 1/s, while the imp01tant characteristic of all springs is that they are petmanent and they do not exert typically karstic behaviour: the relation between minimal and maximal yields is never greater than approximately 1: 1 0 and the reaction to drought is relatively smooth. Pt, Pz 3 zone 'I II 2 bmken zone 3 fissured zone 1 Fig. 5. Cross-section through the main fault zone of the Dzevrin fault (actordirtg to the terminology of CAR & SEBELA, 200 I). Coupe a lravers Ia zone principale deji-aclure de Ia faiile de Dievrin (d'apres Ia /erminologie de CAll & SHBHI.A, 200/).


48 Considering that the precipitation in this area is relatively scarce (average 600-700 mm) and that the sinking streams seasonally dry out, permanent springs (although sometimes with flow rates of only about 0.5 lis) prove that a supply from another aquifer (fissured schists) exists, and also that the karst aquifer is of considerable depth, with good retention capability. The dominant direction of groundwater flow is from south to north, along the ridge, regardless of the W-E direction of the surface streams. However, due to the very small width of the limestone, there is no proven circulation through the whole ridge, but several input-output systems. Output points (springs) are usually in the gorges, and in all cases, further to the north, another system is present. Only in a few cases groundwater directions are in accordance with surface streams-in the case of Mali Drenjar and river; although it is quite disputable whether waters, which occasionally flow through a short tunnel cave, can be regarded as groundwater at all. One of the peculiar characteristics of the karst ofDzevrinska Greda is the presence of three lukewarm springs, all of them with the temperatures 17-19 oc and followed by gases (mostly NIKOLIC OUTCROP c:noe Pnr J. C'alic-Ljubojevic & V. Ljubojevic nitrogen, but also oxygen and C02}. The northernmost spring Dzevrinska Banja, which is now submerged by the Danube "Iron Gates" artificial lake, has a high content of NaCI (85 mg/1}. The total mineralization in all warm springs is up to 530 mg/1 (FILIPOVIC eta/., 1.985). The southernmost point of Dzevrinska Greda is also characterized by the presence of a warm spring-Banjica, with a discharge of about 3-5 1/s. these two locations, Zuli krs limestone outcrop, is located the Banja warm spring, which has quite a compli cated setting. There are actually three places of warm water outflow (on opposite sides of the river bed, and in the river bed itself), while in the very close vicinity there is a temporary cold water outflow from the cave Velika PeCina, and another cold springjust about 50 m upstream. The cave Velika Pecina is hydrologically connected with the cave Nikolica Ponor, situ ated in another limestone outcropNikolic outcrop (see Fig. I and 3 for location of outcrops and Fig. 6 for cross-section). That connection was not proven by tracing, but it is rather obvious from the directions of cave passages, as well as from the presence of schist sand and pebbles in the cave Velika Pecina (the catchment area of Nikolica Ponor is built on schists). Fig. 6. Cross section through the limestone outcrops, along the direction 200-20, with indicated of cold and warm groundwater. Coupe transversale des a.fJleuremen/s calcaires le long de Ia direction 200-20, avec des indications sur/a circulation des eaux souterraines fi'oides e/ chaudes. Mt. Miroc / Dzevrinska Greda w E Fig. 7. Cross section through MI. Miro

Karst 1!1' rid$e Dier,r.inska Greda Pt, Pz schists limestone . ; characteristic cross-sections 30m C C' Kl SISTEM Kt,z flysch < ... '" .. -' '0,5 m 5m Fig. 8. Plan of the cave Drenjarski Sistem (3730 m) and typical cross-sections. Plan de Ia grotte Drenjarski Sistem (3720 m) et sections transversa/es typiques. 49 Hydrogeologists consider that the wanning of the water is caused by the deep circulation along the Dfevrin Fault (due to thermic gradient), as well as by the presence of magmatic bodies. For example, 1.5 km east from the Banja spring, there is a dike of spilites which may act as a warming body for the groundwater (DRAGISIC et a/., 1988). Karstification Extent Considering the small limestone surface and the numerous types of water input, we can consider that karst is considerably developed on Dzevrinska Greda. The density of surveyed cave passages is quite high taking into account the relatively small volume of rock. The presence of the regional fault enabled the deep circulation of the water, so although these parts are im penetrable, characteristics of these springs indicate the exist ence of conduits at greater depths. Unfortunately, it is still unknown whether the limestones of Dzevrinska Greda are currently in contact with the Mt.Mir

so also conspicuous traces of tectonic activity, which can all be seen on several typical cross sections from the Drenjarski Sistem. We already mentioned that on several places surface rivers managed to entrench into the ridge, form through-gorges, and continue their surface flow. There are two cases of caves on opposite sides of a gorge (one of those in the Zuti outcrop, sec Fig. 6). The genesis of such caves is still not clearly understood. One of the possibilities is that these gorges are a younger feature that intersected the existing caves; while other options arc either the formation of caves in accordance to river entrenchment, or even by backflooding of the river. Presently, these caves are out or"permancnt hydrological function; unfortunately, the scallops in them arc insufficiently dis tinct to show the direction of the paleo-flow, while other possible evidence has not been found yet. Concluding remarks Taking into account the fact that the surface of exposed limestone on Greda is quite small (less than 5 km2), total length of all cave passages explored so far (more than 6500 m; when 32 caves are summed up) can be considered as References 13ANDRAilUR, G., SI.AVOACA. D.,I3ANI>RAIJUR. R., & SLAVOACA, R. ( 1999) Considerations hydrodynruniqucs sur lc systcmc karstiquc de 11iirza (Monts ctl'lateau de Mchcdinti, Roumanic). Them: IIppi Karsto/ogy, 11-12. pp. 133-JJR. l3oao,\NOVIC, P., MARKOVIC, V,. DR,\GIC, D., RAKIC, M . IJAnovJ{:, M .. RAJ(:r:vJt D., l'orovJc, V., & Mn.oJEVIC, LJ. (1973) O.movna geo/oska kart a SFRJ I : JO()()(}(}, list Donji Milanovac Savczni gcoloski zavod, 13cograd. C::AR, J. & SEOEI.A, S (200 I) Kra!:kc narivnega stika apnencc-dolomit pri Prcdjami (Karst charnctcristics or thrust contact limestone-dolomite nenr Prcdjama). ;leta C arsologica, 30. 2, pp. 141-156. J. Calic-Ljubojevic & V. Ljubojevic one of the indicators of a significant karst development. One of the reasons for this certainly lies in the strong influence of allogenic water input, which gives to this area an attribute of contact karst and/or fluviokarst. As this kind of water input is combined \'(ith the favourable tectonic conditions next to the Fault, the great density of cave passages per surface is quite comprehensible. The behavior of the fluvial network in contact with the narrow limestone ridge is varied: some of the streams sink at the con tact, while the larger of them en"trench across the ridge. The dominant direction ofkarst groundwater flow is from the south towards the north, along the limestone belt. Due to the inten sive tectonic activity, groundwater is capable con siderable depths. The presence of warm springs and springs of stable discharge throughout the year, strongly points to deep groundwater circulation and to the presence of a karst aquifer with retention capability. It is important to mention that the limestone belt continues to the other bank of the Danube (in Romania), stretching further in the Mehedinti Plateau. Considering that the major geological conditions in this area are quite similar on both sides ofthe Danube, it be expected that a more detailed comparison would perhaps give answet:s to some of the questions on the evolution of this karst area. v .. ru.JrOVIC, 13., & OIMITRIJI!VIC, N. ( 1988) Pojnvc tcnnalnih voda na padinama Zapt:mici S17'skog Geo/oJkog Drustva ::a 1985-1986, Beograd, pp. 225-229 Fu.JI'OVIC, 13., DJMITI\IJEVJC, N., & V. ( 19R5) /lidrogeoloska studija pojava /erma/nih i termomineralnih voda na teritoriji SO Kladovo. Faculty of Mining and Geology, Beograd. 47 p. (unpublished, in Serbian) GRUBJ(: A. ( 1992) Gcoloski prolil Dzcvrinskog brda u Sl Srbiji. Ann. Geo/. l'enins /Jalk 56, I, pp. 17-35. (Geological section of D:levrin Hill in NE Serbia. in Serhian) LAURITZEN, S-E (200 I) Marble stripe karst of the Scandinavian Caledonides : an end-member in the contact knrst spectrum Acta Carso/ogica, 30, 2, pp. 47-79. RoTAI\ll, t\., Butai\R, A .. & PovARA, I. ( 1995) L'ctudc hydrodynami quc du systcme karstiquc de Motru Sec-IJaia de Araml\ (RouJ:nanic). 111eor. Appl. Karstology, 8, pp. 145-155. received : 8 July 2002 revised: 30 October 2002 accepted: 02 November 2002


Abstract Theoretical and Applied Karstology, 15 (22), pp. 51-59 Morphotectonic analysis in hydrogeological research of karst terrains. A case-study of SW Kucaj Massif, Eastern Serbia Igor Jemcov Radmila Pavlovic, & Zoran Faculty of Mining and Geology, Belgrade University, Djusina 7, 110 0 0 Belgrade, Yugoslavia. The fracture structures are recognized as being among the principal factors of karst development by analyzing data obtained by remote sensing. Regional fractures were recognized through analysis of satellite imagery-scanograms, while the detailed fault pattern was ob tained by stereoscopic analysis of aerial photographs. The method of quantitative geomorphologic analysis was used for the identification of neotcctonically active structures. Neotectonic analysis was performed using a morphometric-statistical procedure the calculation of topographic relief. As a new procedure, supplemental to fault pattern analysis, Digital Elevation Model (DEM) was applied. Comparison of the results obtained by fault pattern analysis using remote sensing, quantitative geomorphologic analysis and digital eleva tion model with favored directions of karst groundwater now, revealed a significant control of the position and function of the faults over karstification. General karst groundwater directions and drainage reorientation arc mainly controlled by the morphotcctonic evolution. Key words: karstification, karst groundwater now, remote sensing, Digital Elevation Model, neotectonic analysis. L'analyse morphotectonique dans Ia recherche hydrogeologique des terrains karstiques. Etude de cas sur le sud-ouest du Massif de Kucaj, Serbie de /'Est Resume En analysantles donnees oble."lues.par des plzolographies satellitaires, on a observe que les structuresfracturales son/ parmi les principauxfacteurs quifavorisentle developpemenl du karst. Lesfractures regionales on/ ete reconnues sur des images satel/itaires, landis que le modele structural de detail a ete etabli par analyse stereoscopique des aeroplzotogrammes. Afin de me/Ire en evidence les structures actives au point de vue neotectonique, on a utilise Ia methode de /'analyse geomorpltologique quantitative. Pour /'analyse neolectonique on a utilise une procedure morphometrique et statistique, en ca/cu/ant/'energie de relief En tan/ que nouvelle procedure complementaire pour /'analyse du modele structural, on a applique le modele digital altitudinal( Digi tal Elevation Modeh-DEM). En comparantles resullals obtenus par /'analyse du modele structural a /'aide des images sqtel/itaires, /'analyse geomorplto/o gique etle modele digital altitudinal avec les directions preferentielles d 'ecou/ement karstique, on a observe que Ia karstification est significativemenl influencee par Ia position etlefonctionnemenl des failles. directions generales d 'ecoulemenl soulerrain e/les captures karsliques son/ conlrolees surtout par /'evolution morplzolec/ouiqlte. Mots-clt!s: karstification, ecoulemenls karsliques, analyse des photographies satel/itaires, modele digital altitudinal, analyse neoleclonique. corresponding author. E-mail address: jemcov@pll.yu 0 2002, Editura Academiei Romanc. All rights reserved.


52 Introduction Remote sensing and quantitative geomorphologic analysis are methods intensely used for hydrogeological investigations. The hydrogeology of South-Western karst massif (Eastern Serbia) has been investigated by complex researches including the analysis of regional fault pattern on satellite images, the analysis of detailed structural fabric on aerial photograph, and neotectonic analysis based on topographic maps. Studied area The karst terrains of SW Eastern Serbia occupy two separate areas (CviJic, 1893), with a total surface of about 320 km2 (Fig. I): the western karst zone -Ravanica zone (70 km2 ) the eastern karst zone-Kucaj zone (250 km2 ) Significant lithostratigraphic differences between these two zones resulted from different paleogeographic evolution. Deposition of carbonate sediments in the western karst zone started during Lower and Middle Triassic, and ceased in the Upper Jurassic. In the eastern zone, the deposition of carbonate sediments commenced in the Upper Jurassic, and was completed at the end of the Lower Cretaceous. In the discharge zone the ground waters are either gravita tional or ascending, mainly depending on the base level of karstification. Six karst springs have been identified, with an average yield exceeding I 00 1/s. According to the results of a relatively small number offlow tracings, fictitious rates of groundwater flow range from 0.006 to 0.045 m/s, resulting in an average rate of 0.025 m/s (STEVANOVIC & DRAGISIC, 1992). They also indicate a multidi rectional circulation, and the occurrence of groundwater piracy between adjoining catchment areas. Remote sensing Besides lithology, the fault pattern is one of the principal factors that control the source, the circulation and the outlet of karst ground water. Consequently, analyses of regional and detailed fault patterns were accomplished by extensive use of remote sensing methods. The main requirement for the application of remote sensing in geology and hydrogeology is the comparative analysis of images obtained using different sensors, in different spectral areas and at different scales of sensing (PAVLOVIC & KRESic, 1990). Data collection for the regional fault pattern was achieved using satellite images (LANDSAT missions 2 and 3, scale I :500 000), analyzed by means of the logical comparative technique. The analysis was accomplished on images in different I. Jemcov eta/. channels (Fig 2). Black & white contrast images were taken in green, red and related infrared spectral areas. The results obtained were correlated with the analytical results of color-composite images at the same scale. Faults, classified by significance in structural fabrics, were divided into two ranks (STEVANOVIC eta/., 1996). The first one includes large features of regional importance"preferential fault lines" (PAVLOVIC, 1990). The other one includes faults oflocal significance, the 1anking criteria being based on the scale ofthe investigated area. When an observed fracture was limited to the investi gated area it was treated as a local one; if it extends outside the area's borders it was considered as having a regional signifi cance. Regional faults generally form two systems. One of them has a NW-SE strike (longitudinal structures), whereas the other is ofNE-SW strike (transverse faults) (Fig. 3). Undoubtedly, the widest application among remote sensing procedures in hydrogeology is the analysis of aerophotographs. Following the stereoscopic analysis of aerial photographs, on a scale of I :50.000, the detailed fracture fabric of an investigated terrain can be determined. In general, the observed faults within a studied area show the same orientation as the regional fractures. of a statistical analysis ofphotogeologically determined faults, shows a maximum NW-SE orientation, similar to the orientation of lithostratigraphic units. Therefore, faults of'NW-SE strike could be defined as longitudinal structures. Submaximum orientation shows faults ofNE-SW orientation; these arc transverse faults. The Digital Elevation Model The Digital Elevation Model (DEM), as a digital form of relief presentation has been used in the earth sciences for many decades. Originally oi1ly used as a convenient resource for visualization, DEMs are now a remarkable source for analysis and interpretation of geological data. In this study, DEM was applied for rupture pattern analysis, and for quantitative geo morphological analysis. The DEM was also used as a helpful resource for 3D visualization ofthe landscape (Fig. 4). Three-dimensional landscape models provide a comprehensive view of geomorphologic features, as well as geological elements. Thereby, an improved and comprehensive view of geomorphologic features, differ ent types of geological data, as well as a visualization of their mutual spatial relations, was achieved. Application of the DEM in fault pattern analysis is based on the presence of a relief-shading model. The concept of fau It pattern analysis rests upon simple following the lineaments with sharp morphologic reflection on surface. The advantage of using a shaded relief model as compared to aerial photos and satellite images is the absence ofvegetation, cloudiness, different ob jects and other anthropic structures. Limitations in the appli cation of the shaded relief model rupture pattern analysis were


Morphotectonic analysis in hydrogeological research of karst LEGEND : c=J1 c:=J 2 0 3. "V 4 5 (j] 6 --<. ..... 8. 0 1 2 3 4 5km ---;'J .. ': ;,: I' Fig. 1 T h e kar s t t e rr ains of SouthW es t e rn Mts. ( Eas t e rn S e rbi a). 1. ka r s t t errai ns; 2 n o n kar s t terrains; 3. cav es: 4. s i nkholes ; 5. kar s t s prin g (Q s r > I 00 1/s); 6 tapp e d karst sprin g (Q s r > I 00 1 /s); 7 l a b e led und erg r o und drain age: 8 predic t ed dra in age. 53 Les t err a ins ka r s t iques du sudo u es t du Massif de K u caj (Se rb ie de I 'Est) I terrains ka r s ti ques; 2. t erra i ns non-karstiques; 3. grolles; 4. d o li nes; 5 s our ce ka r stiq u e (Qsr > 100 /Is); 6 sou rce kars ti q u e captee (Qsr > 100 /Is); 7. d r a in age souter r ain demontre ; 8. drainage so ut e rrain suppose.


54 I. Jemcov et a/. Fig. 2. Map of the regional fault pattern on the satellite image. Key : I. regional faults, verified (solid lines) and supposed (dashed lines); 2 local faults, verified (solid lines) and supposed (dashed lines). Le modele structural regional superpose sur /'image sate/lit a ire. Legende : I. failles regionales, conflrmees (/ignes continues) et suppo sees (/ignes interrompues); 2.failles locales, conflrmees (/ignes continues) et supposees (/ignes interrompues). resolved by parallel analysis of multiple generated models with varying position of the simulating lighting source. Data ac quired as a result of shaded relief model analysis highly correlates with the results ofthe analysis and interpretation of fault patterns on satellite images and aerial photos. By merging satellite imagery with the shaded relief model, the plausibility of analysis has beeri improved substantially, pro viding considerable additional information The OEM as a digital form of relief presentation contains all the essential numeric infol1nation of the terrain (x, y, z), required for morphometric analysis. The application of the methods of quantitative geomorphologic analysis is greatly simplified by 360/0 270 Fig. 3. Orientations ofphoto-geologicaly determined faults. Orientations des failles determinees par des moyens photo geologiques.


Morphotectonic analysis in hydrogeological research of karst 55 Fig. 4. The 3D model of the investigated area. Modele tridimensionnel de Ia ::one etudiee. automated extraction of requisite data from digitally recorded data. Errors which were previously common, such as those ocurring during the retrieval of data from topographic maps, are now completely excluded (MARKOVIC eta/., 1996). Quantitative geomorphologic analysis The fault pattern comprises all fractures, irrespective to their size, genesis, time of origin or movement intensity. Nevertheless, fault chronology (the time of activation and the duration of its activity) directly affect the development of karst processes. Considering their influence over hydrogeological char acteristics, recently active fractures have the highest significance. These are the fractures formed in recent geological history, or pre-Neogene faults that were repeatedly activated during the recent geologic periods. The method of quantitative geomorphologic analysis was used in order to define the youngest tectonic activity, i.e. to determine the neotectonic zone:s, the general movements of tec tonic blocks, as well as to establish the direction and intensity of motion of particular blocks. This is in fact a morphometric statistical procedure, carried out by analysis of topographic maps and DEM. The procedure of topographic relief analysis (PENCK, 1924; MARKOVIC et al., 1996) was applied in the study of the SouthWestern Kucaj karst massif. The topographic relief represents the potential energy defined as a result of height differences within measured equal unit fields. Data obtained by systematic measurement of potential energy allow the construction of a map with contour lines. Contour lines represent anomalies with respect to the refer ence level (mean value of the whole measured data set). The interpretation of map anomalies of topografic relief is simple. In general terms positive values of energy of relief anomalies, i.e. reinforced erosion correspond to the areas of uplifting. Conversely, negative values mean rein forced accumulation by subsiden,ce. of corresponding blocks (Fig. SA). Furthermore, contour lines, their orientation and gradient define neotectonic structures and boundaries between


56 ;?:1').(}1) I)(J =::uc_iJO nQ. 1 6().L)(J \4\J.OO 10t1.

Morpholeclonic analysis in hydrogeological research of karst Influence of morphotectonic evolution on flow directions and redistribution of karst groundwater The recent fault pattern is the result of repeated tectonic move ments, and many of the identified faults are still active. The locations of main karst springs are controlled by regional faults. Large karst springs Crnica spring, Toplik spring, Radovansko spring, St. Petka spring, with average yields exceeding I 00 Vs emerge along zones crossing longitudinal and transverse faults. The directions ofkarst groundwater flow are controlled by the geological structure, karstification level and local hydrogeo logical conditions. They are mostly determined by the evolu tion of karst processes. Tectonics and limestone structure are the principal factors of karstification and groundwater circulation. The first stage, which has the strongest influence on subse quent groundwater circulation, is certainly the uplifting and the formation of karst massifs (western part anticline). Longitudinal faults of NW-SE strike (e.g. Senje-Kcpoljin, Samanjac) have established the general directions of groundwater movement towards the regional base-level. In the following stage of karstification, periclinal inclination ofthe terrain, the disposition of impermeable rocks and trans verse faults of NE-SW orientation controlled the paleo-hy drographic pattern and defined the position of erosion base level. At this stage fluvial processes are dominant. Although initially the massif was built of intensely fissured carbonate sediments, due to the immaturity of the joint systems, the dis solution effect of the water was limited. Most of the water still flowed as surface streams, creating fluvial landscape features. In the subsequent stages of evolution, karstification becomes a dominant process as compared to fluvial erosion. Due to epirogenic movements and faster incision of the main drains ... _, ... Fig. 6. Main orientation of passages in caves from Massif. Orientations preferenlielles des galeries des grolles du Massif de Kucaj. 57 at the rim, the erosion base-level was lowered. Stream flows, weakened by the karstification processes, were not able to down cut further and remain overhanged. Further on, the river network continues to degrade and the groundwater levels were lowering. In this phase, the karstification progresses along the directions of the original paleo-flows, mostly defined by trans verse faults. This assumption is suggested by the fact that cave passages coincide with the transverse faults orientation (NE SW)(Fig. 6); passages oriented along longitudinal faults are almost completely missing, especially in the eastern karst belt. Fluvio-karstic processes continue their local development in the western karst belt, where a narrow area of carbonate de posits is criss-crossed by transverse fractures. This in deep incision of bigger streams (Ravanica, Crnica, Grza). The lowering of karstification processes can be easily traced by following the hypsometry of the cavities; the highest ob jects are hydrogeologically inactive, while the lowest ones are temporarily or perennial active (Ravanicka cave, Nemanja 2, Grza, Sisevac cave, etc.) The recent stage of karstification processes (including the present time) is characterized by the rearrangement of the drain ages between some ofthe adjoining areas, as a result of differ ential block movement along neotectonic zones. The rearrange ment of karst groundwater flows follows the regional erosion base-level, with the abandonment oflocal erosion base-levels and creation of smaller catchment areas. Groundwater piracy is characteristic for this phase (Fig. 7). In the eastern karst belt, the reorientation of the groundwater was determined by dye tracing of the Velika Brezovica ponor. Waters originating in the Grza catchment area were re-di rected towards the Radovansko spring. Re-direction ofthe karst groundwater flow towards the Radovansko spring was a con sequence of the relative lowering of the central block (A) with respect to the side-blocks (B and C). Another typical example of groundwater piracy is the re-direction of the karst ground water flow from the Resavica catchment area towards the Crnica spring as a consequence of intensely rising of the north block (E). In the western karst belt, piracy is not characteristic due to the deep incision of surface streams along transverse faults. Only one case of piracy was detected by dye tracing of Pirivoje ponor (Ravanica catchment area); this is due to a relative low ering of the northern (D) block with respect to the southern one (B). Conclusions From the regional structural analysis, cave orientation and main drainage zones, appears that the position and mutual relation ships oflongitudinal and transverse fault-structures plays an im portant role in the formation ofpreferrential directions ofkarst groundwater movement in different phases of karstification. In


58 I. Jemcov et a/. 3 4 Cfj. 5 7 Fig. 6. General directions of karst groundwater and fault pattern on the shaded relief model. Key: I. regional faults (verified and supposed); 2. local faults (verified and supposed); 3. neotectonically active zone; 4. karst spring; 5 swallet; 6. underground drai nage verified connection; 7 predicted groundwater flow direction. Directions gemJrales de I 'ecoulement karstique et modele structural superposes sur le modele du relief Legende: I. failles regionales (conjirmees et supposees); 2 Failles locales (conjirmees et supposees); 3 ::ones actives au point de vue neotectonique; 4. source karstique; 5. perle; 6. drainage souterrain demonlre; 7. direction supposee souterrain.


Morphotectonic analysis in hydrogeological research of karst the last development phase, neotectonically active fractures are especially important. Comparison of the studied underground water movement directions and neotectonically active zones in the eastern karst belt of SW massif, revealed that longitudinal faults have had a n important role in the es tablishment of the direction of karst drainages, and that the main drainage elements of karst groundwaters are connected to neotectonically active zones-Nemanja spring, Krivi Vir spring, Radovansko spring, Izvor spring and Toplik spring. References CviJic, J. ( 1893) Geographical investigation in Kucaj area in Easter Serbia. Geol. Anal. of Balkan Peninsula, V. Belgrade (in Serbian). KRF.SIC, N. & PAVLOVIC R. (1990) Remote sensing as a method for determining of directions of karst groundwater flow. Nas Krs, 28-29, pp. 21-34. MARKOVIC, M., PAVLOVIC, R., & CuPKOVIC, T. (1996) Quantitative geomorphological analysis; a direction of further development 59 of geomorphology. Collection of the papers of Geographic In stitute "Jovan Cvijic"-Serb. Acad. of Science and Arts, 46, pp. 107-119 (in Serbian). PAVLOVIC, R., & KRESIC, N. ( 1990) Remote sensing in hydrogeological research of eastern Serbia karst. Inti. Symposium of Remote Sens ing and Water Resources, Enshede, Netherlands. PAVLOVIC, R. ( 1990) Complex relief analysis as method of geologi cal investigation. PhD. Thesis, Faculty of Min. and Geology, Belgrade University, Belgrade. (in Serbian). PENCK, W. (1924) Die Morpho/ogishe Analyse. J. Engelhorn. Stuttgart. STEVANOVIC, Z. & DRAGISIC, V. ( 1992) Directions of circulation of karst ground waters in the Carpatho-Balkanides, Serbia. In: Tracer Hy drology(HmzL, 1-1 & WERNER Eds.) Balkema, Rotterdam, pp. 291-295. STEVANOVIC, Z., MARKOVIC, M., PAVLOVIC, R., JEMCOV, 1., DRAGISIC, V., & CuPKOVIt, T. ( 1996) Correlation Between Regional Fault Pattern and Karst water Flow Directions Based on Examples from Eastern Serbia (Yugoslavia). Theor. Appl. Karstology, 9, pp. 149-154. received: 22 July 2000 revised: /7 November 2002 accepted: 29 November 2002


Abstract Theoretical and Applied Karstology,lS (2002). pp. 61-72 Environmental isotopes studies and the hydrogeological model of South Dobrogea (Romania) Augustin Tenu*, Florin Davidescu, Razvan & Lacramioara Coarna National Institute of Meteorology and Hydrology. Sos Bucure$li-Ploie$1i 97, 71552 Bucharest, Romania. Due to its natural a n d anthropic features, So ut h Dobrogca (Romania) is a very interesting area from a hydrogeological point of view. In the region there arc two superposed calcare o us aquifers-the upper aquifer (Sarmatian) and the lower aquifer (Barremian-Jurassic)-which form the so-called "karstic system" of a strategic importance Isotopic ('H. 14C, uC, D, 110) was performed over the last 25 years by an integrated research of all natural types of waters, in order to improve the knowledge of hydrogeological parameters and of the regional patlcm flow. The main isotopic characteristics of meteoric and surface waters a r c briefly described in the p ape r Within the karstic system, our study focused on t h e lower aquifer for which a n up-to-date regional model was carried out by correlating all hydrodynamic and isotopic infom1ation. The m ain isotopic features can be synthesized as follows: the recharge area is located mainly in the Pre-Balkan Platform (Bulgaria); the mnin groundwater flow direction is cast-northeast. towards Lake Siulghiol and the flow velocities, at regional scale, vary from I 00 m/ycar for t he secondary groundwater flow to 500-1:800 m/ycar for the main one. Key words: knrstic aquifers, isotopes, monitoring, conceptual model. Etudes des i sotopes environnementaux et le modele hydrogeologique de Ia Dobrogea du Sud (Roumanie) Resume Grace a ses carac/IJristiques naturelles et antllropiques, Ia Dobrogea du Sud (Roumanie) est une region tres interessante au point de vue hydrogeologique. Dans cette region if y a deux aquifores calcaires superposes I 'aquifore superieur (Sarma lien) et I 'aquifere inferieur (Barremien-Jurassique) qui formenl le 11systeme karstique 11 ayant une importance strategique. Le monitorage isotopique (1ft, IJC, 11C, D. INQ) a ete ejfec/ue au COlli'S des 25 dernieres annees pour lOllS les types d'eaux naturelles, afln d'une meilleure connaissance des parametres hydrogeologiques et d 'elaborer un modele regional de I 'ecoulement souterrain Les prin cipales caracteristiques des eaux de surface el des precipitations y son/ brievement decrites Pour le systeme karstique, notre etude a ete focalisee sur l'aquifore inferieur, pour lequel on a realise un modele fonde sur toutes les informations hydrodynamiques et isotopiques. Les principales caracteristiques isotopiques peuvent etre syntlletisees ainsi : I 'a ire d'afimentation de I 'aquifore est localisee dans Ia Plate-forme Pre-Balkanique (Bulgarie) ; Ia direction principale de l'ecoulement sou terrain est E-NE, vers le lac de Siutghiol et les vitesses d'ecoulement varient entre I 00 mlan pour I 'ecoulement secondaire et 500-1800 mlan pour l'ecoule ment principal. Mots-clt!s: aqui.feres karstiques, isotopes, monitoring, modele corresponding author. E-mail address:

62 Introduction South Dobrogea is a region that hosts a karstic system of stra tegic importance for Romania. Until1973, the classical hydro geological prospection has not been able to yield a definitive picture of the groundwater pattern. A preliminary survey (TENU, 1973) using D/H ratio measure ments provided promising results. Consequently, an isotopic and geochemical regional survey program started. Beginning with 1982, a sampling network (Fig. 1), including not only the groundwater but all the water types of the hydro logical cycle, was maintained unchanged and all environmen tal isotope analyses were performed: tritium (lH), radiocar bon ('4C), carbon 13 (13C), oxygen 18 ('80) and deuterium (D) Additionally, in 1994, 8180 and 815N in N03 -measure ments have been carried out as well. This isotopic database covers about a quarter of a century and, currently includes more than 2500 analyses. 0 SOUTH DOBROGEA Sc11l" in m 10000 20000 I 30000 J37 SM102 0 b e u . JJ8 J40 I ....... -1 I e ... -... I I J43 ..-" 1 ... _.......... I I I I I I I \ \ \ A. Tenu el a/. Regional framework South Dobrogea is located in the southeastern part of Roma nia, on the Black Sea shore, at about 44 northern latitude and 28 eastern longitude. This region covers a surface of about 4800 km2 and its altitude is generally below 100 m; it is char acterized by an annual mean temperature of 11.2 oc and a variable amount of rainfall (380-500 mm/year). The perma nent streams are lacking and the runoff is the scarcest of the country: 0.2-0.6 1/s per km2 (6-19 mm/year). Several lakes are situated along the Danube River or the 'stack Sea shore. The karstic system of South Dobrogea is formed by two su perposed aquifers, separated by a discontinuous acviclude: the upper aquifer, is developed within a calcareous Sar matian (Late Miocene) plate which is over 10m-thick especially in the southeastern part of the region; the lower aquifer--:-very important from the economic point of view-underlies most of South Dobrogea, in a unitary, karstified, limestone and dolomitic complex of Jurassic and Barremian (Early Cretaceous) age. J124 J39 54 J45 Lake 'I J44 SM24 J116 J110 '" SM20 J115 SM19 ... -... J123 J47 e> ', Negru Voda .... ....... ... .._. ____ ..... Fig I The geochemical sampling network. Key: Triangles-Surface water; Squares-Upper aquifer; Circles-Lower aquifer. u n!seau d'ec/wntillonnage geoclzimique. Ligende: triangleseaux de sulji1ce; carres-aquifere superieur; cerc/es-aqu(f'ere inferieur.


Hydrogeology of South Scaloln m \ \ ___ ,.-------' _,.\ _, ____ ..,. I b e . Fig. 2. Structural map of South Dobrogea. Carte structurale de Ia Dobrogea du Sud. The thickness of the lower aquifer complex is about I 000 min the southwestern part of South Dobrogea but decreases to about 400 m towards the east and especially in the northeast (Siutghiol Lake area). This complex descends from the South towards the North and Nm1heast and it is affected by two main fault systems NNESSW and WNW-ESE oriented, respectively (Fig. 2). The last fault system includes the great overthrust fault Capidava Ovidiu that represents the northeastern geological boundary of South Dobrogea; this is the youngest fault and has a major hydrogeological significance. Experin1ental procedures and isotopic expression units In order to measure the stable isotopes oxygen-18 ('80), deuterium (D or 2H) and carbon-13 (13C) the water samples were treated as follows: 180 isotopic analyses were carried out by equilibrat ing C02 with the water at 25 C; deuterium isotopic measurements were made by re duction of water to hydrogen by metallic uranium at 600 C; for the carbon isotopic ratios the barium carbonate was treated with phosphoric acid and the carbonate samples were previously roasted at about 450 oc in a helium steam: The oxygen, hydrogen and carbon isotopic compositions are given in the 8 notation in %o vs. SMOW (Standard Mean Ocean Water) for 8180 and 8D, and in %o vs. PDB-1 (Pee-Dee-Be lemnite) standard for 813C. VG Micromass 602 C double collector and Varian Mat 250 mass spectrometers were used for the isotopic measurements. The accuracy of oxygen and carbon isotopic analyses is 0.1 %o and 1 %o for the hydrogen ones. The measurements of tritium (T or 3H) in natural waters im ply a previous enriclunent teclmique; in our laboratory, the fractionated electrolytic process was used. Counting of the 3H activities was achieved by the liquid scintillation teclmique; we successively used a Packard 3320 and a Beckman 5801 spectrometer. The tritium concentrations are expressed as "tritium units" (1 TU = o-:'11983 Bqll) cmTesponding to a concentration of one tritium atom per lO'K atoms of hydrogen.


64 Using these preparation and measurement techniques, a minimum 5 TU detection limit and a % globaf accuracy are obtained in our laboratory; for the sample of 1996 a detection limit below 1 TU was reached atThonon laboratory (France). The radiocarbon measurements included the precipitation as BaC03 of carbonate and bicarbonate dissolved in about 200 I of water. The BaC03 obtained was converted into benzene according to the following reactive steps: f.i(600C) Perlcatal The benzene obtained was counted on the same liquid scintillation spectrometer like the one used for tritium activities The radiocarbon activity is expressed as pMC (percent of modem carbon) and the significance limit is usually 2 pMC. The ac tivity of 1 pMC represents the activity equal to 95% of that of NBS (National Bureau of Standards) oxalic acid (I pMC = 13.585 dpm/g C,0,.1 ) Isotopic Analyses Results An integrative insight over the entire isotopic distribution con tents for all water types at regional scale as average values is presented in Table I One may notice the good individualiza tion of each water type M eteoric and surface waters Meteoric water was sampled at Constanta meteorological station during the period 1980-1989 Tritium was measured within this entire interval. Figure 3 shows the evolution of .11-1 concentration as monthly values (inset) and annual weighted averages; the annual values of Vienna and Bucharest stations were added only for an easy Table I Comparative average isotopic composition of the main water types in South Dobrogea: May 1986. Composition iso topiqu e moyenne compa rative des principawc typ e s d 'emt:-c dans Ia Dobrogea du Sud en mai /986. JH oD 0180 J4c 813C Water type N (TU) {96o} {96o} (pMC) {96o} Meteoric water I 31 -59. 1 -9.0 Danube River 3 38 -64.8 -10. 1 Black Sea 2 28 -23.0 3 6 Upper aquifer 12 15 -{)6.1 -10.0 43.2 -10. 6 Lower aguifer 27 <5 -78. 0 -11.1 18.4 -7. 8 number of sampled points. A. Tenu eta/. comparison. At Constanta station a normal annual distribution of concentrations according to seasonality may be noticed; also, for all stations there is a decreasing trend and a very good correlation to the experimental curves, which may be a conse quence ofthe same continental conditions. For stable isotopes, D and 180, 40 monthly samples were col_lected between 1983 and 1987. The regression line (Fig.'4), defined by the equation OD = 7.12 X 018Q + 1.98 is Very similar to the meteoric waters line (MWL) (R2 = 0.92) Also, we calculated the useful input-function as follows: oD = -74.696o and 8180 = -1196o. SuJ:facc waters include different natural water bodies (Danube River, Black Sea, saltand fresh-water lakes) as well as a re gional network of irrigation or channels. Since these waters partially constitute the recharge, they were systematically sampled, the same annual frequency as the groundwater Table 2 includes the results for 1974, 1986 and 1996, and Figure 5 shows the oD-5180 diagram for the results obtained in 1986. It may be noticed that the stable iso tope concentrations cover, at a regional scale and at the same moment, a great range of values (8110 = -2.8 to -I 0 9 96o and oD = -68.8 96o) reflecting the different water types. Surprisingly, for the same water body, during the 1974-1996 period, only small changes were recorded, the most important being the progressive enrichment of Lake Siutghiol that we consider to be a consequence of the draught installed over the last 30 years. The relationship for all surface waters (oD = 6.5 x &180 4.5) is very similar to the MWL (R2 = 0.92), but our experimental points arc grouped into two areas: one for the waters from the Danube and all the channels and another, with less negative values, for the waters from littoral lakes and the Black Sea Karstic The karstic waters, especially the lower aquifer, have been the targets of our isotopic research in the last 25 years. From the entire period, we selected three moments : 1974, 1986, and 1996, as being representative from the hydrogeological point of view and better illustrated by isotopic analyses The upper aquifer The sampling of the upper aquifer began only in 1985 in a penn anent network, consisting oftwelve water-sampling points (boreholes wells; and sprin gs). Table 3 shows the results for the moments 1986 and 1996. The zonal distributions of8180 and of radiocarbon are in agree ment with the regional piezometric pattern (Fig 6) showing gradients towards the Black Sea. Figure 5 shows the correlation 80-5110 for the upper aquifer in 1986 and its equation: oD = 9.6 X 8110 + 18.9; Rb 0.4. It may be noticed that the contents of stable isotopes are placed,


Hydrogeology of South Dobrogea 10000 r-.--r l ; !,; i '1 ; '! i!' < f : : ';' :'' '------j-1 A l. +. . ,.-, .o::-,, ..., __ .,.. __,.,,.-. .. --:--. 1 In_ ; 181!'1 ,,. ,._,_ 1: .IICI._,. IIIIlS ltM 11M, ; ... 1iM > 1lif ., .' 1 I '' ... .. ' il" \ .. i(!'. : \.. : 100 ---i:,. I i I I i I ! i ! g ! ... -6 VIENA -Q-BUCUREl?TI -+-CONSTANTA Fig. 3. Annual averages of monthly weighted tritium contents of met e oric waters in Vienn a, Bucharest a nd Const a nta (Dobroge a, Romania) in the 1960-1996 interval (Inset: evolution of monthly values in Constanta. 1980-1989). 65 Evolution des valeurs moyennes annuelles ponderees en tritium des ema de precipitation co/lectees a Vienne Bucarest et Constan{a dans I 'intervalle /960-1996 (En nu!daillon : evollllion des valeurs mensuelles a Constan{a, !980-/989) -14 -12 -10 -8 -6 -2 0 0 10 -20 30 -40 -J. 50 Cl c.o -60 -70 -80 -90 -100 Fig. 4 oD-S1RQ diagram for the monthly meteori c Constanta ( 1983 1987). Diagramme de correlation oDo1RQ pour les eaux meteoriques mensuelles a Ia station de Constan{a (1983-/987)


66 A Tenu el a/. Table Isotopic results for the surface waters. Re.wltats des determinations isotopiques les emLT de surface. 1974 1?86 1996 Sampling point T 5180 so T li180 so T s"o SD (TU) (%c.) (%c.) (TU) (%c.) (%c.) (TU) (%c.) (%c.) S I /Danube at Ccmavodl1 112.3 -10.8 -59.8 41 -10.1 -68.8 14 -10.9 -58.2 S2/Danubc-nlack Sea Channel at Medgidia 98.7 35 -10.1 -68.0 13 -10. 7 -61.1 SJ/Lakc Siutghiol at Mamaia Resort 90.1 -5.9 -43.3 28 -4.3 -37.8 1 7 -4.1 -29.6 S4/Lakc Plopeni nt 30 -3.7 -34.0 II -7.3 S5/Dnnubc-Biack Sea Channel at Basarabi 24 -9.0 -64.8 18 -10.4 S6/Biack Sea at Mamaia Resort 43.9 -5.4 -31.8 28 -4.1 -26.1 -4.3 S7/Biack Sen at Eforic Nord Resort 35.5 -4.1 -21.6 27 -3. 1 -20.0 17 -2.8 S!!/Lake Tcchirghiol at wharf 24 -3.2 -23 .6 14 -14 -12 -10 -6 -4 -2 .-----------.------------,------------.-----0 -' ---::-,:' OJ ........ 0 .. MWL ,._.... -------... -........ . 0 ...... .. .... ,.. :: .... ;" .. .,._--::-:=--.... -r.. .. :....... . .. _ ... S ur fac waters ...... _ .. --20 -40 -60 -80 Upper aquifer 0 [ ---------------____ ___J 0 -100 -120 Pig. 5.1>D-Ii1"0 diagram for the surface waters and groundwaters in 1986. Key: I Dia1110nds: Surface waters: I) f)= 6.5 "81"0-4.5, R2 = 0.96; 2. Sl:unrcs: the upper aquifer: liD= 9.6 "1>180 + 18.9, R1 = 0.40; 3. Triangles: the lower aquifer: SO= 7.9" li1"0 + 9.9, 1?2 = 0.93; M\VL: Meteoric Water Line. Diagramme de correlation I\ f) -1\'NO pour /t?s eaux de swface et les c,mx souterraines, en 1986. Legende: I Eau:x de surface: 'f>D =-6.5 "li"O -4.5, R1 = 0. 96 ; 2. Aquifore superieur: liD = 9.6 x S'O + 18.9. R1 = 0.40; 3. Aquifore in.forielll': liD = 7. 9 I\1NO + 9 .9, = 0. 93; Mil' I.: droite generate de correlation des precipitations ,....... *-.._., 0 c.o


Hydrogeology of South Dobrogea 67 Table 3. Isotopic results for the upper aquifer. Resultats des determinations isotopiques pour l'aquifore superieur. 1986 1996 No Sampling point T 14c 5JJC 5180 5D T J4c 513C 5180 5D (TU) (pMC) (%o) (%o) (%o) (TU) (pMC) (%o) (%o) (%o) SM 18ffechirghioi-P II 8 15.7 -9.3 -10.2 -73. 8 <2.4 -10.2 -97. 7 2 SM 19/Amzacea-P2 <5 52.4 -9.2 -10. 2 -75.2 6.0 -9. 9 -66.5 3 SM20/Credinta-P I 33 85 7 -9.7 -9.5 -60.2 12. 3 79.9 -10 5 -9.4-68.1 4 SM21 <5 19.0 -10 6 -11.2 -82. 9 7.0 -11.4 -75.2 5 CAP 8 30.6 -10.4 -8.8 -74. 2 11.0 -9.9 -66.8 6 SM23ffatlageac-P3 7 17.3 -11.0 -10.1 -92.2 9 0 -10.1 -73.6 7 SM24/Chimogeni-FCAP 10 9 6 -80.4 7 0 -10. 0 -72.3 8 25 30.6 -10.6 9 6 -70.5 12.0 -10 .0 -70.4 9 SM26/Mangalia-P3 -11.4 -99.3 6.0 -10.5 -75.8 10 SM I 0 1/Ccrchezu-well 15. 1 90.7 -12.7 -9.5 II SM I 02/Bllneasa-spring 9 55.7 -8.8 -10. 1 -72.1! 12 SM 103/Movi l i l a wcll 41 81.6 -10. 9 9.9 -66. 5 SOU T H DOBROGEA Scale In m 0 1oooo 2oooo 30000 b e \l -Fig. 6. Piezometric map of the upper aquifer 1986 (beyond the Romanian border, aflcr Puuoo-Boscu eta/., 1997). Carte pidzometrique de l'aquifore superieur en 1986 (au-de/a de Ia frontiere r o umaine, d 'apres PIII.JIX>BoscH eta/., /99 7).


68 with a great dispersion, between those of the Danube and those of the lower aquifer. This dispersion is probably the conse quence of two factors acting especially along the Black Sea littoral: the partially evaporated surface waters and the deeper ascending groundwater. The lower aquifer The study of the environmental isotopes of this aquifer began in 1974 with a few points (TENU eta/., 1975) and has yielded a first regional image of the isotopic distribution and, conse quently, of the pattern flow (Fig. 7). This regional pattern flow was confinned later (Fig. 8), de spite the fact that the piezometric surface has changed during the last 30 years {TENU et a/., 200 I). After 1982, 27 wells were selected as a permanent network for geochemical and isotopic sampling. Table 4 shows the iso topic results obtained at the three mentioned characteristic moments and Table 5 summarizes the same results but only for the three boundary areas of the aquifer. Concerning the tritium, it may be noticed that significant con centrations were measured only for some wells situated next to the Bulgarian border, in the southwestern part of the South Dobrogea named "recharge (SE Ostrov) area" in Table 5 and in the East Medgidia bifurcation zone of the Danube-Black Sea Channel. Besides the information about the placement of the main intake area, these tritium values prove a good isola tion over the entire area against external influences except for the east Medgidia area where there is a minor descending re charge (Fig. 9). A. Tenu et a/. The ranges of stable isotopes D and 180 are, more or less, the same as for other freshwaters in the region but the average val ues are more negative. The equation of their correlation, oD = 7.9 x o180 + 9.9), is practically the same as that ofthe MWL (R2= 0.93) (Figure 5). Tlie zonal distributions of the stable isotopes values were found almost unchanged within the entire research period. This sta bility is proved moreover by the high correlation coefficient of the 15180 values {R2> 0.82) measured during a four years interval, at the same laboratory and locations. The most posi tive concentrations, close to the actual "input function", were registered in the southwestern part of the region, and the most negative ones along the southern Black Sea littoral. The low concentrations measured in 1974 emphasize a con tribution of the deeper groundwater, and therefore the great importance of tectonics in the upward leakage, especially in the Mangalia-Lake Techirghiol area. The progressive increase of this contribution after 1974, empha sized by a decrease in contents, seems to prove that, during the last 25 years, there was a continuous overdraught which has un balanced the natural equilibrium of the two recharge compo nents. For the radiocarbon in Table 4, we preferred its expression as pMC. The variation of measured 14C activities for the same point is generally small excepting some areas (e.g. Lake Mangalia), where the recharge conditions have changed during the last 25 years. Therefore, all isoline maps drawn for the 1974, 1986 and 1996 (Fig. 8) moments show the same main features: a central SW-NE stripe with higher concentrations, limited by two ar eas in the water shows a progressive aging towards Lake Mangalia and Cernavodl\ town. Fig. 7. Zonal distribution of ages for lower aquifer-1974. Distribution des ages 14C pour /'aquifore inferieur en 1974.


Hydrogeology of South Dobrogea SOUTH DOBROGEA Scale in m 0 .... \ 10000 ..... __ ____ 20000 30000 b e u I Fig. 8. Piezometric map of the lower aquifer-1996 Ccirtc pilizomlitrique de /'aquifore inferieur /996. SOUTH DOBROGEA Scale in m 0 10000 20000 30000 b e I Fig 9 14C (pMC) isolinc mnp of I he lower aquifer-1996 C arte des isolign e s (pM C ) pour I aquifore inferieur /996 69


70 A. Tenu et a/. Table 4 Isotopic results for the lower aquifer. Determinations isotopiques pour I 'aquifore injeriew: 1974 1986 1996 No. Sampling point T li13C li"O lin T li1Mo (%.) (TU) (pMC) (%.) (%o) liD T c li13C li1RO (%.) (TU) (pMC) (%.) (%.) liD (%o) (TU) (pMC) (%.) (%.) .J29{fortomanu-F 5044 2 J30/Mcdgidia-F 5091 3 J31/Poarta Alba-F 5036 4 J32/Constanta-P7 5 J34/Constanta-P I 0 6 J3S/Constanta-PI 7 .!36/Mamaia-1' I Mil 8 .!37/0ilina-F 5048 9 J38/Aiimanu-l' Vinalc. I 0 .J391Pc$1era-F 5046 II .!40/Adamclisi-F 5063 12 .141/Garlila-F 13 .!42/[]nncasa-F 5062 14 .!43/Dobromiru Deal-Spring 15 5064 16 .J4S/I'Iopcni-F 5074 17 .J46/Coslinc$1i-F 5068 I 8 .!47/Ncgru Vodn-F 5065 19 .148/Mangalia-F 5082 20 .11 08/l'oarta Alhn-F 5042 21 .J II 0/Ciobanila-1' 5054 22 .1 I IS/G-ml 5066 23 1 116/Cobadin-F 5067 24 ,JJ 22/Rasova-211 25 .1123/Ccrchczu-F 2 26 J 124/Pctrcni F CAP 27 .1125/CaslcluF 5053 -11.6 -68.3 8.2 -6. 7 -12.5 -77.2 10.4 -7.7 -12. 5 -78.8 82.9 -9.2 -11.0 -64.6 54. 8 -7. 7 -11.2 -70.3 4.1 -7.3 -12. 8 -78.2 33. 5 -7. 9 -11.3 -73. 2 <5. 5 7 -7.4 -11.7 -82.6 <0.8 4.6 -7. 0 -12. 2 -82. 9 <5 7 8 -7. 7 -11.3 -79. 7 0.9 6 4 -6. 9 -12. 0 -77.7 <5 13. 8 -7. 7 -10.9 -77.4 -12. 1 -83. 5 II -10.5 -71.6 -10.3 -73.9 <5 -10.7 -74.2 -11.8 -83.5 <5 9 9 -8.1 -11.0 -75. 5 -12.1 <5 9 9 -8.0 -11.2 -77.0 <0.8 10. 9 -8. 2 -12. 2 -86. 7 <5 3 8 -7. 0 -11.0 -75. 8 1.1 4 3 -6. 8 -11.7 <5 11.1 -7.4 -10.2 -65.7 -10.9 -76.8 <5 I 0 6 -8.1 -11.5 -80.5 -88.7 <5 22. 9 -8.9 -11.2 -81.7 1.5 23.2 -8. 8 -11.9 -85.5 II 79. 0 -9.4 -8.8 -63. 3 -10.7 -72.0 <5 17.5 -8. 0 -9.0 -61.3 0 8 17. 0 -8.9 -10.5 -73. 9 7 35. 9 -8. 2 -10.5 -71.6 <5 7 1 -8.8 -10. 8 -75.8

Hydrogttology of South Dobrogtta For the first two stages, 1974 (TENU el a/., 1975) and 1 986 {TENU eta/., 1987), our results-as well as the total radiocarbon data existing in the hydrogeological literature -were expressed as radiometric ages and they have been corrected only for A0 the initiai14C concentration. These "uncorrected" or "partially corrected" values led to an overestimation of the groundwater ages and implicitly to the underestimation of the flow velocities. A "complete 14C cor rection" by evaluating a total retardation factor such' as for the isotopic exchange reactions during the underground movement of the water as well as for the delay caused by the diffusion exchange between the fissures and matrix, was carried out for South Dobrogea by TENU & DAVIDESCu ( 1996). SOUT H DOBROGEA Scale In m 0 1'11000 20080 Key D [fill]l s 4 BQa i _;ii.ti 10 71 The actual velocities obtained in this way: 500 to 1,800 m/year for the central stripe, and only less than I 00 m/year in the Lake Mangalia area, differ up to two orders of magnitude from the values previously obtained. Despite the fact that few hydraulic flow directions were found ("no radiocarbon flow di rection"), these values are in agreement with the velocities hydraulically calculated suggesting that they are close to the ''true values". The regional conceptual model for the. lower aquifer can be synthesized by some main isotopic features as follows (Fig. I 0): recharge area: mainly in the Pre-Balkan Platfonn (Bulgaria); Fig. I 0 Hydrogeological conceptual model of the lower aquifer, updated for 1996. Key: Aquifer type: 1. Shallow groundwater (14C = 2{}-50 pMC. 3H = 0.8 14.3 TU): 2. Partially confined aquifer (locally covered by the Lower Cretaceous acvitard: c = 5-10 pMC, 3H = < 0.8 1 I TU): 3. Confined aquifer covered by the intermediate complex acvitard : c = 1-10 pMC, 3H = < 0.8 TU); Recharge: 4 The front of the main recharge area; 5. Adventive recharge (by surface water) area; Drainage: 6 Natural drainage area {Np below+ 17m); 7 Major (natural and anthropic conditioned) discharge area (Np below+ 2m); Groundwater hydraulics: 8. Flow direction of groundwater, c on tinned by isotopic means and the value of hydraulic gradient; 9. Flow direction of groundwater, suggested by underground hydraulic information but non-confirmed by radiocarbon: I 0 Zero line of the area having the piezometric head above the upper aquifer (up to 15m). Mode le hy d rogeologique conceptuel de /'aqui.fere inferieur. actua/ise pour /'annee 1996 Legende: Types de nappes: / Nappe libre rue= 20-50 pMC, JH = 0,8-/4,3 TU); 2 Nappe partie/lement captive (couverte /ocalement par /'aquitard du Cret ace infori eur: uc= 5-10 pMC, Jfl = < 0,8-/, 1 TU); 3 Nappe captive (couverte par le comple.-

72 recharge water: precipitation, nmoff, small rivers and probably from the Danube; groundwater now: the main direction is east-northeast, towards Lake Siutghiol and the secondary is southeastward, towards Lake Mangalia; other inputs: upward leakage in the Lake Techirghiol -Lake Mangalia area and a minor descending recharge in EastMedgidia area; now velocities: 500-1 ,800 m/year on the main groundwater tlow and up to I 00 m/year on the secondary one. Conclusions The long tern1 isotopic survey of the waters from South Dobro gea allowed the achievement of a comprehensive isotopic da tabase, as well as an integrative insight over all water types and their relationships. Special attention was given to the karstic system, including their hydrodynamic parameters and the assessment of vulnerability, the most important being the lower aquifer for which an improved model was achieved. For the meteoric waters, both the tritium mean monthly con centrations and the D and IRQ concentrations were determined, through multiannual analyses. The isotopic mark was identified at Constanta station. They were used in studying the inter-relations of these waters with the surface waters (especially littoral lakes) or the groundwater in the region. For the surface waters, the environmental isotopes CH. D and IRQ) were analyzed. An important characteristic feature of the region is the great range of values for stable isotopes: 5110 vary from -2.8 to -I 0 .9 %o and 50 from -20.0 to -68.8 %o, a consequence of the very different genetic types of water (the Danube, the Black Sea, lakes). Besides the great utility of these data in establishing the ground-References Puuoo-Boscu, A . MACIIKOVA, M . LorEZ-CIUCANO, M., CALVACIIH, M.L .. DIMITROV, D., CALAFORRA, J.M., VEIJKOV, v .. & Pl!lfiCHEV, P. ( 1997) Hydrogeology of the upper aquifer, Dobrich region, northeastern Bulgaria. Hydrogeology Journal, 5, pp. 75-85. TENU, A. ( 1973) Excmplcs d'application des isotopes du milieu dans l'hydrogcologic roumainc. Meteorology and Hydrology. 2, pp. 21-31. TENU,II.. & DAVIDEscu, F. ( 1996) Some considerations about radio carbon method in carbonate fissured aquifers. Acta Geologica Hrmgarica, (39 Supple mentum) Isotope Works/rap Ill, Budapest, pp. 186-190. A. Tenu eta/. water recharge model, the enrichment in D and 11Q of Lake Siutghiol points out the regional draught that extended the last 30 years. For the upper aquifer, the tests started in 1985 and have been performed up to present-day covering all the range of envi ronmental isotopes. The zonal isotopic distributions are in agreement with the regional piez.ometric pattern, showing gra dients towards the Black Sea. For the lower aquifer, the tritium measurements have proved, that the recharge area of the aquifer is located in the southeast of Qstrov area and, generally, it is well protected from external influences. Both spatial distribution and the correlative relation of the stable isotopes agree with the results of 1H measurements con cerning the recharge area. At the same time, they have pointed out the contribution of the deeper groundwater and therefore the great of tectonics in the upward leakage, espe cially in the MangaliaTechirghiol area. At a regional level, the distribution of stable isotopes values was found almost unchanged during the entire research period. For radiocar6on, the variation of measured 14C activities at the same points was generally small, except for the Lake Mangalia area; here the variations were assigned to both tectonic rea sons and to the continuous local overdraught which has unbalanced the natural equilibrium of the two recharge components. Ali14C-isolines regional maps show same main features: a central SW-NE stripe with higher concentrations, limited by two areas in which the water shows a progressive ageing to wards Lake Mangalia and Medgidia town, respectively. regional model based on the isotope studies can be thus synthesized by the existence of a Pre-Balkan Platform recharge area, a groundwater flow direction mainly oriented towards the Siutghiol Lake and a flow velocity of 500-1 ,800 m/year. TENU, A., NOTo, P., CoRTOCCI, G., & Nun, S. (1975) Environmental isotopic study of the Barremian-Jurassic aquifer in South Dobrogca (Romania) ./. of Hydrology. 26, pp. 185-198. TENU, A., DAVIDESCu, F., & St.AvEScu, A. ( 1987) Recherches isoto piques sur lcs caux des fornmtions calcaires dans Ia Dobroudja Meridionale (Roumanic). Isotopes Techniques in Water Re sources Development, Proc. Intern. Symp., Vienna, pp. 439-443. TENu, A., DAVIDESCU, F., R., & STANr:Scu, G. (2001) Variation des rcssourccs soutcrrainc du Dobrogca du Sud (Roumanic), ciTct conjuguc des causes climatiques ct anthropiqucs. Regional Management of Water Resources, Proc. ofsymp. Sixtlr IAHS. 268, pp. 131-137. received: 5 July 2002 revised: 18 October 2002 accepted: 13 November 2002


Theoretical and Applied Karstology, 15 (2002), pp. 73-91 Biokarst on a tropical carbonate island: Guam, Mariana Islands Danko Taborosi lAboratory ofGeoecology, SchfKJ[ of Environmental Earth Science, Hokkaido University, N-/0, W-5, Sapporo, 06o-080 Hokkaido, Japan. Abstract "Biokarst" refers to erosional and depositional karst features formed by direct biologic action. However, since no distinct karst landform is produced exclusively by biologic action, most features known as biokarst arc actually products of both biotic and abiotic processes operating concurrently in intricate interrelationships. Although biokarstlandforms arc often regarded as features of local interest and limited signifi cance, the effects of living organisms on karst geomorphology arc profound on an eeosystem scale, and arc widespread, diverse and of fundamental importance. This is especially true in tropical and coastal environments, where the biologic influence on karst is so common and intense that it makes the distinction between "biokarst" and "non-biokarst"' thoroughly impracticable. A survey of karst on Guam has confirmed that nearly all of the smalland medium-scale karst features documented on the island arc affected by biota, and bear characteris tics of what has been termed biokarst. Keywords: biokarst, phytokarst, carbonate island, Guam. Biokarst dans une ile tropicale ca/caire: /e Guam, archipel des Mariannes Resume Le terme biokarsl se refire auxformations karsliques d'erosion el de depotformees par action biologique directe. Pour/ant, com me if n 'ex isle aucun phenomene karslique produit exclusivemenl par action biologique, Ia pluparl des formes connues com me 11biokarsl son/ en realite des produils de,, processus biotiques el abiotiques qui agissenl de concert. Bien que lesformations biokarsliques soient considerees sou vent com me ayanl des significations et des inlerels limiles, les e.ffets que les elres vivants on/ mr Ia geomorplrologie du karst a I 'echelle de l'ecosysteme son/ pro fonds: en me me temps, ils son/ divers. generaux. et ont une importance fonda men/ale. Ces considerations son/ valables surto/11 dans les zones tropicales el coliere.t, o1i /'influence biologi que sur le karst est ,fi commune el intense, qu'elfe rend impossible toule distinction entre biokarsl e111non-biokarsl. Un exa men du karst de Guam a conflrme que presque louie., les formes karsliques de dimensions peliles ou moyennes son/ a.ffectees biologiquemenl el qu 'elles gardent/'empreinle de ce qu 'on a appete biokarsl. Mots-cles: biokarst, pllytokarst, ile calcaire, Guam. I n troduction The term "biokarst" refers to erosional and depositional karst features produced by direct biologic action (VILES, 1984). The precursory term "phytokarst" was coined by FOLK eta/. (1973) to refer to extremely jaggedly dissected limestone pinnacles in the Cayman Islands. They recognized that the pinnacles were covered by a coating of filamentous cyanobacteria and proposed corresponding nuthor. -mail address: C 2002, Editum Acndcmici Romftnc. All rights reserved. that algal boring was responsible for the observed geomorphology. The term phytoka.-st was later applied by other re searchers to a variety of dissimilar features in which biologic action was suspected. Buu.. & LAVERTY ( 1982) tried to orga nize all biologically mediated karst features under the term phytokarst, which they redefined as sensu Jato, and grouped them into biolithogenic, destructive (erosional), physical (tec tonic), and C'on..,tructive (depositional) types. The term phyto karst was later replaced by biokarst (SCHNEIDER & TORUNSKI, 1983) to recognize the involvement of organisms other than plants.


74 The organisms whose effects on limestone have studied include bacteria (e.g., SMYCK & DRZAL, 1964; SAND eta/., 1991 ; CASTANIER eta/., 1999), cyanobacteria (e.g., VILES, 1987; MERZ, 1992; SCHNEIDER & LE CAMPION ALSUMARD, 1999), fungi (e.g., DE LA TORRE et a/., 1993 ), algae (e.g., KonLUK & RISK, 1977), lichens (e.g., AcsASO eta/., 1982; CHEN et al., 2000), plants (e.g., HERWITZ, 1993; WALL& WILFORD, 1966), terrestrial and. freshwater invertebrates (e.g., STANTON, 1984; DRYSDALE, 1998), marine invertebrates (e.g., SCHNEIDER & TORUNSKI, 1983; TRUDGILL, 1987; ALBERGARIA MOREIRA, 1996), 'and vertebrates {e.g., HOOPER, 1958; MILLER, 1990; BUTLER, 1995).AsimiJarJy wide range of processes (depositional, diagenetic, and ero sional) and karst phenomena is affected by biota. This includes travertine (e.g., CASANOVA, 198); BAYARI eta/., 1994; DRYSDALE, 1999), calcrete (e.g., KHADKIKAR eta/., 2000; ALONSO eta/., 1998; KLAPI'A, 1979), beachrock (e.g., WEBB et al., 1999), karst breccia (JoNES & KAHLE, 1985), desert stromatolites (e.g., KRIJMBEIN & GIELF., 1979), tufa (e.g., FORD, 1989; PENTECOST, 1985), speleothems (e.g., BARTON eta/., 2001; WENT, 1969; reviewed in FoRTI, 2001), cave wall deposits (e.g., CA!IIVERAS eta/., 200 I), moonmilk and related features (e.g., NoRntUP & LAVOIE, 200 I; JONES, 200); CAUMARTIN & RENAULT, 1958), limestone pavement (e.g., JoNES, 1965; VERRECCHIA, 1990), coastal karrenfelds (e.g., FoLK eta/., 1973), solution runnels (FIOL et a/., 1996), solution basins (e.g., Mos!>S & SMITH, 1993), smooth surfaces and pits (e.g., DANIN, 1983), karst pinnacles (e.g., ZHANG & BAO, 1994), shafts (e.g., TRICART & DASILVA, 1960), soi I pipes (e.g., W AJ..SH & MoRAWJECKA-ZACHARZ, 200 I), etc. Such an incredible diversity of organisms, processes, and landforms involved clearly indicates the high incidence and com plexity of biologic influences on karst geomorphology. This is, perhaps, nowhere as clear as it is in tropical and coastal settings, where nearly all karst features are profoundly affected by biologic processes. The purpose of this paper is to illus trate the widespread influence of biota on karst geomorphology in tropical and coastal environments by providing examples from Guam. A survey of exokarst features on the island has revealed that nearly a !I small-and medium-scale features documented are affected by biologic processes, and bear charac teristics of what has been termed "biokarst". Nonetheless, landforms commonly referred to as biokarst are also profoundly affected by inorganic processes. Biologic action, however intense, is typically just one of the factors involved in shaping a particular geomorphic feature, and organic processes operate concurrently in intricate interrelationships with various inorganic processes to produce any given landfonn. In the most thorough review of the subject, Vu.ES ( 1988) states that no one distinctive landfonn can be related directly to biologic action, and that biologic and inorganic processes interact and occasionally outweigh each other. To support and illustrate this statement by examples from Guam is the second aim of this paper, which reviews various biokarst features and discusses both biologic as well as inorganic processes associated with them. D. TaboroJi Finally, the observations presented contribute new data to Ja:towledge ofthe global distribution of various karst phenomena. Study area Guam, the largest (549 km2) and southernmost of the Mariana Islands (Fig. 1 ), is elongate in shape, 48 km long, and 6-19 km wide. It is divided into two parts by the Pago-Adelup Fault: southern Guam, a rugged volcanic highland with several limestone outliers; and northern Guam, an undulating limestone plateau fringed by precipitous coastal cliffs, locally fronted by a narrow coastal plain. In northern Guam, Late Miocene vol canic basement units are overlain by detrital Mio-Piiocene Barrigada Limestone. It extends to the surface in the interior of northern Guam, but elsewhere grades laterally and upwards into the Plio-Pleistocene Mariana Limestone, a reef and lagoonal deposit dominating the northern plateau and east coast of southern Guam. In southern Guam, volcanic units dominate much ofthe terrain, and are locally oyerlain by Miocene Bonya and Alifan limestone outliers. The principal geologic reference for Guam is still the USGS report by TRACEY et a/. ( 1964), which includes a I :50 000 scale geologic map. Karst featClres of Guam can be grouped into three broad cat egories: surface features (the epikarst, closed contour depres sions, and features related to surface flow), caves (pit caves, stream caves, and flank margin caves), and the discharge fea tures (coastal and submarine springs and seeps). In northern Guam, the karst is typical of carbonate island karst (MYLROIE & CAREW, 1997), whereas the karst ofthe southern highlands resembles that of the tropical continental settings. A com pre. hensive inventory of the karst features of Guam was carried out by TADOROSJ (2000). MYt.Rom et al. (2001) amended this work by describing the karst geology of Guam in tern1s of the Carbonate Island Karst Model (CIKM), a general model of carbonate island karst (M YJ.ROJE & V ACJ IER, 1999). The climate of Guam is tropical wet/dry, with an equable mean annual of27 C. January through May is the dry season, broken by occasional showers. The wet season lasts from July to November, during which heavy rains and tropical storms are common. The mean annual precipitation averages 215-250 em/year (LANDER, 1994). Northern Guam has noperennial streams and its vegetation is dominated by a thick shrub jungle in the undeveloped portions of the plateau and limestone forests in the coastaf plains. The rugged uplifted volcanic highlands of southern Guam are dominated by deeply incised river valleys and grassland vegetation, with shrub vegetation associated with limestone outliers. Biokarst Features on Guam Karst features known to be biologically produced, mediated, or influenced encompass nearly all of the small-and medium scale exokarst features on Guam. They occur in a wide range of environments, including coastal (reefs, intertidal benches


Biolwrst on a tropical carbonate island Pacific Ocean 0 Limestone Units: Mariana Limestone 75 Argillaceous Mariana Limestone r: Barrlgada Limestone llil Alilan Limestone -Sonya and Maemong Linestones -Janum Limestone 2 z (..) Rota f., tJ G ua m \ 0 50 100 Kllorneters N A Locations mentioned: 1 Pago-Adelup Faull 2Agana 3 Tumon 4 Amant&s Point 5 Tanguisson 6 Haputo Such 6 Uruno 9Aitidan 10 MerQa9an Poinl 11 12 Pati Poinl 13 Lane Point 14 lalac Po i nr 15 Mati Point 16 Pagat 17 Sasajyan 18 Taogam Point 19 Pago 20 Togcha 21 Asiga 22 lnaraj a n 23 Cocos 24 Orote 0\aml'l 25 luminao 26 Fl'na Fig. I Locat ion or Guam. with a simplified g eolog i c map showing locales mentioned in text. 5 "ituati o n du Guam et cart e g eologique simplifiee, ave c l e s localiti s cities dans l e texte and t errace s, marine notches, beachrock, etc.}, epikarstic (karr en. solution pans and pits, soil pipes, etc.}, and hypogean (caves and cave deposits) (Fig. 2) The following discussion includes several landforms that are not intrinsically karst, such a s coral and algal reefs and beachrock deposits, but which become thoroughly karstitied and form an integral part of karst systems on carbonate islands. Although not karst features per se, the y may be considered biokarst, as the latter is not limited to disolutional features but also includes depositional and biolithogen i c structures . Re efs Coral and algal reefs are not karst features but bioconstructed d eposits. N e vertheless, they have been r eferred to as examples o f depositional biokarst (Vn...ES, 1984) and are included in this discussion because of the important role they play on tropical carbonate islands. They are not only found on most such islands, but also represent their principal builders. The entire northern half of Guam is an uplifted Plio-Pleistocene reef, whereas the southern half contains several smaller uplifted Plio-Pleistocene and Miocene reefs. Modem fringing reefs almost entirely surround Guam, being absent only in those parts of the northeastern coast dominated by sheer cliffs. The reefs range in width from narrow benches spanning a few meters across, to broad reef flats that are up to 800 m wide. The largest have developed within the coastal embayments of Agana, Tumon, Tanguisson, Tarague, Pago, and Togcha. In addition to fringing reefs, there are two barrier reefs on Guam: Cocos, which fully encloses a large lagoon (7 .5 km2); and Luminao, which partially encloses a smaller one. The geomorphology of modem reefs on Guam has been dissussed by EASTON e t a/. (1978), RAN DALL & HOLLOMAN ( 1974), and RANDALL & SIEGRIST (I 988). RIDING (2002) has provided a review and classification of organic reefs in general.


76 D. Ta.borosi ._,:! . .. '.;:: ., :. s Fig. 2. Bioconstructed landforms and biologically produced, or influenced karst features on Guam: a. fringing reef: b. intertidal bench; c. rimmed terraces; d. marine notches; e. champignon surfaces and karrenfelds; f. karren; g. bioerosional karren; h. solution pans; i. case hardening on clifTfaces; j. calcrete; k. solution pits; I. soil pipes; m. root holes; n. tufaceous stalactites; o. "directed phytokarst"; p. moonmilk; r, s. oxidation of organic materials accumulated at the freshwater lens pycnoclines enhances dissolution; not shown arc beachrock, root grooves, rhizoliths, and freshwater tufa.. Formes de refiefbioconstruiles etformes karstiques produites ou influencees biologiquement de Guam: a. recif; b. pfage intertidafe; c tenasses en gours>!; d. encoches marines; e. surfaces <

/liokarst on a tropical carbonate island Fig. 3. Rimmed terraces on the intertidal bench in lnarajan Terrasses <

78 Fig. 4. Marine notches in the coastal cliff at Taogam Point. Encoches marines creusees dans /afa/aise de Taogam Point solution pits, separated by ridges and sharp tips in the nodal points On Guam, they exist in a variety of fonns, from the most extremely jagged pits and points and even completely penetrating holes in the coastal areas (Fig. Sa), to somewhat more rounded inland fonns having a more subdued reliefbut similar overall appearance (Fig. Sb). There is variation in color as well: the most extremely eroded coastal surfaces are black or dark gray, whereas those inland are light gray to white. The intensity of color is thought to reflect the amount of organic coating by endolithic and epilithic organisms (JONES, 1989) Perhaps the most remarkable feature of this type ofkarren is the apparent continuum offonn across a range of scales from millimeters to meters. In an almost fractal-like pattern, the mmscale and em-scale points and pits appear to be repeated on a larger scale and cover extensive areas, creating extremely rough karrenfelds of amazingly jagged pinnacles and irregular pits among them (Fig. 6). Despite being the predominant karren assemblage on young reef limestones there is no precise tenninology associated with this morphology. Its most intensely corroded variant is reported from ml}ny tropical I imestone coasts and is commonly referred D. Taborosi to as phytokarst (FoLK eta/., 1973). Other tenns used include champignon surface (STODDART et a/., 1971 ), lacework mor phology and sponge-work morphology (BuLL & LAVERTY, 1982), and coastal karren (MYLROIE & CAREW, 1995). All of these are more appropriate than the tenn phytokarst, since they do not invoke a specific genetic mechanism. Considered one of.the most variable and least understood karren types, its ap pearance is thought to be largely influenced by endolithic and epilithic organisms (JoNES, 1989), but their impact on the to pography has never been convincingly demonstrated (VILEs, 2001 ). Nevertheless, the tenn "phytokarst" is used for both em-scale sculpturing, as well as the similarly jagged m-scale karrenfeld landscapes. Despite the semblance of form across a range of scales, the processes involved in shaping these features are thought to be scale-dependent (VILEs, 2001), albeit poorly known. Whereas the impact of endolithic microorganisms on rock morphology may be considerable at small scales, the large scale pinnacle morphology is probably the result of differential erosion due to meter-scale heterogeneities, such as variations in mineralogy and cementation (TRUOOILL, 1976) and structural weaknesses (VILES & SPENCER, 1986). In addi tion to biologic erosion and dissolution, the mechanisms con tributing to geomorphology probably include physico-chemi cal dissolution, wave action, wetting and drying, and salt weath ering and hydration (FoRD & WILLIAMS, 1989). The relative contribution of each process to the overall morphology most likely varies with local conditions. Observations on Guam indicate that this type of topography, al though most characteristic at the coast, is also found in inland areas, with some minor differences. On the em-scale, the inland features are less extremely sculpted; their points are less sharp; pits are shallower; completely penetratiQg holes are less common; and the pinnacles commonly tenninate in jagged knife edge forms, instead of the sharp points typical to the coast. Rudimentary rillenkarren are present in inland areas, indicat ing an increased relative contribution of gravity flow to the overall fonn. On the meter-scale, the inland topography is generally reduced; the pinnacles are lower ; the pits among them are partially filled with soil and support vegetation; and the shallow flat-bottomed solution pans, common among the coastal pinnacles, are not found. Despite these differences, the continuum between coastal and inland pinnacle fonns is quite clear, and their overall appearance is very similar. Karren Karren features on Guam belong to two broad categories: those found on young reef limestones, and those found on diage netically mature re-crystallized limestones The former are highly irregular and jagged, and are comprised of densely packed solution pits and points. Showing little no hydrody namic control, this type of karren appears to be a subdued variety of the previously discussed champignon surfaces (STODDART et a/., 1971 ); therefore, biologic activity is prob ably highly significant to its development. Conversely, the karren that developed in the massive, re-crystallized limestone


Biolwrsl 011 tropical carbonate island Fig .5. lrrcguhir and jagged karrcn features: a) extremely dissolved coastal type, north ofHaputo B . :ach; b) inland typ.:, Double l.apies irreguliers et ruiniformes: a type cot i er fortement dissoru, nord de Haputo Beach; b type d 'interieur. Double! Reef Fig. 6. Pit nnd pinnacle karrcnfeld on the coasi s-:uth (note pt:rion tor scale). Lapi.u 'il pinacles sur Ia cote sud de Haputo Beach (a noter le personn::zge pour /'ec!relle). 79


80 outcrops in southern Guam are more regular; and although polygenetic forms are present, the dominant features are solution runnels and other gravity controlled karren. However, even such classic karst features as solution runnels, generally thought to form by chemical dissolutional processes, have recently been associated, on a fundamental level, with biocorrosion by algae (FIOL et a/., 1996). Bioerosional karren Karren features found in the subtidal, intertidal, and supratidal zones of limestone coasts are distinct from the rainfall solution karren of inland sites, as well as the irregular champignon surfaces or phytokarst ofFoU< eta/. (1973). These features are produced largely by bioerosion (typically grazing or boring by marine invertebrates) and occur in an astonishing variety, matching D. Taborosi the diversity of marine bioeroders. In some cases, the grazing scars and boreholes of individual organisms are recognizable (SCHNEIDER & ToRUNSKI. 1983). On Guam, the coastal microrelief is extremely variable and chaotic, with various bioerosional scars overprinting each other. Some basic types can be recognized, and roughly correspond to supratidal, intertidal, and subtidal zones. In certain sheltered coastal sites, such as the walls of sea caves, unusual bioerosional scars similar to Lithophaga boreholes are found conspicuously oriented along consistent axes (Fig. 7). Solution pans Solution pans (Fig. 8) are small, flat-bottomed basins found in karst areas worldwide. They are influenced by inorganic processes such as wetting and drying and salt crystallization, but Fig. 7. Oriented bioerosional karren in the walls of a coastal cave in Tin ian, an island north of Guam (flashlight tor scale is 12 em long). Lapies de bio-erosion sur les parois d'une grotte cotiere a Tin ian, ile situee au nord du Guam (Ia /anterne qui sert d 'echel/e a une longueur de 12 em). Fig. 8. Coastal solution pan in Sasajyan. Note the intersecting joints marked by arrows. Kamenilza a Sasajyan A noter /'inter section des fissures indiquees par /es /leches.


Biokarst on a tropical carbonate island the crucial factor in their development is thought to be dissolution by stagnant water, made corrosive by C02 from accumulated plant debris (WHITE, 1988). More direct biologic action, such as mechanical removal of rock fragments by lichens, has also been suggested (Moses & SMmt, 1993) Solution pans typically form on rock pavements and lightly vegetated or bare rock areas (FoRD & WJWAMs, 1989). On Guam they are only found nestled among the coastal karst pinnacles, and are curiously absent in the rest of the island. The solution pans examined on Guam have flat and smooth floors, indicating the absence ofbioerosion and abrasion. Many of them exhibit structural control and commonly develop at the intersections of joints or along a single joint, in which case they are elongate. They have overhanging margins and are rarely associated with overflow, or either inlet or outlet channels. If active, they may contain accumulated rainfall, evaporating seawater of various chlorinities, or sea salt crystals. Whereas the deepening some supratidal pools is thought to be a direct result of differential bioerosion (SCHNEIDER & ToRUNSKJ, 1983), the fact that their remarkably flat floors are covered by pervasive organic veneers seems to indicate that biologically mediated dissolution is instrumental. Although wamt stagnant seawater is incapable of dissolving limestone, it is conceivable that there is a thin layer of undersaturated water, which is in contact with both the organic layer and the bedrock. Beachrock Beachrock is lithified beach sand dipping gently toward the sea (BERNIER & DALONOEVIu.E, 1996). It is most commonly found in the intertidal zone of tropical and subtropical coasts. On Guam, it occurs locally on ail beaches along the northwest and north coast of the island, the most extensive deposits being located at Tanguisson, Uruno, and Ritidian. The beachrock typically dips seaward, following the unlithified beach surface, and its composition reflects that of the surrounding cal careous sand. On some beaches, such as Tanguisson, the beachrock is conglomeratic and contains coral rubble and other fossils surrounded by a matrix of sand-sized calcareous grains. The surface texture of the beachrock on Guam is highly variable, and ranges from flat, smooth algae-covered surfaces to the rough,jagged surfaces pockmarked by characteristic karren, presumably reflecting the relative age of the deposits. A number of hypotheses have been set forth to explain the precipitation ofbeachrock-forming cements. They involve both physiochemical processes, such as the evaporation of seawater, the mixing of freshwater and seawater {MooRE, 1973), or C02 de-gassing due to groundwater discharge, wave agitation, and/or increasing temperature (GJSCHLER & l..oMANoo, 1997); as well as biologic processes, such as the uptake of photosynthetic C02 by intertidal microflora (MERz, 1992) and/or the increases in pH associated with heterotrophic microbes (WEBD e t a/ 1999). An interplay of concurrent inorganic and biologic precipitation was suggested by KRUMBEJN (1979). 81 Case-hardened surfaces and calcrete Biologic influences on the deposition and consolidation of limestone have been recognized in the formation of case-hardened surfaces, calcretes, tufas, travertines, oncolites, stroma tolites, and speleothems (VILES, I 988). The exact role of organisms, however, is often insufficiently known, but most'likely varies from insignificant to dominant. Case-hardening, first recognized in the Caribbean, is the phenomenon of strengthening rocks via vadose diagenesis, and is thought to be assisted by blue-green algae, bacteria, and other organisms. VILES (1988) found case-hardened surfaces on Aldabra Atoll coinciding with the presence of endolithic, epilithic, and chasmolithic microflora. Case-hardening is an important phenomenon on emergent coral reefs; where primary pores within the coral-algal framework become in tilled by precipitated calcite, thus making the rock in the vadose zone less permeable. On Guam, such infilling of primary voids by calcite is easily observed in quarry walls and road cuts where the exposed surfaces are commonly case-hardened. This phe. nomenon has also been noticed in thin sections from drilling cores (JENSON & SIEGRIST, 1994). Calcrete occurs in areas where evaporation exceeds precipitation. It involves the chemical precipitation ofCaC03 binding soil, alluvium, or weathered rock by carbonate-rich waters (G6UDJE, 1983); however, biologic action can also be instrumental (KLAPPA, 1979; VERRECCHJA. 1990) On Guam, calcrete can be found locally inside flat-floored sinkholes. Such sinkholes commonly act as sediment traps, and their flat floors can indicate a thick layer of sediment infilling (WHITE, 1988) Analysis of rock samples from a sinkhole in Tarague area, northern Guam, revealed them to be calcrete floatstones con taining coral and mollusc fragments, cemented in a calcified paleosol matrix. Solution pits Solution pits are round-bottomed cavities, normally up to a meter in diameter. They are thought to be one of the most widespread forms ofkarren (FORD & WtwAMS, 1989). Biologic activity is commonly associated with pit development, particularly on a small scale. DANIN (1983) suggested that em-scale hemispherical pits in limestone are directly caused by euendolithic cyanobacteria. Comparable features coated by a black organic layer, presumably cyanobacteria, have been documented on walls in the twilight zone of coastal caves on Guam. On a larger scale, biologic processes are generally not thought to be instrumental in the development of solution pits, but they are known to accelerate the pit deepening. Organic processes in the soil, accumulating in or surrounding the pits, can make deepening extremely rapid. SWEETING ( 1966) reported the dissolution ofJ-5 em deep pits within telf'years by water enriched by organic acids. LoNGMAN & BROWNLEE ( 1980) suggested that vegetation plays an important role in the development of pits and pinnacles in Palawan (Philippines), and indicated that the decay of organic matter


82 accumulated in depressions as small as trittkarren facilitates the dissolution process, producing pits and even sinkholes. Plant root action has also been suggested as a contributing factor to the development of pits in limestone (TRICART & DASILVA, 1960). On Guam, solution pits are found in all exposed limestones. They are somewhat rare in finer grained units such as fora miniferal limestones, and are most common in reeflimestones, whose heterogeneity is probably favorable to their initiation. The pits are found in the highest densities within the jagged pinnacle karrenfelds and limestone forests. In limestone forests, the pits are partially tilled with soil and decaying organic materials, enabling them to support dense vegetation, including large trees commonly rooted inside them. Pig. 9 Root grooves on the surface of a collapsed boulder, Tarague. C ann e lures d e racines sur Ia surface d'rm bloc, a Tarague. D. Taborosi Soil pipes Soil pipes are subrosional soil or sediment-filled cylindrical pits in the epikarst. They are a common feature on Guam, usually up to 0.5 m wide and 3-4 m deep, and can be observed in high densities in quarry walls and road cuts. Soil pipe casts of lithified paleosol are also common on the island. Soil pipes develop by the dissolution of limestone below the soil cover and are thought to form along structurally deter mined zones. There is evidence, however, that the initiation of soil piping is determined by chemistry and the variable per meability of rocks, rather than by structural controls (WALSH & MORAWIECKAZACHARZ, 2001), and this appears to be the case with very heterogeneous rocks such as the young reef limestone of carbonate islands. Organic deposits, such as guano, also promote soil pipe development in raised reeflime stone (JENNINGS, I 985). Following the initiation of soil pipes, Fig 10. Cave entrance concealed by numerous banyan tree roots, Nimitz Hill. Inset: root holes close-up (lens cap is 55 mm in diameter). Entree d une groue ,;,asquee par de nombreuses racines d e banians a Nimitz Hill En bas : detail des trorts de racines (le couvercle de l'obje ctif a un diametre d e 55 mm).


Biokarst on a tropical carbonate island organic processes are instrumental in their continued development. Soil pipes are enlarged by percolating meteoric water acidified by biogenic C02 in the soil. Some authors have hypothesized that soil pipes are a direct result of dissolution by acidified water concentrated around tree roots (BRINK & PARTRIDGE, 1980; WALSH & MORAWIECKA-ZACHARZ, 200 I). Root karst In addition to soil pipes, a variety of small-scale karst features, both erosional (WALL & WILFORD, 1966) and accretional (WILfORD & W 1965) are thought t? form by plant root action. BuLL & LAVERTY ( 1982) discuss those features as types of physical phytokarst and propose the term rhizokarst, and KLAPPA ( 1980) provides a review. On Guam, at least three types of small-scale features appear to be produced directly by plant roots. Root grooves (WALL & WILFORD, 1966) have been documented on the freshly exposed surface of a single collapsed boulder in Tarague. They appear as an irregular ramifying network of hemispherical grooves (Fig. 9). Root holes are made in limestone rocks by roots of trees. The best examples on Guam are associated with the banyan trees (Ficus prolixa) that commonly grow around the entrances of caves (Fig. I 0). The roots completely penetrate the cave roof, emerging from widened holes in the rock (Fig. I 0, inset), and reach the cave floors by growing considerable distances through the open air of the cave Upon reaching the floor, they grow into the soil, anchoring the tree and tightening the sub aerially exposed portion of the root until it becomes completely straight and firm. The roots are typically several centimeters in diameter, and probably cause the widening of holes in the bedrock in which they grow, indirectly by acidifying the rain 83 water flowing down the roots, or directly by the pressure they exert on the host rock as they grow. Some cave entrances are almost entirely concealed by dense curtains consisting of hun dreds of banyan tree roots. Rhizoliths, most likely root tubules (KLAPPA, 1980), have also been documented in the reef limestones on Guam. Found standing in the centers of small pits (of primary or solutional ori gin), these features are tufaceous stalks, up to 7 em tall (Fig. II). The base of the stalks, at the bottom of the pits, is weak enough to be broken by hand. There are partially in filled holes, a few millimeters in diameter, running the length of the stalks. Freshwater tufas Freshwater tufils, travertines, stromatolites, and similar deposits can be precipitated by waters supersaturated with respect to calcium carbonate. CaC03 can be precipitated indirectly by the removal of C02 by cyanobacterial and algal photosynthe sis (CASANOVA, 1981 ), or directly by organisms (CHAFETZ & FoLK, 1984). The role and contribution of biota to the indi vidual processes is subject to debate. Some studies suggest that the influence of biota on sedimentation is less than is gen erally thought (PENTECOST, I 985) or even negative, because of bioerosion (ScHNEIDER, I 977). Biologically produced fresh water deposits are typically not found on small carbonate islands, sirice such islands usually lack surface waters in karst In Guam, however, in the cockpit karst terrain in central southern Guam, several carbonate-rich streams emerge from caves and flow moderate distances as surface streams over the alluvium-covered carbonate units. In those environments, minor tufa and travertine deposits are associated with cascades and small waterfalls, and are shaped as laminar tapered domes, reflecting the trajectories of cascading water. Fig. II. Tufaceous stalks (root accretion features), Double Reef The filter is 55 mm in diameter. Tiges tufieres (formation d 'accretion autour des racines), Double Reef Le filtre a un diametre de 55 mm


84 D. Taborosi Fig. 12. Tufaceous stalactites in cliff face settings. a. breached flank margin cave, Ritidian. The largest stalactite is 2.meters long (note person for scale, lower le.ft); b. horizontal. notch containing numerous tufaceous stalactites, Tarague (note person for scale, at the bottom, off center). Stalactites tujieres. enfalaise. a. grotte de type breche de bord dejlanc, a Ritidian .. La stalactite Ia plus grande a une longueur de 2m (a noter le pe;sonnage d'en bas. a gauche, pour /'echel/e); b. encoche horizon tale contencmt de nombreuses stalactites tujieres, a Tarague (a noter /e personnage d 'en bas pour l'echelle).


Biokarst on a tropical carbonate island Subaerial tufas (tufaceous stalactites) Stalactites made of tufa-like deposits are fairly common in the humid tropics and many are reported from cave entrances. They have not been subject to detailed studies and their formation mechanisms are poorly known, but the activity of cyano bacteria, algae, and mosses is often suggested. BULL & LAVERTY ( 1982) classified these features as depositional phytokarst and called them non-vertical stalactites. VILES & GoUDIE (1990) refer to the terms "aussen Stalactiten" and "Stalactiten vorhang" from German literature on tropical tower and cone karst. Terms applied to analogous features include cave karst stromatolites and heliotropic speleothems (FUXING & JIANt-JUA, 1993), phototropic phytospeleothems and phytoerratics (LICHON, 1992), subaerial stromatolites (Cox eta/., 1989) and stromatolitic stalagmites (JAMEs eta/., 1994). On Guam, tufaceous stalactites, which can be thick and massive, are ubiquitous in cliff-side entrances ofbreached flank margin caves (Fig. 12a), and may also be found in horizontal notches on coastal cliffs (Fig. 12b). Unlike normal cave sta lactites, the growth axis oftufaceous stalactites is not always vertical. Their leaning can be a result of preferential deposition by bacteria or algae on the side facing the light, as suggested by BuLL & LAVERTY (1982); but groups of stalactites leaning in conflicting directions, and specimens exhibiting curved shapes, appear to indicate more complex growth histories. Tufaceous stalactites, typically Jacking the crystalline luster of normal cave speleothems, are lightweight, porous, and friable, and feel powdery or earthy. Thf;y can be easily broken byhand, revealing white chalky interiors, homogeneous from Fig. 13. A lufaceous stalactite from Ritidian. The scale is in centimeters. Stalactite tufiere de Ritidian. &helle en centim"jtres. 85 the periphery to the core, and generally deposited in concen tric layers (Fig. 13). Somewhat harder micritic stalactites, simi lar to those described from the Cayman Islands by JoNES & MoTYKA (1987), have also been documented They appear to be transitional forms between tufaceous and normal cave speleothems, since they commonly contain sparry calci t e in addition to chalky and microcrystalline fabrics. X-FaY. diffrac tion (XRD) analyses of selected samples from Guam have indicated that they are uniformly composed of low-Mg calcite. The origin oftufaceous stalactites is poorly known. Although the most likely mode of deposition is the cyanobacterially assisted growth of subaerial tufa in cave entrances, the decay of cave (normal) speleothems, when exposed to open atmo sphere by cliff-retreat. can also be a factor. Understanding the origin oftufaceous stalactites in coastal cliffs is crucial to karst studies on carbonate islands because it makes possible the easy distinction between notch-like breached flank margin caves and marine notches (M YLROJE & CAREw, 1997; CARNEY et a/., 1997). The two commonly appear similar, but whereas both are associated with former sea levels, only the flank margin caves are related to freshwater discharge and may provide clues about the paleodynamics of the freshwater lens. Until recently, the presence of extensive speleothem deposits in the cliff-side voids was considered the key factor in distinguishing breached caves from marine notches; but the discovery of massive tufaceous stalactites, apparently growing in open at mosphere conditions has blurred that distinction. "Directed phytokarst" Unusual small-scale, light-oriented features are sometimes reP<>rted from the twilight zones of tropical caves. Apparently biologically produced erosional forms, they have been termed photokarren by WALTHAM & BRooK (1980) and "directed phytokarst" by BuLL & LAVERTY (1982), based on examples from insular Malaysia Morphologically similar features have been documented on Guam, where they grow on the sides of tufaceous speleothems in certain cave entrances They appear to be composed of ca lcified algal filaments, up to 5 mm in length, intertwined and grouped in clusters pointing toward the light (Fig. 14). Although superficially comparable to ero sional forms termed photokarren and "directed phytokarst'' the structures observed on Guam are clearly depositional. As such, they represent unique and, to the best of my knowledge, previously unreported karst features. No specific term is being proposed yet as these features are currently under active study. Moonmilk Moonmilk, considered a speleothem, has been described as "soft. plastic and pasty'' when wet and "crumbly and pow dery" when-dry (HILL & FoRTI, 1997). It is a microcrystalline aggregate, containing calcite and other minerals. It is thought to be precipitated inorganically (BoRsATO eta/., 2000), as well as biologically (NoRTHUP & LAVOIE, 200 I) by bacterial action


86 ;rill. -.: Jl : ... 11 5F:\1 Fig. 14. Light oriented thread-like depositional features on the surface of a tufaceous stalactite, Ritidian. The scale is in centime ters. Structures jilamenteuses j{1iblement orientees, deposees sur Ia surj{1ce d'une stalactile tufiere de Ritidan. chelle en centimetres. (DANIELL! & EDINGTON, 1983) and fungal action (PHILLIPS & SELF, 1987), both intra-and extra-cellularly (GRADZINSKI et al.j 1997). Moonmilk can also form by speleothem decay (HILL & FORTI, 1997). BULL & LAVERTY ( 1982), therefore, classify it as both a constructive (depositional) and a destructive (erosional) form ofphytokarst. On Guam, moonmilk-like loose powders and wet pasty sub stances and biomats are found commonly on speleothems and cave walls and can be up to several millimeters thick. They are typically purple, green, and gray (Fig. 15) and only in cave areas where light penetrates. The best developed ones are found in breached flank margin caves, which are ex posed to light and outside atmosphere by cliff retreat or roof collapse. They are also found periodically submerged in caves having freshwater pools, such as Marbo Cave, where they cover the walls in the range of tidal fluctuations. Similar biomats, reported from Mexican cenotes, are thought to be products of photosyr:tthetic purple bacteria (GARY, 2002). D. Taborosi Fig. 15. Organic coating (scratched by hand), covering a stalagmite in a breached tlank margin cave in Ritidian. Color of the coating is purple, with green and ochre patches. Pe/licule organique (grallee a Ia main) couvrant une stalagmite d'une grol/e de type breche debord dejlanc de Ritidan. La pe/licule est de couleur pourpre, avec des taches vertes et ocres. Conclusions Karst features that are thought to be directly related to organic processes have been termed phytokarst (BuLL & LAVERTY, 1982) or biokarst (SCHNEIDER & TORUNSKI, 1983; VILES, 1984). They have largely been regarded as features of local interest and of limited significance (VILES, 1988). Whereas this may be the case with certain individual organisms and unusual small-scale features, the effects of living organisms on karst geomorphol ogy are profound on an ecosystem scale, and they are wide spread, intense, diverse and of fundamental importance. As a rising number of karst features are recognized as biokarst, the roles of biota in processes beyond the small-scale phenomena are becoming the focus of important studies. Organisms are now considered instrumental in rock surface weathering (VILES, 1995) and speleogenesis (SASOWSKY & PALM!'R, 1994; MARTIN eta/., 2002). The evolution of entire karst landscapes is thought to be biologically controlled through the interrelationships of


Biokarst on a tropical carbonate island Table I. Karst features of Guam and the most important bio!ngic and inorganic processes thought to be associated with. them. References arc given in text. Formations karstiques de Guam etle.f plus importants procc:.fSUS biologiques et inorganique qui paraissent y eire associes. Voir le Iexie pour le.f references bibliographiques. Karst features Coral reefs /. Intertidal benches and rimmed terraces lnlluenccs on observed geomorphology Biologic Inorganic QicX:Onstruetion bj cOiaJs; eonilline algae; Wave action; freshwater discharge; . .: .. Bioconstruction by coralline algae, vcnnetid Dissolution by wave agttatoo seawater; mollusks; Diocrosion by rocks Bioerosion' (bioid)mion by Srazersi< Di5Solution by wave agit8ted seawater; r' biocorrosion by. endolilf:as) -ffiechMical abrasion . -Marine notches Champignon surfaces and pinnacles Karren Biocrosional karrcn SOlution pans Biocrosion (biocorrosion by endoliths and epiliths) Dissolution; wetting and drying; salt weathering and hydration Biologicweathefini dissohidon by wdter by organic COz\ -. : :: : by, Diocrosion (bioabrasion by, endoliths) Dissrilution by organically acidified watci-; mechanical action OflichcnS -' ; : precipitated meteOric water -.' PhysicO-chemical by accumulated meteoric water._ "': Beachrock lnllucncc ofintcrtidalmicrollora muJ hctcrotrophs on pi I mKI C02 levels C02 degassing due to evaporation, wave action, groundwater discharge Case-hardening Wld calcrete Solution pits Soil pipes Root grooves and root holes Rhizoliths Freshwater tulil deposits Tufaccous stalactites Direct and indirect precipitation of caJcite by bacteiia, algae, lichen&.... .. lliocorrosion; dissolution by organic acids in water; root action : Dissolution by organically acidified water. root .. .. ; .. = .. :. Dissolution by orgm1ic acids from plants: mcclmnical action of roots Binding effect of plant .roots Indirect precipitation hy autotrophs {C02 removal): Direct precipitation Indirect precipitation by autotrophs (C02 removal); Direet precipitation "Directed phytokarst" Direct or indirect precipitation or corrosion by autotmphs precipitation of calcite from carbonate-rich . Physico-chemical dissolution by pcrcolnting meteoric water Ph)'sic eta/., 1995), and some authors have suggested that the process ofkarstitication is essentially ofbiogenic character (BARANY-KEVEI, 1992). Bio logic influence is being demonstrated in an increasing variety of karst features, and even such quintessential abiotic and hy drodynamically controlled fonns as rillenkarren have recently been I inked, on a fundamental level, to biologic precesses (Ftot. et a/., 1996). All this is perhaps nowhere as obvious as it is in tropical and coastal settings, where the distinction between "biokarst" and "non-biokarst" becomes thoroughly impracticable. _as nearly all karst features are affected by biologic processes, in addi tion to the inorganic ones (Table I). This survey of karst on Guam has revealed that practically all smalland medium scale exokarst features documented on the island arc biologi ca 'ly mediated and bear characteristics of what is known as "biokarst". The biologic influences on karst actually extend even further:-On tropical carbonate islands, even the fonna tion of large caves is thought to be greatly dependant on or ganic processes. Some of the most common ishind caves, known as flank margin caves (MYLROIE & CAREW, 1990), have


88 developed in improbably short time spans if the on_ly mecha-, Acknowledgments nism invoked is inorganic dissolution (SANFORD & KoNtKow,. 1989). Their speleogenesis is largely a result of enhanced'dis. The research reported here. is part of a project funded-by: lhe solution through oxidation of organic materials transPorted U.S. Geological Survey, Water Resources Division, through by vadose water from the epikarst and accumulated at the 1 the Water Research Institute, Regional Competi pycnoclines in the freshwater lens (MYLROIE & BAlcERZAK, live Award Program, award no. 1434-HQ-96-GR-02665.Inj-1992). All this indicates that, in tropical and coastal environ. tial work was also supPxted by the Guam Hydrologic ments, the biologically, influenced karst are -"ot small-; scale peculiarities, but actually encompass of, if not all :. of the Western Pacific, University of Guam, wtiicli \\ras initi.l karst phenomena. / .. ated 'by the 24111 fhf author i s References AcsAso, C., GALVIN, J., & RooRIOUI!ZPASCUAL, C. (1982) The weathering of calcareous rocks by lichens. Pedobiolngla, 14, pP. 219-229. ALBERGARIA MoREIRA, M.E.S. ( 1996) DiO-crosional features in rock reefs at coasts of Mozambique and Drazil. Zfilr Geomorpho. Iogie S11pplemenlband, I 01, pp. IS 1-168. AI.ONso, Z.A.M., Sn.vA, P.G, Gov. J.L., & ZAZO, C. (1998) Fan-surface dynamics and biogenic calcrete development: interactions during ultimate phases of fan evolution in the semiarid SE Spain '(Murcia). Geomorphology,14, 1-3, pp. 147-167. RARANVKF.vm, I. ( 1992) factcurs ccologiqucs duns lalhrmatic)n du kursl. In: Kmwt "' evollllitm.f clillltllitiiii'.Y (SAlOMON; J.N. & MAIKE, R .. EdY.),I'n.:s.'ICS Univcrsituires de Bordeaux, Bordcuux. ,pp .SI-59. , John Jensan, Dr. JohnMylroie, andDavid Vanri.for in,;; valuable support in'thb field, to or; David J.'I.bwe for an exJ ccptionally thorougli review and indispensable and to Dr. JOhn Mylroie and Dr. Paid Williams for importlnt critical comments : Guidance 'and support of Dr. ; Kazuomi Hirakawa are aisd great_l_y . CAiiiVI!RA$, SANcui!ZMORA'-'.s.,.SLOER, v., & SAizJIMENez.C . (2001) Microo,rganisms and microbi_ ally .induccd fabrics in cave walls. GeomicrobiologyJournal, 18, 3, pp. 223-240 . .:. .. ' CAKNP.V, S.M., MvLRom, J.E., & CAREW, J.L. (1997) Bioerosion notches versus nank margin caves; viability as palco sen level indicators.,J. ofCave and Kar:rl Sl11die:r, 59, 3, p. 169. CASANOVA, J. ( 1981) Morphologic et biolithogcncse des barrages de travertines. Bull. de I'A:r:rocialion de Gengraphie 479-' 480, pp. 192-193. J ... = .:"1-1 l: I J I ...... I I I CASTANII!R, s .. Ll; MliTAVI!R-LI!VRfll., G., & PERTIRJISOT, J.-1'. (1999) Ca-carbonatcs precipitation and limestone genesis-the microbiogcologist point of view. Sedimenlary Geology, 116, /-4, pp. 9-23. 1.\ 0 .. CAIIMAII'IlN, V. & ltliNAIII:r, P. ( 19S8) l.n corrosion hiochimiquc dans un ct In gcncsc llu incmdmilch. Note.f biospel., IJ, pp. . . DARTON, H.A., SPEAR, J.R . & PAcF., N. R. (2001) Microbial Life in ,-CuAFI!TZ, 1-I.S. & FoLK, R.L. (1984) Travertines: Depositional mor-lhc Underworld: Biogenicity in Secondary Mineral Formations. .phology and the bacterially constituents. J, SedimenGeomicrobiology Jo11rnal, 18, J, pp. 359-368. lory Pelrology, 54, pp. 289-316 . ' 11AVARI, C.S., KuRrrAs. T.,'& TF.MEL, A. (1994) Hydrogeology, Cm!N, J., DUJME, H.-P., & BEVERM, L. (2000) Wcatherif!g of rocks hydrogeochcmislry. and geochcmislry oft he biogenic travertines induced by lichen coloni1.ation -a review. Calena. 39, 2 pp. in I he lower Zamanti Basin (ca.o;lem Taurids-Turkey). Cave and 121-146. -<-. 1 Karst Science, 11, /., p. S. p v .. Cox, G., JAMES, J.M., LEGGETT, K.E.A., & OsBORNE, R.A.L. (1989) P. & _DAI.ONGI!vu.u:, R. ( 1996) Mediterranean coastal changes Cyanobacterially deposited speleothems: subaerial stromatoliies. recorded in beach-rock cementation. z. filr Geomorphologie Geomicrobiolog)'JOIIrnal, 7, pp. 24S-2S2. . Sllpplementband. I Ol, PP 18S-198. DAI.ON

Biokarst on a tropical carbonate island DICKINSON, W. R. (lOOO) Hydro-isostatic and tectonic influei1e:es on emergent'Holocene paleoshorelincs in the Mariana Islands, We:-1-ern Pacific Ocean. J. of Coastal Research, 16, 3, pp. 735-746. DRYSDALI!, R.N. ( 1998) Aquatic insect larvae as geomorphic agents in travertine-building: a case study from the Barkly karst, Australia. Geografla Fisico e Dinamica Quaternaria, Suppl., 3, pp. 53-59. DRYSDALI!, R.N. (1999) The sedimentological significance of hydro. psychid caddis-fly larvae (Order: Trichoptcra) in a travertine-de positing stream: Louie Creek, northweStern Queensland. J. a/Sedimentary Research (Section A), 69, I, pp. 145-150. EASTON, W.H., Ku, T.L., & RANDAI.L, R. H. (1978) Recent and shorelines on Guam. Mlcronesica, 14, /, pp. 1-11. EMERY, K. 0. (1962) Marine Geology ojG11am. 403-0, U.S. Geo logical Survey Professional Paper, U.S Government Printing Office, Washington, D.C. 76 p. F1o1., L.l., FoRN6s, J.J., & GINEs, A. (1996) EITccts ofbiokarstic processes on the development ofsolutional rillcnkarrcn in limestone rocks. Earth S11rjace Proc. and Landforms,21, 5, pp. 447-452. FISCIIER, R ( 1990) Biogenetic and nonbiogenctically determined morphologies on the Costa Rican coast Z.ftlr Geomorphologic, 34,J,pp.313-321 FocKn, J. W. ( 1978) Limestone cliff morphology on Cumr;no (Neth erlands Antilles), with special nttcntion to thcorigin of notches and vcrn\ctid/cornllinc algal surf benches ('cornices', 'trolloirs') Z. jilr Gemnorphologie,2, pp. 329-349. foLK, R.L., Roni!R1'S, H.ll. & MooRE, CJI. (1973) Black phytokarst from Hell, Cnymnn Islands, British West Indies. Geological Society of America fl11ll., 84. pp. 2351-2360 FoRo, D.C. & Wn.LIAMS, P. W. (1989) Kar.vl geomorphology and hy drology. Unwin Hyman, Winchester. Massachusetts, 320 p. foRo, T. D ( 1989) Tufa-The Whole Dam Story. Ca1e Science. 16, 2. pp. 39-50. rorm. P. (200 I) Biogenic speleothems: an overview. Inti J. Speleol . 30A. I pp. 39-56. FuxrNG, W & JrANBUA, C ( 1993) Karst cave stromntolites. In: Geol ogy, Climate, Hydrology and Kant Formation: IGCP 299Netv,flel/er (SHI!NOYOU, X. & DAOXIAN, Y., Eds.), The Institute ofKarst Geology, Guilin, China, 79 p GARY, M. (2002) UndcrstandingZacat6n: Exploration and initial interpretation of the world's deepest known phreatic sinkhole andrelated karst features, southern Tamaulipas, Mexico. In: Hyrirogeolog) and Biology of Post-Paleozoic Carbonate Aqttifors, Symp. Proc.: Gainesville. Fl, (MARTIN, J.B., Wrci:S, C.M. & SASOWSKY, J.D., Eds ) Karst Waters Institute, Gainesville, Florida, pp. 141-145. GISCIILER, E. & LoMANDO, A.J. (1997) Holocene cemented beach deposits in Belize. Gcol., UO, pp. 277-297. Gouom, A.S. ( 1983) Calcrete. In: Chemicai Sedimenls Geomorphology: Precipitates and Re.fidua in the Near-S11rjace Enl'imnment(GouoiP1A.S. & PYE.. K., Ed.f.),Academic Pres.<;, London, pp. 93-131. . GRAoZJNSKI, M., SzuLC, J., & SMYK, B. ( 1997) Microbiai agents of moonmilk calcification. In: Proc. I )th Inti Congress ofSpeleology. /,a Chaux-de-Fonds, Stvitzerland (JilANNIN, p. Y., Eif.), pp. 275-278. 89 HI!RWfTZ, S.R. (1993) Stemflow influences on the formation of solution pipes in Bermuda aeolianite. Geomorphology, 6, pp. 25J:-271. C.A. & Fom, P., Eds. (1997) Cave minerals of the world. NaSpeleological Society, Huntsville, Alabama, 238 p. Hooi'Eit, J.H.D. ( 1958) Bat erosion as a in cave formation. Nature, 182, p. 1464. Hurr, C.R., OSTERKAMP, W.R.,-& HowARD, AD., Eds. {1995) Biogcomorphology, terrestrial and fresh Willer systems. Geomorphol ogy Special Issue, 13, /-4, pp. 1-347. JAMES, M.J., PATSALIDI!S, E., & Cox, G. (1994)Amino Acid Compo sition of Stromatolitic Stalagmites. Geomicrobiology Journal, 12, pp. 183-194. JENNINGS, J.N. (1985) Karst geomorphology. Basil Blackwell, Ox-ford and New York, 293 p. JI!NSON, J. W. & SrooRLcrnchom Peninsula, central Greece. Marine Geology, 179, pp. KIIADKIKAR, A.S., L.S. & RAMI!SH, R. (2000) The charac ter and genesis of calcrete in Late Quaternary alluvial deposits, Gujarnt, western India. and iL'i bearing on the interpretation of ancient climates. Palaeogeog,.aphy, Palaeoclimatology, Palaeoecology, 162,3-4, pp. 239-261. KLAPPA, C. F. ( 1979) Lichen stromatolites: Criterion for subaerial exposure and a mechanism for lhc formation of laminar calcretes (caliche). J. of Sedimentary Petrology, 49, pp. 387-400. KLArrA, C.F. (1980) Rhizoliths in terrestrial carbonates: classification, recognition, genesis and significance. Sedimentology, 27, pp.613-629. KoBLUK, D.R. & RISK, M.J. ( 1977) Rate and nature of infestation of a carbonate substratum 'by a boring alga. J of Experimental Marine Biology and Ecology, 27, pp. I 07-115. KRUMUI!IN, W.E. (1979) Phololitholrophicand chcmoorganotrophie activity of bacterin and algae ns related to beachrock forn1ation and degradation (Gulf of Aqaba, Sinai). Geonricrobiology Jom"nal, I, pp. 134-203


90 KRUMDEIN, W.E. & Guir."e, G (I 979) Calcification in a coccoid cyano bacterium associated with the formation of desert stromatolites. Sedimentology, 26, pp. 593-{)04. LANDER, M.A. (I 994) Meteorological factors associated with drought on Guam. Water and Energy Research ln.rlilule of the Western Pacific. University of Guam, Technical Report No. 75. LICHON, M.J. (1992) The phototropic phytospcleothcms of Moss Palace, Mole Creek, Tasmania. fle/ictite, JO, /, pp. 8-10. LoNGMAN, M. W. & BROWNLER, D.N. ( 1980) Characteristics of karst topography, Palawan, Philippines. Z. ji1r Geomorphologie, 24, 3, pp. 299-3 I 7. MARTIN, J.B., WICKS, C.M., & SASOWSKY, J.D . Eds. (2002) Hydrogeology and Biology ofPost-Paleo7..oic Carbonate Aquifers, Proc. Symp ... Karst Frontiers: Florida and /lelated Environments". Karst Waters Institute, Gainesville, FL, 212 p. MERZ, M.U. (1992) The biology of carbonate precipitation by cyanobacteria. Facies, 26, pp. 8 I -10 I. MILI..RR, T.E. (I 990) Bell holes; biogenic (bat) erosion features in tropical caves. Geo (I I th Friends of Karst meeting; program and ab stracts, National Speleological Society). 17,2-3, p. 76. MooKE. C. II .. JR. ( 1973) Intertidal carhmmte ccmentation, Grand Cuynmn. West lndic.'> .1. Sediment. l'etrol., 4J, pp. 591-(,()2. Mos1;s, C.A. & SMrnl, B.J. (1993) A note on the role of the lichen Co/lema auriforma in solution basin development on a carbon iferous limestone substrate. EarJ/r Surf Lane/forms, 18, pp. 363-36R. MYLROIE, J.E. & BALCr:RZAK, W.J. (1992) Interaction ofmicrobiology and karst processes in Quaternary carbonate island aquifers. In: Proceedings oft he Firstlntemational Conference 011 Ground Water Ecology (ST,\NFORD, J.A. & SIMONS, J.J., Eds.), American Water Resources Association, Bethesda, MD, pp. 37-46. MnRoJE, J.E. & CAREW, J.L. (1990) The nank margin model for dissolution cave development in carbonate platforms. Earth Sur face aml/.am{forms. 15, pp. 413-424. Mvr.ROIE, J.E. & CARRW, J.L. (1991) Erosional notches in Bahamian Carbonates: biocrosion or groundwater dissolution? In: Proc. Fifth Symposium on Geology of the Bahamas, Bahamian Field Station, Port Charlotte, Florida, pp. 85-90. MYLROIE, J.E. & CAREW, J.L. (1995) Chapter 3: Karst development on carbonate islands. In:. and Porosity in Car bonate Strata (llunD, D.A., 1-lARRIS, P.M . & SAr.Lr:R, A, Eds.), 63, Am. Ass. of Petroleum Geologists, pp. 55-76. MYI.ROIE, J.E. & CAREW, J.L. (1997) Land usc and carbonate island karst. In: The Engineering Geology /lydrogeo/ogy of Karst Terranes (BECK, ll. r .. & STI:PIIENSON, J.B . Eds.), Balkcma. Rollcrdam, pp. 3-12. Mvi.Rom, J.E., JI!NSON, J.W., D., JocsoN, J.M.U., VANN: D.T., & Wnxr:L, C. (2001) Karst features of Guam in terms of a general model of carbonate island karst. J. of Cave and Kar.fl Studies, 63, 1 pp. 9-22. MYLROIE, J.E. & V ACIIF.R, 1-I.L. (I 999) A conceptual view of carbon ate island karst. In: Karst Modeli11g, Symposi11m Proceeding.r: Clwrlollesvil/e, VA (PALMER, A.N., PALMER,' N.Y., & SASOWSKY, J.D., Eds.), pp. 48-58. D. Nom1ur. D.E. & LAVOIE. K.H. (200 I) Geor,1icrobiology ofCavcs: A Review. 18, J, pp. 199-222. PI!NTI!COST, A. ( 1985) Association of cyanobacteria with tufa dcposil'>: Identity, enumeration and nature of the sheath material revealed by .histochemistry. Geomicrobio/ogica/ Journal, 4, pp. 285-298. P1m.urs, S.E. & SELF, P.G (I 9S7) Morphology, crystallography and origin of needle-fibre calcite in Quatcmary pedogenic of South Australia. A11stralian Journal of Soil Research, 25, I, pp. 429-444. PIKAzzou, P.A. (1986) Marine Notches. In: Sea-level Research: a Man11al for the Collection a11d Evaluation of Data (vAN DE PLAsscne, 0., Ed.), Gco Books, Norwich, UK, pp. 361-400. P1KAZ7.0LI, P.A. (1996) Sea-l..eve/ Changes in the Las/20,000 Years. Wiley, New York, 211 p. RANI>AI.I., R.H. & HoLLOMAN, J. (1974) Coastal survey of Guam. Marine Laboratory. University of Guam, Technical Report 14. 404 p. RANDALL, R.H. & Srr:oKIST, H.G. JK. (1988) Geomorphology oflhc fringing reefs of northern Guam in response to Holocene sea level changes. In: /'me. 6th /nltrlltrlimm/ Cura/ Reef Symfm .fium. Amtlr11lia, J, pp. 473-477. RIDING, R. (2002) Structure and composition of organic reefs and carbonate mud mounds: concepts and categories. Earth-Science Review.f, 58, pp. I 63-23 I. RuST, D. & KRRSIIAW, S. (2000) Holocene tectonic uplift pallcrns in northeastern Sicily: evidence from marine notches in coastal outcrops. Marine Geology, 167, 1-2, pp. 105-126. SAND, w.,AIILI!KS,Il., & BocK, E. (1991)Thc impactofmicroorganisms-especially nitric acid-producing bacteria-on the deterio ration of natural stones. Science, Technology and European Cultural Heritage (BAEK, N.S., SAIIDIONI. C., & SoRs, A. I., Eds. ), Oxford. pp. 481-484. SANI'ORU, W.E. & KoNIKow, L.F. ( 19R9) Porosity development in coastal carbonate aquifers. Geology, 17, pp. 249-252. SASOWSKY,I.D. & P,\I.MER, M.V., Eds. (1994) BreakJhroughf in Karst Geomicrobiology and Redox Geochemistry. Karst Waters Insti tute, Colorado Springs, CO, II I p. ScHNEIDr:R, J. ( 1977) Carbonate construction and decomposition by cpililhic and endolithic microorganisms in saltand fresh-water. In: Fossil algae (FwoEL, E . /:.a.), Springer-Verlag, Berlin, pp. 248-260. ScBNr:IDEK, J. & LE CAMPION AI.SUMARD, T. ( 1999) Construction and destruction of carbonates by marine and freshwater cyanobacteria. .lour11al of l'hycology, 34, pp. 417-426. SciiNRIDER, J. & ToRUNSKI, H. (1983) Biokarst on limestone coasts, morphogenesis and sediment production. Mari11e Ecology, 4, pp. 45-63. SMYCK, B. & DRZAL, M. ( 1964) Untersuchungen ubcr den Einnuss von Mikroorgmiismcn auf das Phanomen der Karstbildung. EJdku11de, 18, pp. 102-113. Sri!NCER, T. & Vn.I!S; (:I.A. (2002) Bioconstruction, bioerosion and disturbance on tropical coasts: coral reefs and rocky limestone shores. Geomorphology, (in press).


Biokarst on a tropical carbonate island STANTON, W.l. ( 1984) Snail holes in Mendip limestones. Proceedings, Bristol Naturalists' Society, 44, pp. 15-18. STEARNS, H.T. (1941) Shore benches on north Pacific islands. Geological Society of America Bulletin, 51, pp. 773-780. STODDART, ri.R . TAYLOR, J.D., fARROW, GR., & FOSIII!RO, F.R. ( 1971) The geomorphology of Aldabra. In: A discussion on the ruults of the Royal Society Expedition to Aldabra, 1967-1968. (WESTOU.,, T.S. & STODDART, D.R., &Is.), Phil. Trans. Roy. Soc. London, pp. 31-65. SwEETING, M.M. ( 1966) The weathering oflimestones (with particular reference to the Carboniferous limestones of Northern En glimd). In: Essays In Geomorphology (DuRY, GH., Ed.), Heinemann, London, pp. 177-210. T D. (2000) Kant foaturu on Guam. MSc. Thesis, Water and Environmental Resean:h Institute of the Western Pacific, University of Guam, 324 p. TRACEY, J.l., JR., ScHLANOI!R, s.o . STARK, J .T., DoAN, D.O., & MAY, H.G (1964) General Geology ofGuam. US Geological Survey Paper 403-A, US Government Printing Office, Washmgton, DC, 104 p. TRICART, J. & DASILVA, T.C. ( 1960) Un exampled. evolution karstique en milieu tropical sec: le mome de Bon Jesus de Lapa (Bahia, Bresil). Z.ftlr Geomorphologie, 4, pp. 29-42. TRuDOtLL, S.T. ( 1976) The subaerial and subsoil erosion of limestones on Aldabra Atoll, Indian Ocean. Z filr Geomorphologle Supplementband,l6, pp. 201-210. TRuDOtLL, S.T. ( 1983) Measurement of rates of erosion of reefs and reef limestones. In: Perspective on Coral Reeft (BARNES, D.J., Ed. ), Austral. lnst. Mar. Sci Contrib., lO, pp. 256-262. TRUDOILL, S.T. (1987) Rioerosion of intertidal limestone, Co. Clare, Eire 3: Zonation, process and form. Marine Geology, 74, pp. IJ 1-121. V ERRECCHIA, E .P. ( 1990) Litho-diagenetic implications of the calciumoxalate-carbonate bio&eochemical cycle in semiarid calcretes, Nazareth, Israel. GeomicrobiologyJournal, 8, pp. 87-99. VILES, H.A. (1984) Biokarst: review and prospect. Progress in Physi-cafGeography, 8, pp. 523-542. 91 V1U!S, H.A. ( 1987) Blue-green algae and terrestrial limestone weath ering on AldabraAtoll: an SEM and light microscope study. Earth Surface Processes and Landforms, ll, pp. 319-330 VtLES, H.A. (1988) Organisms and karst In: Blogeomotphology {VILES, H.A., Ed.), Basil Blackwell Ltd., New York, pp. 319-350 Vn.ES, H.A. (1995) Ecological perspectives on rotk surface ing: Towards a conceptual model. Geomorphology, 13, pp. 21-35. VIU!S, H.A. (2001) Scale issues in weathering studies Geomorphol-ogy,41,pp.63-72. VILES, H.A. & Gouote, A.S. ( 1990) Tufas, travertines and allied carbonate deposits. Progress in Physical Geography, 14, pp. 19-41. V1U!S, H.A. & SPI!NCI!R, T. ( 1986) 'Phytokarst,' blue-green algae and limestone weathering. In: New directions in Korst (P ..m!RSON, K. & Swl!lmNo, M.M., &Is.), GeoBooks, Norwich, pp. 115-140 WALL, J.R .D. & Wn.FORD, GE. (1 966) Two smaiJ scale solution features of limestone outcrops in Sarawak, Malaysia. Z ftlr Geomorphologie, 10, pp. 91-94. WALSH, P. & MoRAWIP.CKAZACHARZ, I. (200 I) A dissolution pip;c: palaeokarst ofmid-Pieistocenc age preserved in Miocene lime stones near Stasmw, Poland. Palaeogeogroplry, Palaeoclimato-logy. t:alaeoecology, 174, 4, pp. 327-350. WAL11tAM, A .C. & BROOK, D.O. (1980) Cave development in the Melinau limestone of the Gunung Mulu National Park. J. Geogr., 146, pp. 258-266. WEBo, GE., JeLL, J.S., & BAKER, J.C (1999) Cryptic intertidal microbialites in beac:;hrock, Heron Island, Great Barrier Reef: implications for the origin of microcrystalline beachrock cement. Sedimentary Geology, ll6, pp. 317-334. WENT, F. W. ( 1969) Fungi associated with stalactite growth. Science, .16, pp. 385-386. WHrre, W.B. (1988) Geomotphologyand hydrology ofkarstte"ains Oxford Press, New York, 464 p. WtLPORD, GE. & WALL, J.R.D. (1965) Karst topography in Sarawak. Journal ofTropical Geography,ll, pp. 44-70. ZHANG, J. & 8Ao, H. (1994) On modem biokarst erosion on Yongan stone pinnacles, Fujian province. Mountain Research, ll, 3, pp. 136-144 5 2002 17 Nollf!mber 2002 accepted : 29 NOIIf!mber 2002


Thc:orctical and Applied Karstology, 15 (2002), pp. 93-99 The Quaternary morphogenesis of the Lagoa Santa tropical karst, Minas Gerais State, se Brazil Heinz Charles Kohler 1 & Joachim Karfunkel2 I Ponli/(cia Univer.fidade Cat6lica de Minas Gerais, A v. Dom Jose Gaspar, 500, Eucarlstico, Horiz.onte, Brazil. 2 I Departamento de Geologia, Univer.fidade Federal de Minas Gerais, Campus Pampulha, Av. Antonio Carlo.f 6627, 31270-901 Belo Horiz.onte, Brazil. Abstract The tropical karst of the Lagoa Santa region, north ofBelo Horizonte, in South-Eastern Brazil was studied, in order to describe the Quaternary morphogenesis and to elucidate the morphodynamie evolution of this area 11te karst developed over an interfluvial block, at altitudes between 650 and 850 m above sea level. Structuml alignments. consequence of the Brasilian geotectonic cycle, were reactivated through the Cenozoic. These alignments. together with climate oscillations through time, control the main trends of the karst scenario and allow the authors to postulate a morphogenetic history from Tertiary through Holocene. Key words: geomorphology, tropical karst, Precambrian carbonates, lithostratigraphy, karst morphogenesis, Quaternary climate fluctuation. La morphogenese quaternaire du karst tropical de Lagos Santa, Minas Gerais, sud-est du Bresil Resume Le karst tropical de Ia rdgion de Lagoa Santa sllue au nord de Be/o Horizonte, dans le sud-est du Bresil, a ete etudld dans /e but de decrire sa morphogenese quaternaire et d 'e/uclder /'evolution morphodynamlque de Celie zone Ce lcorsts 'est developpe sur un bloc interfluvia/, a 650-850 m d'a/lilude. Les alignements structurauxformes dans le cycle geo(ectonique bresi/ien ont ete reaclives durantle Neozolque. Ces a/ignements, alnsi que les oscillations climatiques, ont contr6/e les principa/es caracterisli ques du pays age knrstique et permettent une reconstitution de /'histoire morpho-genet/que depuis /e Tertiaire et jusqu 'au Ho/o-. Mots-ells: geomorpho/ogie, karst tropical, carbonates precambriens, litlrostraligrapllie, morphogenese lcorstique, variations climatiques quaternaires. Introduction The tropical karst region of Lagoa Santa is located approximately 30 km north ofBelo Horizonte, the capital of Minas Gerais state. The karst region is situated on a mountain plateau in the northeastern part of Lagoa Santa urban area, developed over an interfluvial block at altitudes between 650 and 850 m above sea level (Fig. 1 ). To date, it is the oldest karst area studied in Brazil, comprising-karstological, paleontological, corresponding author. E-mail 0 2002, Editura Acadcmici Romine. All riahts reserved. archaeological 'and speleological interests. This JWst is out standing due to itS extension, economical, and ecological im-portance. The Rio das Ve/has is the main river that drains the region of Lagoa Santa. It is situated at the eastern side of the karstic block and represents the base-level (650 m) of the karst aquifer. With a flow rate of 206 000 m3/sec it is also one of the largest tributaries of the Slio Francisco River. Towards the south and the west the block is bordered by the smaller stream


94 Ribeiriio da Mota, whereas the Serra de Santa Helena maiics its northern border. The whole block is tilted towards the east. Consequently the superficial drainage follows this direction (e.g. Samambaia, Mocambo, Bebida creeks), with a flow rate around I 0 m3/sec. AULER {1994) has measured in the Mocambo basin flow rates of 1.55 m3/sec during the dry season (October, 1993) and 2.8 m3/sec i!l the rainy season (February, 1994), with tern-. peratures varying between 21.9 C (July) and 23.9 C {February). Amounts of dissolved calcium carbonate for these creeks are, according to KOHLER (1989), between 127 and 165 ppm, whereas the pH-values oscillate between 7.2 and 7.8. The climate recorded during the 1970 decade (CoPAER, 1980), showed an average pluviometric index of I ,381 mm, with a daily maximum of 162 mm and totaling 105 rainy days/year. January and August were the most rainy and driest months, respectively, however only 20% of rains have had durations of more than four hours. The average annual temperature was 20.8 C with an average maxima and minima of 27.2 C and. 15.4 C, respectively. The natural vegetation on top and around the karst terrain is represented by deciduous and semi-deciduous forests, while hydrophile and hygrophile vegetation predominate in permanent and temporary lagoons. In places with a covered karst a mosaic-like cerrado (savana-like vegetation) is ubiquitous. Geologically the region belongs to the southern portion of the Silo Francisco Craton, composed of low grade metamorphic pelites and calcitic/siliceous limestones of Precambrian age (Upper Proterozoic). The aim of the present study is to analyze morphologically and morphogenetically the features of the exokarst scenery, in order to elucidate the morphodynamic evolution of the area." Geologic settings The major geotectonic feature in central-eastern Brazil is represented by the Silo Francisco Craton (ALMEIDA et a/., 1977) with its adjoining fold belts, belonging to the Mantiqueira Province (ScHOBBENHAUS eta/., 1984). Initial deposition of the Proterozoic in this area of Brazil, is related to an extensive form, which has been stabilized during the end of the Transamazonic cycle(2 Ga). This platform have suffered taphrogenesis due to crustal extension around 1.8 Oa ago. The crustal rifting processes gave rise to an intense acid to intermediate volcanism, dated in the. Atlantic shield (e.g. tween 1770-1710 Ma (U/Pb ). Between 1000 and 900 Ma (KARFVNKEL & HoPPE, 1988; PEDROSA-SOARES et a/., 2000) the craton went through important upheaval movements, with dramatic climate changes, leading to a continental glaciation: the Siio Francisco Glaciation (KARFUNKEL et a/., 2000). Much evidence of this period has been from the Upper Proterozoic fold belts around the craton during the last 30 years (e.g. KARFUNKEL & C H. Kohler & J. Karfunkel KARFUNKEL, 1977; HErnCH, 1977; KARFUNKEL & HOPPE, 1988). Negative diastrophic crustal movements of the area (-750 Ma) resulted in a new climatic change and the forma:. tion of a large epicratonic sea, in which the psamo-pelitic carbonatic sediments of the Bambui Group were The last large geotectonic cycle in central-eastern Brazil, the Brasiliano cycle (-650-550 Ma) affected the Proterozoic uri its in the Atlantic Shield by tectonic and metam.orphic changes, mainly at craton borders and in the fold belts. Sediments of the Bambui Group suffered such processes at the southeastern part of the craton (e.g: the Lagoa Santa region), whereas sediments from central parts of the epicratonic Bambui-sea show little tectonic and metamorphic changes. Stratigraphic subdivisions jn the Lagoa Santa region, at the south-eastern border of the Silo Francisco Craton, are based on pioneering works of BRANCO & COSTA (1961 ), BARBOSA ( 1965) and BRAUN ( 1968). Many difficulties occured in estab lishing a stratigraphic subdivision valid for the .whole area; they were mainly caused by facies interfingering in a large region with heterogeneous depositional conditions. Therefore, in the present paper, we use themodified subdivision suggested by ScHOLL (1972), DARDENNE GROSSI SAD & QUADE ( 1985), and TuLLER et a/. ( 1992). According to DARDENNE ( 1978) the Bambui Group shows at its basal part a paraconglomerate (Carancas Formation), covered by a carbonate-mud stone sequence (Sete Lagoas, Santa Helena and Lagoa do Jacare formations). The Carancas Formation is not an equivalent ofthe older glacial deposits of the Silo Francisco Glaciation; this formation rather represents subaquatic, reworking levels and aquatic detritus flux formed in paleodepressions of the cratonic basement, then covered by uniform marine seGROSSI SAD & QuADE (1985) established three formations in the Lagoa Santa region: Vespasiano, Sete Lagoas and Santa Helena. The first is composed of the Carancas Formation, oc curring very locally as relicts and of the Pedro Leopolda facies of ScHOLL (1972), designated by TuLLER eta/. (1992) as the Lagoa Santa Member. All these terms and names show clearly, that there is only partial agreement between the authors, concerning the nomenclature in the Bambui sequence. The carbonatic rocks in the Lagoa Santa region, which underwent low-grade metamorphism, are subhorizontal or slightly tilted (5-10 E). Two litho-types can be distinguished: (i)siliceous limestone at the base, with 75-90% CaC03 <7% Si02 and 0.5-3% MgO, (ii) calcitic limestone at the top, with 94-99% CaC03 <6% Si02 and

The Qttatenwry morplwgene.ri.r of the Lagoa Santa tropical karst traced easily on airborne images and photographs by doline (sinkhole) alignments, or by straight, narrow streams, creeks and caverns (KoHLER, 1989). The karst The exokarst scenery of the Lagoa Santa region is character ized by closed depressions (e.g. dolines, uvalas, poljes), that transform in temporary lagoons (Fig. l) when the karst aqui fer level emerges as a consequence of the reloading during maximum rainfall periods (summer). Thus it can be classified as a typical karst of intermil/entlagoons. The geologic map (I :500 000) displays the Lagoa Santa karst region as an enclosed window incised in the Precambrian metapelites. Outstanding is a narrow mountain range, with a NW-SE direction, representing a remnant of the "Sui Ameri cana" surface (KING, 1956), as well as the anomalous course of the C6rrego da Mata (Mata creek), flowing to the south, in . : . : ..... . -.Jjj;;:.' -"._ .,.,, 95 contrast with the general direction of the hydrographic net work. Analyzing the region on a larger scale ( l :50 000), four geo morphologic compartments have been identified (from SW towards the NE): (i) a topographically higher area, up to 850 mat the Serra do Ferradores (covered karst with remnant lateritic de tritus, covered by allochtonous soils}, containing a flu viokarst with blind valleys and do lines (Palmeira-Mo cambo); (ii) a semicircular belt comprising dolines and uvalas (ex:tending from Confins to Macambeiro); (iii) a lower elevation plateau containing a complex system of asymmetric do lines of different ages, aligned along ancient, Precambrian structural trends (NW-SE and NE SW), expressed by straight segments of abrupt walls. Fig. I View of Lhe Cerca Grande massif during Lhe (a) summer, and (b) winter. Vue dtt ma.r.rif de Cerca Grande durant /'ete (a) etl'hiver (b).


96 The different ages are marked by twin-do lines inside a larger, older do line. This compartment incorporates the Samambaia, Macacos, Bau and Cerca Grande massifs; (iv) the lowermost compartment at approximately 650 m with many poljes and ponors, is situated next to the main river, the Rio das Velhas. At a scale of 1: 1 000 horizontal and vertical karren (Schichtenkarren and Rund-Rillenkarren, respectively) at the walls of the massifs can be seen (Fig. 2). As a consequence of dissolu tion dynamics by pluvial water, the karren occur in a major concentration on the upper portion of the high walls (>20m) together with some oblique joints. The morphogenetic features have been imposed by sedimentary facies, hence mineralogically the depressions (e.g. do lines, uvalas) represent areas of calcitic limestones, while topographic posi tive relief areas (e.g. humes, towers) are composed of siliceous carbonate rocks. Cavern systems (endokarst) develop at the bottom of the walls near the actual doline and ponor level. Several chambers are coated with speleothems that can fill, in some cases, ancient conduits. Under the stalagmitic floor re mains of extinct mega-fauna have been discovered. Karst n1orphogenesis The karstic landscape of the Lagoa Santa region developed mainly during the Quaternary; however its origin is remote and probably related to the breakdown of Gondwana during the Late Jurassic Early Cretaceous, when extensional crustal dynamics gave rise to the opening of the South Atlantic (Me sozoic reactivation according to ALMEIDA, 1977). On top of the later formed Sui Americana continental plat form the drainage system had been reorganized and the actual Rio Sao Francisco basin formed. This basin fonned mainly on outcrops of Upper Proterozoic metasediments ofthe Bambui Group. After a long period of apparent stability, probably during rvtid Tertiary, a large surface (etchplain), the Sui Americana Sur face ("master surface" of KING, 1956), was formed at an ac tual topographic level of approximately 850 m, was formed. A thick red lateritic soil developed as the consequence of a hot and humid climate. Within carbonate units under the pelites of the Bambui Group (< 830 m actual topography) the endokarst developed a cave system filled by karst water. Evidence points towards corro sion conduits that formed under hydraulic pressure. Accord ing to Boau ( 1978) endokarstic corrosion processes start when the water reaches the soluble rock. Analysis of the present-day Lagoa Santa drainage system (Fig. 3) points towards an important anomaly. The general of the hydrographic system with its largest river, the Rio das C. H. Kohler & J., Karfunkel Fig. 2 Schichtkarren and Rund-Rillenkarren at the wall of the Bau massif. The cactus (mandacaru) is a reminiscent of a dry savanna climate. Schichtkarren et Rund-Rillenkarren sur Ia parois du massif de Ba1i. Le cactus (mandacarzi) est un relique du c/imal de savane sec. Velhas and its tributaries, is northwards. Solely one of its tribu tary, the Ribeiriio da Mata (Mata Creek) flows in an almost opposite direction, towards the southeast. This anomaly sug gests that the Rio das Velhas once occupied the actual course of the Ribeirao da Mata, flowing at that time in a northwestern direction. Geomorphic evidence supporting this hypothesis are: anomalous hydrographic flow ofthe Ribeirao da Mata; the width of over 100 m of the Ribeirao da Mata (bed) is incompatible with its present flow rate; . the Rio das Velhas shows, after its confluence with da Mata, an undulatory course with many rap ids and waterfalls; gravel of fluvial origin at 850-900 m; Quaternary terraces;


I TOPOGRAPHY +845 Spot eleva/Jon JiQQ, Contour line every 50m 2 EXOKARSTIC FORMS Ooline Ooline w11h wall Uvolo PoiJje HU/116 Slope with "worts" Tower Korslic wall (<20m >20m/ Blocks 3 ENDOKARSTIC FORMS [[] [I] Cove Rock sheller 4 HIDROGRAPHY Perennial water flow Tempory water flow Abandoned meander Swamp Ponor Spnng Probable .subterranean water flow Main d1reclion of the subterranean water flow Lake 8 STRUCTURAL ELEMENTS ,_, rn Inferred lineations Oip 5 NO KARSTIC LANDFORMS 9 MAN MADE STRUCTURES E=?.:!J Terc1ory surface (South American surface) 17-AJ Co/uvial cone A/uvial con e '""'' J Gully 6 SUPERFICIAL FORMATIONS Lateritic so/Is yellowred, deep, slltic clayish] dury crust, defrifJc povements(cerrodO/.> 850m Lateritic so1/s, red, deep, ( c errodo) B00-850m Lateritic s01/s, dark red,deep,c!oy;sh, (cerrodo); outcrop rocks,tforesfJ. 700m LoferJIJc soils, yellow-red, low, delrilic J pavements, gullies, tcerrodo). < 700m 7 LITOLOGY J_r-J ldJ 00 [[ill Main rood Barrage Minmg exploration Main city Airport WATER RECHARGE WATER DISCHARGE Serra de Sonia Helena Formation (fililes) Sete Logoos Formation (limestone) Vesposiono Formation (colciofilite) Archaean (gram tic-gnoJssicJ J BAMBU{ GROUP NEOPROTEROZOIC NO NM\frG .. mea" declination, 1976 SCALE 500m 0 1000 KARST SCENERY OF THE LAGOA SANTA REGION-MG-BRAZIL CARTOGRAPHIC BASE: KOHLER, H. Ch. et ani. Cart du Karst Pedro Leopolda Logoo Santo. caen. CNRS,I97B. A roo---CANYONS HEIGHT WALLS Serra dos Ferrodores ...-----------,-...,----POLIJE ------, cl D NNE METOOOLOGY: Comission tltts phtfnomintJs Korstiqus du Comitt!Nolionol d Gt!ogrophi, Poria, 1965. (Cidapl
PAGE 100

Tile QIUUernary morp/wt:ene.ti.f of the Lt1gou Santa tropical karsl fluvial gravels inside the endokarst; occurrence of fluvial gold, similar to that which occur in the Rio das Velhas terraces, inside the endokarst. The origin of the gold can be attributed only to the gold deposits derived from the Quadrilatero Ferrifero; the surface is tilted in a southeastern. direction, while the strata dip towards the northeast; regional evidence for neotectonic processes. According to voN RrormornN ( 1886) the age of a river is con temporaneous with the massif which accommodates its springs and tributaries. The Rio das Velhas originates in the Serra de Ouro Preto and reaches the locality of the Ribeirno da Mata in the Lag0a Santa region, only after passing through several sec tors of the Quadrilatero Ferrifero (Iron Quadrangle). As a sequence of a long period of morphostructural evolution of this massif, the Rio das Velhas has been dislocated. Its actual bed, due to superimposition, lies inside old structural align ments (VON FREVDERG, 1932). The Rio Paraopeba, a tributary of the large Silo Francisco River, located upstream to the confluence of the latter with the Rio das Velhas, passes the extension ofthe Serra do Curral (north east of the Quadrilatero Ferrifero) and the Fecho do Funil in a narrow and straight, ravine-like course, without waterfalls and chutes. The only apparent anomalous behavior displays the Rio das Velhas, that shows, after its confluence with the Ribeirno da Mata,juvenile features, with dozens of waterfalls and rapids, down to the Silo Francisco River at Barra do Suarruf. During the Pleistocene, when the denudation cycles of the Rio das Vel has dissected the Sui Americana surface, this river de posited gravel, found today at topographic levels of850-900 m on top of the carbonates and metapelites. During this period the Rio das Velhas, after passing the Serra do Curral, continued towards northwest, along the structural alignment between the basement and the rocks of the Bambui Group-a course that is currently occupied by the Ribeirno da Mata bed. When the dissection reached the limestone level, today at 820 m, it turned into a karstic river. The passage of the Rio das Vel has through the Lagoa Santa endokarst is shown by the presence of well rounded gravels of quartz and quartzite inside cavern speleothems of the Porros region in Matozinhos (KouLER et a/., 1976). Connections with the Paraopeba basin can be ob served in the Sete Lagoas area, in the Gruta Rei do Mato (Rei do Mato cave), situated on the actual water-divide between Velhas and Paraopeba rivers. PAULA Couro ( 1975) affirms the probable existence of"a very good synchronization between the fossil mammal fauna in the caverns of Minos Gerais and Siio Paulo and the Ludien.fe Argentino of Upper Pleistocene age". This affirntation translates in time the beginning of the incision and the consequent denudation of the interfluvial block towards the Mid Pleistocene. In the Upper Pleistocene the 97 denudation and corrosi9n processes have already open the caverns, which were subsequently occupied by the Ludinense fauna. It is important to remember that the EuroJ?e80 quaterna rists have established a magnetic inversion and the beginning of an imP<>mint glaciation during the Mid Pleistocene, at ap proximately 700 ka BP. The evolution of this karst region can be explained simply, if we admit a much higher flow rate at that time, than that of today, which is around 10 m'/sec. This would satisfactory ex plain the paleokarst processes that were responsible for the configuration of the actual landscape. In short, it can be postulated, that the Rio das Velhas, after crossing the Quadrilatero Ferrifero, sought to adapt its course to the easier passage towards the north. Its course passed the lithological contact basement/Bambui and reached the karstic region in the vicinity of Matozinhos, entering into the under ground. The Rio das Velhas, allochtone to the karst, transports its wa ter at 200 m,/sec through a labyrinth of endokarstic channels. Along its route it floods the caves and deposits clay and grav els; in addition it corrodes the walls of conduits, widens pas sages and removes calcium carbonate These widened con duits and passages can form collapse dolines, creating a flu.viokarst scenario. Archaeologists (e.g. PROUS, 1978) describe the clay as "reddish" In the reddish clay beneath the stalag mitic level, fossil bones have been discovered. Gravel was found cemented in the roof of passages (e.g. at Balet and Curral de Pedras) Finally the watercourse reappears as a tributary of the Rio Paraopeba. The lack of bauxite in karst can be ex plained by its age (-700 ka BP), although the overlying lithol ogy is composed of rocks rich in aluminium phyllosilicates. Concluding remarks Structural alignments, consequence of the Brasiliano cycle, that were reactivated during Cenozoic time, control the main trends of the karstic landscape of the Lagoa Santa region The superimposed effects of tectonics and climate over the carbonatic lithologies through time, left evidence that allowed the reconstruction of its structural history. The morphogenetic events in a preliminary chronological ap proximation can be described, according to KoutER & Pu..6 ( 1991) in the following way: Tertiary-the interfluvial block was peneplaned by the Sui Americana surface during a warm/humid cli mate. Thick laterite soils were developed ; Pliocene/Pleistocene dissection of the Sui Ameri cana surface due to different Velhas cycles. Alternat ing climate between cold/dry and warm/humid;

PAGE 101

98 Mid Pleistocene the denudation cycle Velhas has rea ched the carbonatic rocks level creating the upper passages of the endokarst. Alternating climates have triggered a superficial karstification; Late Pleistocene (>22 ka BP) strong incision of the topography. Formation of built-in surfaces with doline levels between I 0 and 12 m above the actual level during a warm/humid climate. The maximum corrosion was recorded between t 3 and 22 ka BP and speleothems were precipitated into the cave passages during a cold/dry climate. Holocene-maximum incision during climatic crises between 3000 and 5000 years BP. Periodic inundation ofplains and formation ofdolines and uvalas. References Ar.MHIOA, F. F. M., de ( 1977) 0 Craton do Silo Francilico. Rev. Bra.t. Geoc., 1, 4, pp. 349-364. A ut.F.H, A. ( 1994) Hydrological and Hydrochemical characteri'Ultion oj the Matozinlws -Pedro Leopoldo, Brazil. MSc. Thesis. Dept. Geography and Geology, Kentucky Univ., Bowling Green, IICl p. (unpublished). BARBOSA, 0. (1965) Scric Bambui. Simp. das formayOes Eodo Brazil. XIX Congr. Brct.t. Geol., SBG, Rio de J unci ro, B mzil. Boou, A. ( 1978) Kar.tthydrographie und plly.ti.tclle .tpeleologie. Springer, Berlin. 279 p. BRANCO, J. J. R. & .. ,.A. M. T. (1961) Rotciro pam cxcurliiio Bclo Horizonte -ao 14. Congr. Bra.t. Geol. Puhl . 15, UFMG. Belo Hori1.ontc, 25 p. B HAUN, 0. p G. ( 1968) Contribuiyito a do Grupo Bambuf. Anai.t 22. Congr. Bm.t. Geol . Bclo Horizontc, pp.I5S-166. Coi'AHH, B.H. ( 1980) A con.ttnu;t1o do aeroportn e tl prolet;llo do meio ambiente e o.t valores culturai.t. Belo Hori1.onte (Brazil), Ministcrio da Aeronautica, 288 p. DARDENNH, M.A. (1978) Sfntese sobre a cstratigrafia do Grupo Bambui no it cenlral. Anai.t 30. Congr. Brat. Geol. ,l, SBG., Recife. pp. 597-610 F'RHYRHRG, n., voN (1932) Ergebnisse geologischcr Forschungen in Minas Gerais, Brasilien. N. Jb. Min. Geol. Palaeont., Sonderbd. 2, Stuttgart, 403 p. SAn, J. H. & QuAoo, H. ( 1985) Reviliiio do Grupo Bambuf (Bioco Oriental) em Minas Gerais. Anai.t do Ill Simp. de Geol. De Mina. t Belo Horizonte, pp.68-83. Thus the successive karstic features are not interpreted as the result of single cyclic events, but the consequence of complex Quaternary climatic oscillations which favored changes ofthe flow rates and chemical characteristics of the waters. The scenario of the phenomena over interfluvial block Ribeirlo da Mata-Rio das Velhas represents a remark able example of initiation, genesis and dynamics of an inter tropical karsta true case-study for interfluvial surfaces from other regions. Acknowledgements This study was partly supported by the Consetho Nacional de desenvolvimento Cientffico e Tecnol6gico (CNPq), as well as by the de Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG). We thank 0.8. Hoover for reviewing the final manuscript. HHTncH, M. '(1977) A glaciayilo Proteroz6ica no centro-norte de Minas Gerais. Rev. Bra.t. Geoc., 7, pp. 87-101. KARFUNKI'J., B. & KARI'UNKI!I., J. (1977) Faziclle Entwicklung dcr mittleren mit bc.wndercr Berilckliichtigung des Tillit-Problcms
PAGE 102

The Quaternary morphngene.d.t nf the IAxna Santa PROUS, A. ( 1978} areve hilil6rico das pe1KJUili8li liObre 0 homeni na regiilo de Lagoa Sa'!taiMinu Gentili. De Geogr., USP, Col6quio lnterdisciplinor Franco Bra.dleirn, 2, Silo Paulo, pp. 21-27. RICKrnOFJ:N, F., vON ( 1886} Falver jllr'r!i.fl!nde. Anleitung zu Beobochtungen Uber Gegen.ttilnde der phy.ti.'fCher Geographie und Geologie. Oppenheimer, Berlin. 734 p. 99 SCHOURF.NHAUS, c .. D. A DER7..F., G. R & ASMUS, H. E. (1984} Geologia do Brasil. Bra.o;i_ lia, 501 p. ScHOu., W. U. ( 1972} Der liUdwCiitliche Randbereich der Zone, Minllli Gerais, Bruilien. Geol. Rdtch., 61, pp. 201-216. Tuu.EK, M.P RIRiliRo, J. H., & DANDERFEt:R, A. ( 1992} Mapeamenlo geolOgico da area do Prnjelo VIDA. CPRM. Belo Horizonte. 6 August 2002 18 November 2002 ocapted: 18 November 1002

PAGE 104

Theoretical and Applied IS (2002), pp. 101-110 Small mammals of the cave sites in the Baikalian region Margarita Erbajeva *, Fedora Khenzykhenova, & Nadja Alexeeva Geologicalln.ftitute, Siberian Branch, Russian Academy of Sciences, 6a Sahianova Str., 670047 Ulan-Ude, Ru.fsia. Abstract This paper deals with the small mammal assemblages from the cave sites of the Baikal ian region. In the region there are more than 300 caves and shelters (FtuPPOV, I 993a, I 993b), some of which contain numerous small mammal fossils. The oldest fauna is the Middle Miocene from Aya containing Eurolagw, llelerosminlhrts and Gobicricelodon. In other caves, more than 30 taxa of insectivores, Jagomorphs and rodents were discovered, mainly Pleistocene and Holocene age. Most ofthe small mammals belong to modem species. which inhabit the region nowadays; however, specimens of I.Agurus lag11n1s and Dicrostonyx sp., whose area of distribution lies outside of the region currently, were nlso found. Key words: Mammalia. rodents,Jagomorphs, caves, Baikalian region,l'rebaikalia. Tmnsbaikalia. Miocene, Late Pleistocene, Holocene. Les mlcroniammlferes des sites speltilques de Is region Bsi1csl/enne Resume Dans ce travail on fall une presentation des ossociolions de micromommifires des grolles de Ia region Bofkaliene II y a icl plus de JOO grotles el obris-sous-roche (Filippov, 199Jo; 199Jb), doni quelques-uns conliennent de nombreux restes fossiles de micromammifires. La faune Ia plus anciemre est d'age Miocene !Jioyen, provienl de Ia grolle d'Ayo el inclul Eurolagus; Hcterosminthus el Gobicricctodon. Dons d'oulres grolles, plus de JO taxons d'inseclivores, logomorplres et rongeurs onl ete de couverls, Ia pluporl elonl d 'age Pleistocene superieur et Holocene. La mojorile des micromommiflres son I des especes recentes, qui peuvent eire trouvles ocluel/ement dons Ia region, mois des exemploires de Lagurus lagurus el Dicrostonyx sp. dontl'oire de distribution est moJntenont plus_ reslreinte onl elllgolemenl identifies Moa-dis: Mammalia, rongeurs, lagomorphes. grolles, region Bofkoliene, Prebofko/ia, Tronsbolko/ia, Miocene, Pleistocene superieur. Holocene Introduction The Baikalian region is located in the center of Asia. The terri tory stretches from South to North between 47 and 58N; from West to East it extends between 102 and ll4E (Fig. 1). It is situated on south-eastern borderland of the Siberian Platform, one of the stable blocks of the Asiatic continent and a part of the Central-Asian folded belt Conned during the long geological evolution of Palcoasian ocean. The region is spread within the Central-Asian mountain belt and it is characterized by the alter nation of large, detennincd relief forms. The stud ied area includes the territory of south Eastern Siberia - corresponding author E-mail address: crbl\ 0 2002, Editura Acadcmici Romtnc. All rights reserved. Prcbaikalia and Western Transbaikalia which has more than 70 recorded caves and shelters, the most of them being karst caves, distributed unevenly across the territory in a variety of climate, geologic, physiographic, and hydrologic regimes. Their was provided by FtuPPOv ( 1993a, 1993b ). Systematic studies of the caves had been carried out at the end of XIXth and beginning of the XXth Century. Geolo gists P. Kropotkin and I. Tschersky and archaeologists B. Petri, G. Debets, G. Vologodskyi and others were the early investigators that reported the first data on the caves from the western shore of Lake Baikal. Around the 80's of the last century detailed, multidisciplinary studies of the most caves of the Prebaikalian National Park:, located on the western coast

PAGE 105

102 56 52 .............. ............... ....... ...... ....................... .................... 0 l2011m ..._ __ __, M. Erbajeva et al. t06 116 Russia. 106 Fig I Sketch map of the Baikal ian region showing the location of cave sites. Carte de Ia region Bai"kalienne avec Ia .duwtion des .fite.f etudiee.f. I. Uzurskaya; 2 Shamanskaya; 3. Boro Khukhan; 4. Khurganskaya; 5. Tonta; 6. Bolshaya Baidinskaya; 7 Sluchainaya; 8. Aya; 9. Kurtun I ; I 0. Kurtinskaya; II Kadilinskaya; 12. Kulcitovaya; 13. Brckchievuya; 14. Unyllikayu. of Lake Baikal were provided by different specialists spe Jeologists, geologistS, paleontologists, biologists, climatolo gists, archaeologists etc (GoRYUNOVA et al., 1996). At that time many fossil-bearing caves were discovered in the region and the first remains of small mammal were obtained here. The fossil remains are not well preserved, there are some frag ments of lower jaw and maxilla with different numbers of teeth or without teeth. The main materials consist of isolated teeth and fragments of the postcranial skeleton. Probably they originate from owl pellets; some of these may come from animals which died in the cave itself. As usually, the bone deposits of most caves are subject of the activities of preda tors, either animals or birds, or both. The faunal data have provided significant information on biotic and climatic changes that have occurred in the area from the late Pleistocene to the present. Studied fossils were collected mainly by the geolo gist and speleologist Dr. A Filippov and his colleagues, and in part by F. Ktienzykhenova. All small mammal materials (rodents and Iagomorphs) arc stored at the Geological Institute in Ulan-Ude. Caves and small mammals assemblages The paper deals with the small mammalian faunas of some caves located on Olkhon Island (sites Uzurskaya, Shaman skaya, Boro Khukhan) and on the west coast of Jake Baikal, mainly in its mid and southern part (Khurganskaya, Tonla, Bol'shaya Baidinskaya, Sluchainaya, Aya, Kurtun-1,

PAGE 106

Small mammals ill the Baikalian regio11 Kurlinskaya, Kadilinskaya), as well as Brekchievaya and Unylskaya in the Sayan region and the only one cave in Transbaikalia (Kalcilovaya) (Fig.l ). The oldest site with small mammal faunas is the unique Miocene cave Aya in Eastern Siberia. It is situated on the Priolkhon Plateau, more than 200 m above of Lake Baikal surface in the Aya Bay, over 30 km southwest of Olkon Is land. It is a phreatic cave developed in graphite marbles of Upper Archean-Early Proterozoic age. The length and depth of this cave are 550 m and 40 m, respectively. At this site eight layers of Late Cenozoic deposits were describc<:J by EROAJEVA & FlLJPPOV (1997). The uppermost two layers are Late Pleistocene in age (based on the analysis of sediments and faunas); the later include fishes, amphibians, bats and rodents (Microtus sp. ). In the next, Middle Pliocene, redeposited dark-brown soil, contain ing grit and rock debris grouting with black loam cement, remains of fishes, bats and undetermined small mammals were collected. Abundant fossils, including turtles, amphibians, birds, fishes and were found in layers 4, 5 and 6 (FILJPPOV et al., 2000). The small mammal fauna is not numerous and it includes: Eurolagus cf. .fontannesi (DEPERET, 1887), Hetero mzinthus erhajevae LOPATIN, 2001, Gohicricetodmz sp. nov. SEN & ERRAJEVA (in press), Insectivora gen. indcl. In this fauna Jago morphs arc a dominant group. In the lowermost three layers (7 and 8) no fossil remains were found. The next group of caves arc known on Olkhon Island: Uzurskaya, Shamnnskayn and Uoro Khukhan. The first is located on the north-eastern part of the island, ncar the vil lage of Uzur. The cave is developed also in graphite marbles of Archean-Early Proterozoic age. Its length, width and depth arc 5 x 1 .5 x 2 m, respectively and it is filled with grey sandy loam 30 em-thick. The next small mammals recognized here were : Lepus sp. Ocllotmu1 cf. hyperhorea PALL. Spermophilus sp., Cricetulus sp. Alticola sp., Lugurus lagurus PALL. The fauna includes both forest and sreppe inhabitants However, it is characterized by the predominance of the steppe inhabit ants among which Lagurus lagurus was the most abundant fonn (more than 50% of the total population). The sites and Boro Khukhan arc located in the middle area of Olkhon Island. Fossil bearing layers arc rather deep in the first cave (around 1.1 m) and shallow at the sec ond site (0.2 m). Their faunas are similar and include com mon taxa: Lepus timidus L., Spermophilus undulatus PALL., Cricetulus barahensis PALL., Clethrionomys cf. rutilus PALL., Alticola sp. Micro/u.s oeconomus PALL., Microtus cf .fortis BuECHNE R They belong to modern species that inhabit the region at the present time. However fauna of Shamanskaya cave, in addition to the above-mentioned taxa, contains Clethrionmnys ru.focanus SUNDEVALL and Laguru.s logurtts PALL. The Jailer species docs not exist in the recent fauna of the region. 103 Small mammal associations of these two sites and quantita tive ratio of taxa show that open landscapes were distributed rather widely; however forest biotopes were dominant. The other group of studied caves are located mainly on the western coast of Lake Baikal, along Primorskyi Mountain, while one cave, Kalcitovaya, is located on the eastern coast of Lake Baikal, in the Trans baikal area." They are all devel oped as well in graphite marbles of Archean-Early Protero zoic age. Two caves (Unylskaya and Brckchievaya) are situ ated in East Sayan mountainous region Fossil remains from these caves were collected from differ ent layers reffcred to the Late Pleistocene and Holocene but we shall mainly discuss the Late Pleistocene small mammals. The number of fossil remains in each cave differs very much. Huge number of specimens (more than 900 specimens from three layers) were collected in Kurtun-1 site; around 500 specimens in both Kurtinskaya (two layers) and Tonta (five layers) caves, 25 -in Bol'shaya Baidinskaya, 42-in Sluchainnya, 70-in Khurganskaya (two layers) and olh" crs (FlLJPPOV et al., 1995). All these caves contain faunas of similar taxa. The list of the small mammal species is given in Table 1. The species composition of the faunas show, that all these caves contain mainly recent taxa that inhabit Prcbaikalia nowa days except for Lagurus lagurus and Dicrostonyx sp. During the late Pleistocene and early Holocene, LL1gurus lagurus was widely distributed in the Baikalian region, its area of distri bution extending to the south-cast of Eastern Transbaikalia. The nearest modern area of distribution of this species is Mongolia in the south and the steppes of Khakasia and Kazakhstan in the west (GRoMov & ERRAJEVA, 1995). Next species, the arctic lemming Dicrostonyx torquatus PALL. may be found in Eurasia, at ihc present time. only within the tundra and the northern forest tundra zone (GROMOV & ERDAJEVA, 1995) In total, the faunas of these caves demonstrate that in the region a mosaic-type landscape have existed, including forest, tun dra-steppes, steppes, meadows and river valleys. The forest inhabitants (Lepus timidus, Oclwtona hyperhorea, Pteromys volans Sciurus vulgaris, Tamia.s sihiricus, Cletlzri onomy. s rutilus, C. Aiticola argentatus, Lemmus sp., Myopus s c histicolor, Micromys minutu.s and others) were pre dominant groups in the fauna of that lime The typical steppe dwellers such as Lagurus lagurus and Microtus greg a lis, in habitants or the dry meadows -Spermoplzilus undula!IIS, Cricetulus sp., Microtus arvali.s, M. maximoviczi and moist meadows (Microtus oeconomus) had a rather sporadic distri bution The analysis of species composition of the sites Kurlun 1, Brckchievaya and Unylskaya has revealed that the small mammal faunas was a disharmonious one, including both tun dra and steppe dwellers-Dicrostonyx and La gums, the ani mals which inhabit at present time quite different natural zones. The disharmonious faunas were widely distributed

PAGE 107

104 M.Erbajeva et a/. Table I The list of the small mammals from different caves of Lhc Baikalian region La lisle des differents micrOITUimmifores groues de Ia ;egion Bai'/.:alienne. Cayes/Groues: 4: Khurganskaya; 5: Tonta; 6: Bol'shaya Baidinskaya; 7: Sluchainaya; 9: Kurtun-1; 10: Kurtinskaya; II: Kadilinskaya; 12: Kalcitovaya; 13: Brekchicvaya No. Species 4* 5 6 7 9 10 11 12 13 14 lnsectivora 1. Sorex sp. + + + + + + + Lagomorpha 2. Lepus timidus L. + + + + + 3. Lepus sp. + + + + + 4. Ochotona hyperborea PALLAS + sp + + + sp sp sp + Rodentia 5. Pteromys volans L. + + 6. Sc:iurus vulgaris L. + + + 7 T"mias sihiricus (LAXMANN) + + + + + + 8. Spermophilus PALLAS + + sp + + + 9. Marmota sp. + 10. Cricetulus sp. + + + + + + II. Alticola argentatus (SEVERTZOV) + + i2. Alticola sp. + + + + + + + + 13. Clethrionomys rutilus PALLAS + + + + + + + + + 14 . Clethrionomys rufocanus SUNDEV. + + + + + + + + + + 15. Clethrionomys sp. + + + + 16. Dicmstonyx sp. + + + 17. U.1gurus lagurus PALLAS + + 18. Arvicola terrestris L. + + + + + 19. Microtus oeconomus PALLAS + + + + + + + + + 20. Microtus .fortis BUECHNER cf + cf + cf 21. Microtus arvalis PALLAS cf + + 22. Microtus gregalis PALLAS + + + + + + + + 23. Microtus maximoviczi SCHRENK + 24 sp. + + 25. Lemmus sp + + + 26 Myopus schisticolor (LILLIJEBORG) + + 27 Myopus sp. + + + 28. Apodemus sp. + 29 Mus sp. + + + 30. Mic.:romys cf. minutus (PALLAS) + + +

PAGE 108

SmaLl mammal.r in the Baika/ian region lOS during the Pleistocene cold period in the Northern Eurasia Short review of some taxa from caves from the Western to the Prebaikalia. In contrast to Prebaiakalia the Transbaikal area, at that time a vast territory was occupied by dry cold steppes with arid climate where Lasiopodomys brandti, Ochotona daurica, Lagurus lagurus, Marmota sibirica, Allactaga sibirica were the dominant groups among the small mammals. No fossils of collared lemmings were found here ever. 7 mm 1 0 2 5 8 Among small mammals collected in the cayes of the Baikalian region around 30 taxa arc established (fable 1); one form is refferedto insectivores; three-to lagomorphs and other 26 taxa belong to the rodents. Of the rodents, the most abundant and characteristic forms are Clethrionomys rutilus, C. rufo canu.f and Myopus schi.fticolor, as well as Alticola and dif ferent species of the genus Microtus. They all currently exist in modern fauna of the region The most important and sig nificant elements of the cave faunas are Dicrostonyx and Lagurus which are not present any more in the region. 3 6 9 Fig. 2. Alticow argenta/us (SEVERTZOV, 1879) the caves of the Baikal ian region Alticola argcntatus (Si>YERTZOV, 1879) des grol/es de Ia rlgion Bai'kalienne. M1 : I. Kurtinskaya; 3, 6 Uzurskaya ; 4 Tonta; 7 Bolshaya Baidinskaya; 9. Sluchainaya. M1 : 2 Kurtinskaya; 5. Tontai 8 Sluchainaya.

PAGE 109

106 Rodentia BowoJCH, 1821 Cricetidae FISCHER, 1817 Arvicolinae GRAY, 1821 A/tico[a BLANFORD, 1881 Alticola argentatus (SEVER1ZOV, 1879) The remains of Alticola occurred in all caves except Khur ganskaya and Brekchievaya (Table 1). The dental pattern is identical to that of the recent Alticola argentatus (Fig. 2, 1-9) The modem species occupy the rocky areas of the mountain chains of Pamir-Alai and Tian-Shan in Kirgizia and Tadjikis tan, Altai-Sayan mountainous area in Russia, Hangai and Kentei Mountains, and Khubsugul area in Mongolia. In the Fig 3. First lower molar (M1 ) of Clethrionomys from the caves of the Baikalian region. Clethrionomy f rufocanus SUNDF.VAU., 1846: I : Kurtun-1; 2: Tonta; 3: Kadilinskaya; 5: Kurtinskaya Clethrionmpy f rwilu.f PALLAS, 1779: 6,7 : Kurtun-1; 8, 9 : Kurtinskaya. Lower molars (M 1M3 ) of Clethrimiomy.f rtifoc a nus SUNDF.VAI.I 1846 : 4: Khurganskaya Le premier molaire (M1 ) de Clethrionomys des grottes de Ia region Bal'lcalienne. Clethrionomys rufocanus SUNDT>VA/.1. 1846 : 1 : Kurtlln-1; 2: Tonta; 3: Kadilin.fkaya; 5: Kurtinskaya. Clethrionomys rutilus PAUAS, 1779 : 6 ,7: Kurtun-1; H. 9: Kurtinskaya. Molar.f inferieur.f (M,-MJ) de Clethrionomys rufocanus SuNDF.VMJ. 1846 : 4: Khurganskaya. Tuble 2. Dimensions or the molurs of Cletlrrionmn y s from the cuve Kurtun-1 Dimen.fimu des molar.f de Clethrionomys de Ia grotte Kurtun-1. Clethrionomys Clethr ionomys rufocanus rutilus M1 N X lim N X lim Length 106 2.6 2.4-3 .50 64 2.26 2.00-2.50 Width 106 1.15 1.0-1.23 80 0.99 0.85-1.15 . M.Erbajeva eta/. studied area, in Prebaikalia, this species is distributed on the western coast of the Lake Baikal, on the territory of relict Tazheran steppes and on the Olkhon Island (SHVETsov et al:, 1984) Clethriononiys TILESJUS, 1850 Clethrionomy.r rufocanus (SuNDBVALL, 1846) (Fig.3, 1-5; Table 2) Clethrionomys rufocanus is one of the abundant species of the faunas of all caves of the region This species is charac terized by its large size. The molars are relatively broad; the triangles of the masticatory surface are narrow. The antero conid complex of M1 is rather complicated. The anterior mush room-like cap is of asymmetrical type and almost completely separated from T5. 2 3 mm I 0 7 5

PAGE 110

SI1Wll nuunmaLr in the Baikalian region This species inhabits montainous-taiga regions of northern Eurasia, in particular the south-eastern Siberia. Clethrionomys rutilus (PALLAS, 1779) (Fig. 3, 6-9, Table 2) Clethrionomys rutilus PALLAS is next abundant species of the region. Its fossil remains were found in all caves except for Sluchainaya cave. It is a medium size vole. Unpaired anterior loop of M1 has a wide confluence with T5; other posterior loops have connections of different values, which differ this form from c : rufocanus (Fig. 3, 6-9). This species inhabits mainly forest-tundra and forest-steppe zones, as well as shrubs of northern Eurasia from Sweden in the west through the Urals, Kazakhstan, Altai-Sayan moun tain range to Japan in the east and Bering Island in the north east. mm 0 7 0 L-------' I mm z 5 8 107 Dicro.rtonyx GLOGER, 1841 Dicro.rtonyx sp. (Fig. 4, J) Fossil remains of this taxon were discovered in the caves Kurtun-1, Unylskaya and Brekchievaya. They are represented by isolated teeth {M1 and M3 ) of different preservation. Dur ing the Late Pleistocene, Dicrostonyx was widely distributed in the studied region; however it was not found in all known sites. At the present time this species is characterized by a circum polar distribution in the tundra and the northern forest tun dra, far northern than their previous Late Pleistocene inhabit ing area. 6 9 Fig. 4. Molars of lemmings from the caves of the Baikalian region. Molaires de lemmings des groues de Ia rt!gion Bai1ca/ienne. Dicrostonyx sp. I Ml, Kurtun-1. Myopus .rchisticolor (LIIJJEDORG, 1844) 2. M2 3. M3 Kurtun-1; 4. M1 Kurtinskaya; 5, 6, 8, 9. M1 Kurtun-1; 7. M 1 Kurtun-1.

PAGE 111

108 2 5 6 9 10 0'-------' 1 m m Fig. 5. Molars of the voles from the of the Baikalian region Molaires de souris des de Ia region Bai1calienne.M.Erbajeva et al. 3 4 7 M1 : (PALLAs, 1779): /, 2. Kurtun-1; 3. Kurtinskaya; 4. Khurganskaya. lagurus (PALLAS, 1773), 5-8. Uzurskaya. BuECHNER, 1889: II. Shamanskuya, 12. Kadilinskaya. M3 : lagurus (PALLAS, 1773): 9, 10. Uzurskaya.

PAGE 112

Small mammals in the Baikalian region 109 Table 3. Dimensions of the molars of Lagurus lagunLr, extinct and extant. Dimen.rion.r des molaires de Lagums Jagums, eteint et exi.rtant. Dime nsions Caves (mm) Uzurskaya n X M' Length Width Ml Length Width 1 M' Length 2 2.12 Width 2 0.97 M, Length 7 2.84 Width 7 1.07 Mz Length 5 1.64 Width 5 0.94 MJ Length 1 Width 1 Myopu.1 Mli.Lim, 1910 Myopus .KIIisticolor (LILUBRORG, 1844) (Fig. 4, 2-9) lim 1.75 1.10 2.10-2.15 0.95-1.00 2.65-3.15 1.00-1.10 1.50-1 75 0.85-1.00 1.75 0.80 Myopus was widespread in the Late Pleistocene faunas of the region. The dental pattern (Fig. 4, 2-9) and size of extinct forms arc identical to that of the recent forest lem ming. The modem area of distribution occupies vast forest territories from Sweden and Norway to Far East through Si beria and Baikalian region. l.agurus GLOGER, 1841 Lagurus lagurus (PALLAS, 1773) (Fig. 5, 5-10, Table 3) The remains of Lagurus laguru,r in the Prebaikalian caves are not very numerous; they were discovered in the caves Shamanskaya, Uzurskaya, Kurtun-1 and Unylskaya. They are known as well from some archaeological sites (Malta, Krasnyi Yar, etc.) of the region. The occlusal pattern of the molars is typical for the recent Lagurus lagurus (Fig. 5, 5-10). The teeth are rootless, and there is no cement in the re-entrant angles. The dimensions of molars for studied specimens differ from those of recent lagurids, being slightly larger, especially for the specimens from Uzurskaya cave (Table 3). Lagurus laguru.r inhabits plains and steppes from the eastern borders of Europe and the western parts of Asia, through north ern part of Ka7..akhstan, south Siberia to north-west China and Mongolia. During the Late Pleistocene the western part of iLc; range has reached Britain and France, Germany, Ro mania a.o. and the south-eastern part of the Trans baikal area. Shamans k ay a Recent (Kazakhstan) n 1 1 X n X lim 2.2 1.1 1.8 1.1 2.65 42 2.61 2.25-2.90 0.95 42 0.93 0.80-1.00. 38 1.50 1.30-1.65 38 0.84 0.70-1.10 1.65 22 1.62 1.45-1.75 0 65 22 0.70 0.60-0.75 Microtus ScHRANK, 1795 Microtus (PALLAS, 1779) (Fig. 5, 1-4) The fossils of Microtus gregalis (PALLAS, 1779) were discov ered in all caves except for Sluchainaya and Kalcitovaya This species is close to the recent forms by size and dental struc ture, having both archaic and progressive type of M, (Fig 5, 1-4), which is characteristic for the Late Pleistocene voles Its recent area of distribution occupies a vast territory of Northem Eurasia including mou ntain tundra, forest tundra and diffcrcnt steppes. Microtus fortis BUECHNER, 1889 (Fig. 5, II, 12) This is a large size vole, having relatively broad teeth. All triangles are completely separated This species was not so numerous in the faunas; its fossil remains were discovered in five caves (Table 1). By the dental pattern (Fig. 5, II, 12), as well as by its large size, this form is close to the recent spe cies Modem species inhabits marsh -mea dow and flood plain meadow areas of forest and forest-steppe zones of southeast Siberia through Transbaikal area, Amur region to China and Korea. Conclusions The review of the cave faunas show that almost all of them arc Late Plcist:X:ene and Holocene in age They mainly consist of species inhabiting the region nowdays except of Lagurus1aguru.r and Dicro.rtunyx sp. the area of distribution of which is now outside further to the north and south and west. Only the unique Aya cave hosts Early-Middle Miocene faunas.

PAGE 113

110 Acknowledgements We are greatful to professor Costin RMulescu, Silviu Con stantin, Alex Petculescu and the other organizers of the "Theo retical and Applied Karstology" Symposium for the opportu nity to participate in the meeting and to contribute to this vol ume. This study was supported by the Russian Fund of Fun damental Investigation, current grant 01-05-97239, "BaikalRFFI". References ERnAJEVA, M.A. & Frurrov, A.G. (1997) Miocene small mamma lian faunas of the Baikalian region.In: Actes Congris BiochroM'97 (AGUILAR, J.-P LEGENDRH, S., & MICHAUX Eds.), Mem. Trav E.P.H .E., Inst. Montpellier, 21, pp. 249-259 Fn_rrrov A. G. (1993) The Caves of Irkutsk region. Caves, pp. 71-83 (in Rus.rian ). M.E_rbaj!va et al. FnJPI'OV A. G. ( 1993) The Caves of Buryatia region. Caves, pp. 83-93 (in Russian). Fn.rrrov A.G., ERBAJEVA M.A., & KHilNZYKHENOVA F.l. (1995) The application of the Late Cenozoic small mammals of the south East Siberia for stratigraphy 117 p. (in Russian). Frui'I'Ov A.G., ERBAJEVA M .A., & SvcHEVSKAYA E.K (2000) Miocene deposits in Aya Cave near Russian Geology and Geophy.rics, 41, 5, pp. 755-764 (in Russian) GORYUNOVA 0.1., FILII'POV A.G., VE:l'HOV V.M., & BERDNIKOVA N.E. ( 1996) The Caves of the Pribaikalian National park. Archeo logictlllegacy of Baikal Siberia. 1. Irkutsk, pp 101-109. (in Russian). GROMOV I.M. & ERnAJHVA M.A. (1995) The mammals ofRu.r.ria and adjacent territories, lagomorphs and rodents St. Petersburg, 522 p. (in Ru.r.rian). SIIVIITsov J.G SMIRNOV M .N., & M<>NAKHOV G.I. (1984) Mammals of Baikal ba.rin area Nauka Press, Novosibirsk, 258 p (in Ru.r.rian). received: 26 May 2002 revised : j September 2002 accepted : j September 2002

PAGE 114

Theoretical and Applied Karstology. 15 (2002), pp. 111-116 Marles genus representatives in the Wurmian of Romania Emanoil & Razvan Arghir2 I The "Emil Racovilli" Institute ofSpeleology. str. Frumoasii II, 78114 Bucurqti, Romania. 2 Bucharest University, Dept. of Geology. Bd. N. Blilcescu I, 70111 Bucurqti, Romania. Abstract Remains of various mustelides (Maries martes, Martesfoina, Mustela nivalis, Putorius putorius) haye been recovered from the cave nr. 4 din Scocul Scorotei (Retezat Mountains, Huncdoara county, Romania). This new fossil site of the Sou them Carpathians may be assigned to the Late Pleistocene (Middle WUrmian). Allometrical and morphological study of the skulls and dentition indicate a climate warming that corresponds to the lower sediment layers (3 and 2), followed bX a (;ooling that corresponds to the upper layer, as well as the existence, during the Wilrmian of two different types of Maries. Key words: Maries genus, nr. 4 din Scocul Scorotei, Middle WOrmian, Romania Representants du genre Martes dans le Wiirmien de Roumanie Resume Des des plusieurs mustelides (Martes martes, Martes foina, Mustela nivalis, Putorius putorius) ont ete trouves dans Ia (=grotte) no. 4 de Scocul Scorotei (Montagnes de Retezat, departement de Hunedoara, Roumanie). Ce nouveau site des Carpates Meridionales peut etre attribue au Pleistocene superieur (Wiirm moyen). L 'etude allometrique et morpho/ogique des cranes et de Ia dentition indique un echauffement correspondant aux couches inferieures (3 et 2), sui vi d 'un refroidissement correspondant a Ia couche superieure (/), de meme que /'existence a11 cours du Wzlrmien de deux types de Martes. Mots-cles: Martes: nr. 4 din Scocul Scorotei, Wiirm moyen, Roumanie. Introduction nr. 4 din Scocul Scorotei (=Cave no. 4 from Scocul Scorotei, further on named Scocul Scorotei Cave for simplic ity), is located on the outskirts of the calcareous sector of the Retezat Mountains (Western Jiu Valley). The cave lies at an elevation of 1150 m, on the left bank of Scocul Scorotei Valley, c. 150 m above the river bed. At 75 m from the entrance a 23m-deep shaft is located (PoNTA et a/., 1984) (Fig. I). Diggings carried out at the base of this shaft, in 1986, by members of "Hidrocarst" Club in Vulcan town gave access to a lower level of the cave and also yielded numerous fossil bones, including Canis lupus, Vulpes vulpes, corresponding author. E-mail address: 0 2002, Editura Acadcmiei Rom4ne. All rights reserved. Gu/o gulo, Capra ibex carpalhorum, etc. The researches car ried out by us in 1986, 1987 and 2002 allowed the discovery of a large number of fossil remains belonging to the Maries genus. The marten is less referred for the Upper Pleistocene, as the discovered remains are rare. The origins of the two modem species, Maries maries and Maries foina, are unclear even though the Maries genus is known since the Miocene. In Romania remains of the forest marten (Maries maries) were found in the Upper Pleistocene deposits Muierilor (Parling Mountains), in "La Adam" Cave(Central Dobrogea), Cioarei (Valcan Mountains) and M!igura Valley, Apuseni Mts.). The discovery of nine, al most complete, Maries maries skulls and three Maries foina

PAGE 115

112 skulls in nr 4 din Scocul Scorotei, is interesting be cause this is the first time when the rock marten (Martesfoina) was found in Romania. The good conservation state of some essential pieces, like the skulls, have given us the opportunity of making some observations on the character and the affinities of the forms that populated this area during the Upper Pleistocene, more precisely during the Lower WOrm ian interstadial to the Middle WOrm ian (the equivalent ofWUrm H-Ill of the French chronology). Materials and methods Stratigraphic data The excavations carried out in Scocul Scorotei cave, in E2 and AO squares (Fig . ,) reached the level of 75 and 25 em, respectively. In 2002 other three points were sampled, situ ated at the lower level (codes: PSS4-L, PSS4-I ; and PSS4-2). The deposit, relatively homogenous, consists of a fine, dark-0 5 10m 0 2 4m E. $tiuca & R. Arglrir yellow clay (Code 2,5Y 6/4 in the international MuNSSEL 1954). The, most complete stratigraphic succession was encountered in the E2 square (Fig.l.a, b) and it consists, from top to bottom, of: Layer/: 5 em thick, overlaps a calcite flowstone that does not exceed 1...:1.5 em in thickness; Layer 2: 6-7 em thick, includes a secondary calcite crust at tlui bottom, with the same thickness as the pre vious one; Layer 3: visible over a 60 em depth was divided (Fig. I c) into four levels (3a-3d) based on fossil fauna criteria (RADULESCU el a/., 1991 ). In remains belonging to about I? marten individuals were discovered; in this study we only used five skulls with connected mandibles and other four skulls and mandibles for which it could not be established with certainty if they belonged to the same individuals (Fig. 2). Ocm 50 b. a. -:--------= =1 ------------_2 _3a : :::::::::::::::::::::: _3b 3c 3d c. Fig. I Map of Scocul Scorotci cave: a. the lower passage (surveyed by "llidrocarst Vulcan", 1986, unpublished); 2. excavation sites in 1987 ; 3. stratigraphic log at the limit between E I and E2 squares. Carle de Ia grolle de.Scocul Scorotei: a. Ia gal erie inferieure (d'ap1es Ia carte levee par 11 Hidrocarst Vulcam>, 1986, manuscril); 2. sitpation desfouilles en 1987; coupe stratigraphique entre les canes EIIE2.

PAGE 116

Maries genus representatives in the Wurmian of Romania 113 Fig. 2. Martes maries: I 2: mandible (oclusal and lateral view); 3, 4, 5 : skull (ventral, oclusal and dorsal view). Martes martes: 1, 2: mandibule (vue occ/usa l e et latera/e); 3, 4, 5 : crane (vue ventra/e. occ/usa/e et dors a/e). A preliminary statistical analysis that took into account all dis covered fossil remains of Maries martes yielded a non-nor mal with a variability coefficient V > 10. The fos sil material was separated accordingly into two dimensionally homogeneous fonns: a bigger one characteristic to layers 2 and 3 and a smaller one, characteristic to layer I ; the latter is accompanied by Marlesfoina. Statistical analysis For the statistical analysis we also considered, besides the nine above-mentioned specimens, the following: a maxilla fragment from "La Adam" Cave (Central Dobrogea, DuMITREscu eta/., 1963), the skull discovered in MAgura (TERZEA, 1970) 35 skulls of modern forest marten and 12 skulls of modern rock marten from Romania (Southern Carpathians and the Apuseni Mountains). In order to clarify the connections with the modern fauna we also took into consideration the data from the literature re garding Martes maries maries from Europe (ANDERSON 1970) and Martes martes rulhena, a smaller-size subspecies from the European part of Russia (OGNIEV, 1931 ). In the first phase, in order to compare the main cranial and dental characters of the discovered fossil remains with the ones taken from literature and the actual forms, we drew a ratio diagram applying SIMPSON'S method ( 1941) (also SIMPSON et al., 1960) (Fig. 3). In this diagram the horizontal distance be tween two points is proportional to the difference in size be tween two animal groups because the method consists in sub tracting from the logarithm of a parameter characteristic for the com(Jared species th e logarithm of the same parameter for the standard species. The standard species utilized by us was Maries mari es maries from Norway.

PAGE 117

114 0 0 07 0 06 0 05 0.04 0 .03 0.02 O.D1-+ 0 .01 0 02 0 .03 0 04 0 05 0.06 0.07 0.08 0.09 0.10 0.11 1 2. 3. 4. wee ... 5. 6. WM1 LM 1 ....... 8. e = l.ll e ._, eBL, PL, wee, POe, '8 WPP, INB, POP e ... ....... -t: a t: E 1;:1 Q) e-g M. maries maries (Europe ) M. maries ( Romania modem) 0 00939 M. maries rulena (modern) 0.01177 0.01040 M. maries (Europe fossil) 0.01090 0.01812 0.01429 M. maries (Romania fossil+) 0 01450 ct.l2 1 1 0.0 1 258 M maries (Romania fossil-) 0.01460 0.00893 0.01369 M foina (modern) 0 01521 0 02255 0 01886 E.. & R. Fig 3 Ratio diagram for different fossil and modern marten forms from Europe vs. the standard. I. fossil, Europe {MARTIN, 1968; AROANT, 1991; & RAuscH, 1959; 0RLov, 1941)2. postglacial Europe (MILLI!R, 1912); 3. modem, Romania; 4. fossil Romania, large-sized form; 5. fossil Romania, small-sized form; 6. standard, Norway. See the text for abbreviations Diagramme des rapports morphometriques des diverses formes-fossiles et recenles de martres de / Europe par comparaison au standard norvegien. J fossile, Europe (MARTIN, /968; ARGANT. 1991; KuRTEN & RAuscH, 1959; 0RLOV, 1941); 2. postglaciaire, Europe (MILLER, 1912); 3 actuel, Roumanie; 4 fossi/e, Roumanie.forme de grande taille; 5 fossile, Roumanie,for_me de petite laille; 6. standard norvegien. Voir le Iexie pour les abrevia lions Fig. 4. The matrix of reciprocal variation between different species of marten. Matrice des variations reciproques pour les di.fferen-tes especes de mar/res : 1' e ;;; 'iii .E ..,.2 c2 _, .! ""' .! 8. a a -.,:,;x: <._, 0.01732 0.02258 0 01957 0 00616 0.01909 0.02752

PAGE 118

Maries genus representatives in the Wlirmian of Romania Rr I i a. 0.00 I I 0 05 I I 0.00 I 0.01 I b. 115 Ef A Fig. 5. a. The diagram of relative differences for the skulls of Maries maries: b. The diagram of relative differences for the inferior dentition of Maries maries: R< ""the small-sized fossil froin Romania; R =modem form from Romania; R> =the large-sized fossil form from Romania; r =Maries maries rnthena (Asia); E =the modem form from NW Europe; Ef= the fossil form from NW Europe; A= Maries maries from "La Adam" cave (Romania). Diagramme des differences relatives pour les crones de Maries maries (a) et pour Ia dentition inferieure (b). R< = Ia forme fossile de petite tai/le de Rouma11ie; R = Ia forme actuelle de Roumanie; R> Ia forme de grande taille de Roumanie; r = Martes maries ruthena (Asie); E = Ia forme actuelle du NO de /'Europe; Ef= laformefossiledu NO de /'Europe ; A= Maries maries de Ia grolle La Adam" (Rouma11ie) For the skull, the following seven characteristic measurements were taken into account: the condylobasallength (CBL), the palata/length (PL), the width across the canines (WCC), the width across carnassial (WPP), the inter-orbital breadth (INB), the width at postorbital apophyse (POP), and the length of postorbital constriction (POC). Based on these measurements, the relative distances between different types were calculated and the matrix of reciprocal variation was established. 'This matrix (Fig. 4) establishes the Maries Maries Martes maries maries maries ruthena maries (Sweden) (Germany) (CSI) CBL/WCC M M CBL/POC M M Mr+ POP/POC M M WCC/WPP M, CBL/LCMt/ M M LCM/LC4 M M M+ LC4/WMI M M + F LMI/WMt M M M L/W P3 M L/W P4 M LP3/LP4 M differences; or relative distance between zoological or anthropo logical types as a series af nwnerical data proportional to the standard deviation between the analyzed types. Based on the calculated data and using a simple descriptive geometry method we have represented these differences in Fig. 5. Figure 5a shows that the two fossil types from Romania and the fossil type from Europe are at the biggest distance to one another. The modem European marten is closer to the large sized type from Romania and also to the European fossil type. Maries Maries Maries Maries maries maries maries maries Maries Iorenzi (modern (fossil (fossil vet us (CSI) Romania) Romania) Europe) M+ M F F M Mr+ Mr M M M M M M M M M/F M/F M+ F F F M M F F F F M M M M F F M M F+ F/M F F F F F F M Fig. 6 Mosaic distribution of different characters for Maries species(+: strong character;-week character ; M : Martes mart F: Maries /oilra; F/M: M. foina/M. maries intermediary; Mr. Maries ruthena. Distribution en mosal'que di.fferents caracteres des espcces de Maries{+: caractcrc puissante; caracti:rc faiblc; M: Maries maries; F: Maries foina; F/M: intellTlCdiairc M.foina/M. mari es; Mr: Maries ruthena

PAGE 119

116 On the other hand, the recent forest marten from Romania and the Maries maries rulhena are dimensionally close by the two fossil types from Scocul Scorotei: the first one with respect to the small-sized shape, and the second one with respect to that of the large-sized type. For the upper dentition we taken into consideration the following seven parameters: the length and the width PJ, the length P4 the width at principal cuspid P4 the length and the width M 1 and have calculated the distances between the given types as shown above (Fig. 5b). Discussions and conclusions The faunal association from nr:4 din Scocul prove the existence of a forestry environment, which persisted during the sedimentation of layers 3 and 2. Sedimentation of layer one corresponds to a colder stage, which favored the grassland extension against the forest. (RAouLESCU el a/., 1991) The difference between the forest marten and the rock marten remains still a difficult paleonological problem because, un like other species or genra, they both coexisted in the same geographic area and show mixed morphological characters (the so-called mosaic distribution) (Fig. 6). From the graphic representation in Fig. 5 one may notice that the large-sized type from Scocul Scorotei and the smaller one References ANDERSON, E. ( 1970) Quaternary evolution oflhc genus Martcs. Acta Zoo/. Fennica, 130, pp. 1-131. ARGANT, A. ( 1991) Carnivores quartcnaircs de Burgognc. Doc. Lab. Geol. Fac. Sci. Lyon, 115, pp. 1-300. DUMITRESCU, M., SAMSON, P., TERZEA, E., RADULESCU, c., & GHICA, M. (1963) "La Adam", statiune pleistocenA. Trav. lnst. Speol. "Emil Racovitza",l-11, pp. 229-284. -I KuRTEN, B. & RAUSCH, R. (1959) Biometric comparisons between North American and European mammals. Acta Arctica, XI, pp. 1-44. MARTIN, R. ( 1968) Les mammiferes du gisement du quaternaire de Villereversure (Ain). Etude des carnivores, des cervides et equides. Doc. Lab. Geol. Fac. Sci. Lyon, 27, pp. 1-153. MILLER, G S. ( 1912) Catalogue of the Mammals of Western Europe. Collection of the British Museum, Cloves and Son. MUNSELL ( 1954) Soil color charts., Baltimore. E. $tiuca & R. Arghir are relatively distinct one from another even though they be long to the same area. The large-sized type, like the modem European marten shows some affinities with the fossil martert from Europe. Also noticeable is the p.osition quite isolated of marten type from the "La Adam" Cave in Central Dobrogea. The. analysis has also shown that the European forest marten, the large-sized fossil type from Scocul Scorotei and that from "La Adam" Cave, show affinities with the Pleistocene fossil types of Europe, while the small-sized type from Scocul Scorotei is closer to the modern forest marten from Romania. In conclusion, after this first analysis of Maries fauna from Scocul Scorotei we have shown: Two types of Maries martes coexisted during the Upper Pleistocene, more precisely during the interstadial Lower-Middle WUrmian. The large-sized forest marten was replaced by a type with a smaller size and having the appearance of a rock marten, which shows the occurrence of a climU:tic dete rioration at the level of layer I. The similarity between fossils marten and the present marten from Romania with the Russian subspecies Maries maries rulhena, shows a filiation that has in cluded a series of eastern immigrants. OoNmv, S.l. ( 1931) The fauna of USSR _and of neighboring countries. Moscow. (in Russian). 0RLOV, l A. ( 1941) Tertiary Mammalia and localities of the remains. Tertiary Carnivora of West Siberia. Ill Mustelinae. Trav. lnst. Paleont. Acad. Sci. USSR, 3, pp. 30-39 (in Russian). PoNTA, G, ALolcA;G, BAoEScu, D., PANAIOTU, C., & SoLOMON, A. (1984) CercetAri speologice in perimetrul Jiul de But. Speologic, 8, pp. 77-117. R.i.ouLEscu, C., SAMSON, P., & l)TIUCA, E. (1991) Les faunes de mammiferes wurmicns de Ia Grottc No. 4 de Scocul Scorotci (Dep. de Hunedoara, Roumanie). I. lnsectivora, Rodentia. Trav. lnst. Speol. "Emile Racovitza", XXX, pp. 75-37. SIMPSON, G G (1941) Large Pleistocene Felines of North America. American Museum Novitates, pp. 1-27. SIMPsmi, G S., RoE, A., & LEWONTIN, R. C . (1960) Quantitative zoology., New York, 440 p. TERZEA, E. (l970) La faune de mammiferes quaternaires de Ia grotte MAgurade (Bihor, Roumanie). Trav.lnst. Speol. "Emil Racovitza", IX, pp. 201-230. received: 26 September 2002 revised: 22 October 2002 accepted: 25 November 2002

PAGE 120

Theoretical and Applied Karstology, 15 pp. 117-123 The of karst resources exploitation: an example of iron industry in Kranjska (Slovenia) Andrej Kranjc Karst Research Institute ZRC SAZU, Titov trg 2, Sl-6230 Postojna, Slovenia. Abstract In Slovenia the karst surface covers 43% of the territory. About 57% is Dinaric and 21% Alpine karst. Dinaric karst consists of Jurassic and Cretaceous limestone, while Alpine karst ofTriassic carbonate rocks. Everywhere the iron ore is present. Two types can be distinguished, the form oflimonite pieces, arid the "ironstone" (iron hydroxides), so-called "broad bean ore" In the 8th Century BGthe iron industry prospered in two karst regions: i n theA Ips and on Dinaric karst. Technology of extracting i r o n from the ore was simple. The early metallurgists gathered iron ore o n the surface and in karst underground The ore was melted in shan furnaces with dug-in fireplaces. Because of the ironwork technology, iron industry was scattered all over the country thus no important concentrated pollution appeared Yet, local impact could have been important in places. The consumption of wood and charcoal wns significant including cutting down the forests in the lOth Century iron industry restarted. Between the 1 3th and the 15th centuries ironworks began to develop, based on water energy. Extracting a n d working of iron moved from the plateaux down to the valleys and the ironworks (joint of smelting furnace. fireplaces to heat the iron red-hot. bellows and ironwork hammer) resulted. In 1581, the list of ironworks for Kranjska enumerates 20 smelting furnaces. TI1e direct impact on karst consists oftraces of digging on the surface, in do lines, caves and potholes, disused mine shafts and galleries, and "ore pools" scattered across the land. As a direct impact. large quantities of charcoal were consumed and the water was polluted. There were much less forests left than nowadays. The indirect impact consisted mainly in the high concentration of industrial plants and inhabitants. Key words: iron industry on karst, pollution, Kranjska, Slovenia Resume Histoire de /'exploitation des ressources du karst: l'exemple de l'industrie du fer en Slovenie Le karst couvre 43% du territoir.e slovene, doni 57% represente le karst dinarique et 21% /e karst alpin Le premier est forme de roches carbonateesjurassiques et cretacees, landis que le deuxieme, de roches triasiques. Le minerai defer se trouve partout, en general so us deux formes : pieces de limonite et hydroxydes de fer ( Bohnenerz "). Au cours du 8-eme siecle av. J. -C. I 'industrie dufer fleurit sur le karst dinarique, ainsi que dans les Alpes. Les premiers nufta/lurgistes obtenaient le minerai defer aussi bien de Ia surface du karst que de gissements souterrains, et Ia technologie d'extraction du metal etait simple Le minerai etaitfondu dans des fours enterres avec le foyer creuse a Ia base Par suite de cette technologie, I 'industrie de fer eta it dissemine a travers tousle pays et il n y avail pas une pollution concentree Cependant, I impact aura it pulocalenzent plus fort La consommation de bois et de charbon eta it Ires importante ella consequence en Jut un fort defrichage A partir du I 0-eme siecle, l'industrie de fer a recommence a se developper, cette-fois-ci basee sur Iaforce de l'eau. Les activites ont ell! transferees des hauls plateaux kars tiques dans les val lees et les vraies fonderies (comprenantles hauls fourneaux /es foyers pour chauffer au rouge le fer, les souf {lets de forge et les masses) son/ apparues En /58/, dans Kranjska (Carniole) il y avail 20 hauls fourneaux. Tout cela a eu une influence directe sur I 'environnement, com me l atteste, par exemple, les traces de foui/les a Ia surface, dans les dolines ou dans les grottes et les gouffres, /es galleries de mines abandonnees et les basins de flottation disperses dans tout le pays II faut men tionner Ia consommation importante de charbon et Ia pollution de /'eau La surface desforets eta it plus restreinte qu 'aujourd 'hui. L 'impact indirect s 'est manifeste par une fort concentration des en/reprises industrielles et de Ia population. Mots-ells: industrie dufer sur /e karst, pollution, Kranjska, Slovenie corresponding author E-mail address: 0 2002, Editura Acadcmici Rom4ne. All rights.rcscrved.

PAGE 121

118 Related to human activity and its impact on karst it interest ing to see, what are the consequences of some past intensive activities, including industries which do not exist any more. Not only the impact is important, but also to see if the effects of these a ctivities are still visible today ; this either means that they are permanent or, in some cases, that through centuries and m i llenaries the nature regenerated and retu r ned to its natu ral state. In this paper we shall analyze this using the example of iron industry In Slovenia karst surface covers about 45% of the territory About 57% is Dinaric and 21% Alpine karst. From a morpho logical point of view Dinaric karst is divided into highland and lowland karst. Highland karst forms high karst plateaux (800-1200 m of altitude) stretching from the NW Slovenia ( Friuli border) to the Kolpa river in the SE. They continue to w ards SE across Croatia On both sides of the highland karst l ies t he lowland k a rst" On the south western sid e this is lit tora l (P e r ia dria t ic) while on th e n01t h-easterri s ide lies the karst ofDolenjsko (Peripannonian). They consist oflower plateaux less than 500 m asl. Most of the Slovene Alps namely Julian ADRf,!.TIC SEA .. j A Kranjc and Kamnik Alps consists of karst terrains. This central, high mountainous range (above 1500 m) is surrounded by p l a teaux (1000-1500 m) like Pokljuka, Jelovica or Menina (Fig: 1) Dinaric karst consists mostly of Jurassic and Cretaceous de posits, while Alpine karst ofTriassic carbonate rocks. Every where the iron ore is present in non-consolidated se dim ents covering the rock base or in the rock itself. In Slovenia, on the surface or in the underground there are two types of iron ore relatively easy accessible. Iron ore in the form of limonite pieces is found mostly in the Dinaric ka r st. The second type is the "ironstone" (iron hydroxides : Fe203nH20) wi t h onion like structures, f01: ming concretions up to 6 cmlarge. This type is called "broad bean ore" ("Bohnenerz" in German) and it is found mostly in the Alpine karst. Broad bean ore contains 30 to 50% iron. Iron ore can be found also in the form of bulbs and geod e s (STRMOLE, 1987) . The position of the Slov e n ia n territory ly ing on th e crossroad of two main directions, was very important for the de v elop ment of iron i ndustry. The first dir ection leading along the -Rauroad. + Airport 1 0 3Dkm 10 20ml Fig I Map of S l oven i a w ith t h e lo cat io n of the mai n localit ies m e ntioned in th e text. Ca rt e d e Ia S l ov e nie avec I a s itu a t io n d e s prin c i pales l oca l it e s m e n tio nn ee s da ns le t ex te.

PAGE 122

The history of karst resources exploitation . Danube; has spread the influences from the East, while the other one has spread the influence from the Mediterranean Italy across the Adriatic. In the 13th Century BC the Hittites started to extract the iron; the Greeks mastered this technique in the 11th Century BC an d this knowledge the Slovene relatively early. 'fhe oldest iron o bject imported in Slovenia dates to the I Oth Century BC and was found in Mu!ja jama cave near (Kras). The Early Iron Age (Hallstatt) extended, in Slovenia, from 750 to 300 BC. The inhabitants not only knew the new metal, but also used and extracted iron from the ore. On karst, iron industry first prospered in the 8th Century BC already. It mostly developed in two karst regions: in the Alps, in the region Bohinj including the Triglav massif, and on Dinaric karst of Notranjsko and Dolenjsko, including Bela Krajina. Archaeologists classify Hallstatt inhabitants in six culturct groups. Among those working in the iron industry, there are members of"Sveta Lucija" (western) group in Bohinj, as well'as the members of"Dolenjska" and "Notranjska" group {DULAR & BofiC, 1999). The technology of extracting the iron from the ore was rela tively simple. The iron ore was simply gathered on the surface and underground; seldom was it excavated. The greatest con centrations of broad bean ore lie at the bottom of do lines and in karst caves and shafts {R.IAZANCEV, 1966). The ore was melted in shaft furnaces with dug-in fireplaces. For a better melting, limestone was added (RJAZANCEV, 1963a). Some examples at Novo mesto show that the furnaces were about 2 m-high clay cones. The air was directed through special conduits; there fore the location of the furnace was very important, specially .. Fig. 2. Cross section of"wind furnace" above the Studor village (Bohinj}(R.JAZANCEV, 1962b). Coupe du fourneau a vent sUue au-dessus du village de Studor (BohinjJ (Rwn.;v l962b). 119 its exposure to wind. Furnaces were often located on the slopes to allow a good use of the upwards wind (RJAZANCBV, 1962c; RJAZANCEV, 1963d) (Fig. 2). Branches Qf spruce and beech charcoal were used as a fuel (RJAZANCEV, 1962a). This iron work requires a location that must be near the iron ore deposits, near the source of wood, and also exposed to appropriate wind; therefore the iron industry was not concen trated but scattered all over the country. This is proved by archaeological during which slag was found in the most oppida, in the settlements themselves as well as out ofthe walls. At the foot of high karst plateaux numerous re mains of prehistorical smelting furnaces were discovered: under Trnovski Gozd, near the important karst spring ofHubelj, under Nanos, near the village of Strane, and near the Bela sinking stream (LAMtrr, 1988). Recently, in the caveAntonkov Skedenj at Dolenjsko, the remains of a smelting furnace were discovered. It was placed in the cave, near a small shaft, thus using the air current from the cave towards the entrance {BIZJAK eta/., 2001) (Fig. 3). During the Hallstatt the Sloven ian territory was densely popu lated, even in those parts of karst which are now practically unpopulated. It was a cultural landscape and the Dolenjska and Notranjska groups of Hallstatt people have reached a protourban degree of civilization. Without doubt the role of iron industry was very important as well as the location of ore on karst. Today the impact of the iron industry on the karst environ ment is difficult to asses. There was no substantial concen trated pollution because the iron industry was relatively small and dispersed all over the country. Nevertheless subsequently, these changes can seriously impact upori the local areas. Antonkov skedenj sklca petl --TOm 1 e CD Fig. 3. Location qrthe prehistoric furnace in the Antonkov skedenj cave (BrtJAK et at . 200 I). (Vhod =entrance, = furnace, !Jindra= slag). Situation dufourneau pre-historique de Ia grotle d'Antonkov skedenj (8t7JAK eta/., 2001). (Vhod = entree, pee= fourneau, ilindra ,.; scorie).

PAGE 123

120 In the surroundings of Strafa village (near Novo mesto) the remains of24 smelting furnaces 1.5 m apart, were found (OuJ.AR & Bo!JC, 1999). Without doubt, the consumption of wood and charcoal was substantial. Cutting the forest could locally trig ger an accelerated soil erosion as proves the practice in the near past, when farmers in the mountains have burned forests in order to create arable land (KRAJIIJc, 1979). Iron industry has forced people to be more mobile. They had to search for iron ore deposits and carry the iron products at great distances. People from Bohinj have transported the flint used as the al loy to the ore from the Karavanke Mountains, which is at least 8 hours of walking distance (R.IAZANCEV, 1963e). During the Late Iron Age (La Tene) and the Roman period there were no essential changes concerning the iron industry. Traces are found in places of ancient towns, villages, hamlets and even single farms. The Slovene territory was included into the Roman Empire but its iron industry did not held any im portant position. Deposits of iron ore were relatively small and privately owned (HoRVAT & SASEL-Kos, 1999). The situa tion was different in the NW of Slovenia where Bohinj lies. A. Kranjc This area belonged to "Regnum Noricum"-the union ofNoric tribes. It mostly prospered during the second century BC. It was annexed by the Romans in the year 10 BC, and became a Roman province under Claudius (41-54). Iron from Noricum contained titanium and manganese of similar cjuality as steel and this is why it was highly appreciated throughout the empire (R.IAZANCEv, 1963c). As they say, even Homer have spoken about "ferrum noricum". This was the type of iron from Bohinj area (R.JAZANcEv, 1962b; R.IAZANCEV, 1966) and its prod ucts were exported mostly to Rome. In Noricum there was a special administration of mines having its head office at Virunum (SASEL-Kos, 1993). After the fall ofthe Roman Empire, the ancestors ofSlovenes settled this land, and, after a hiatus of 500 years, at the end of the Middle Ages, the iron industrY became again an important economy branch. In a document from 1004, the village Stara Futina (Old Ironwork) was called Staro Kladvo (Old Ham mer) which is the oldest written proof. There are more indirect written proofs than material remains. In 1340, the bishops of Freising moved the special workers for ironworks from Friuli Fig. 4. A furnace as a part of ironworks in the 16th Century (AGRICOLA, 1950). Un fourneau faisant partie d 'un attelier siderurgique du 16-eme siecle (d 'apres AGRICOl-A, /950).

PAGE 124

The history of karst resources exploitation to their estates in Slovenia (BAT & VRANICAR, 200 I). In 1381, the iron mine under Golica in Karavanke Mountains got the mine act from Frederick of Orten burg (LAMUT, 1988). In the.beginning, this iron industry was similar to that in the prehistory and antiquity, that is "a simple forest ironwork". The metal workers have searched iron ore on karst plateaux and melt it near the deposits using charcoal from the surround ing forests. Between the 13th and 15th centuries ironworks began to develop, based on energy (Fig. 4). Extracting and working ofiron has moved from the plateaux down to the valleys, close to the water courses. Extracting and working grew up as an industry in the modem sense of the word and the-ironworks resulted. Ironwork of the time meant a joint of smelting furnace, fireplaces to heat the iron red-hot, bellows and ironwork hammer. Sometimes the forges (blacksmith's workshops) were included into the ironworks too. In the be ginning the furnaces were small (AGRICOLA, 1950), similar to the ones from prehistoric times, but they became bigger and bigger until they reach the dimensions of real smelting fur naces. Streams are rare on the karst and so the water availabil ity was a restrictive factor and big karst springs become very important. The base for the ironworks in this period was the same as before, the deposits of broad bean ore and limonite iron ore in pieces which were gathered and dug out in Julian Alps, in Notranjsko and Dolenjsko region. New were the iron ore deposits in Karavanke Mountains ridge, in the carbonate rocks ofGolica and Stol Mountains. In 1581 the list of ironworks for Kranjskaand (county of Gorica-Goritia) enumerates 20 smelting furnaces. Many of the ironworks used the water energy of big karst springs and rare karst rivers to run bellows and hammers. Such were the ironworks of the Alps region in Bohinj (the village Bo hinjska Bistrica lies on the bank of the Bistrica stream with an important karst spring), in the foothills of Jelovica plateau (on the strong karst streams Kroparica and Lipnica), in the Notranjsko region -in the foothills of the plateau Tmovski gozd (karst spring Hubeij and the Bela stream), and in Dolenjsko region mainly along the Krka river (Zagradec, Dvor). Iron ore was gathered both on surface and in karst caves. Bet ter extraction technology also allowed the use of other types of iron apart from the broad been ore. Iron ore was mined up to altitudes of about 1700 m. As the concentration of iron ore is the highest in the caves, they were preferred for digging. In the Alps, several shafts are known where the traces of digging and the remains of mining installations can still be seen nowa days. In Brezno na a 26 m-deep shaft, one may found the remai,ns of a wooden construction. In Brezno na ( 1630 m a.s.l., Fig. 5), a 17m deep shaft, a wooden winch, wooden scaffold and poles were found (GosPODARJC & PoHAR, 1966). Two such shafts are also known on mevnik. Do lines were used for other two purposes. They were a source for clay used to cover the furnaces; in some of them quartz sand was found which was added to the ore in order to lower its melting point. Some do lines were transformed to "ore pools" 121 Fig. 5 The Brezno na shaft (Julian Alps) with traces and remains of mining activity (GosPODARIC:: & PoHAR, 1966). I.e puits ae Brezno na Jli.fevniku (Alpes Juliens) avec des traces d'activite miniere (Gosror>ARIC & PoHAR, 1966). were the rainwater was collected for ore washing. Old place names (Stara Fufina [Old Iron work] in Bohinj, Zelemiki in the Segka Sora valley, Futina [Iron Foundry Ironwork] pri Zagradcu), orographic names (Rudnica) [Metal Ore Hill], toponyms (Rudno polje [Metal Ore Field], Rudna dol ina [Metal Ore Valley], Rudno), water names (the Plavtevka [Blast Fur nace] stream) and house names (Pri Plavfarju [at the Smelting Furnace], Fefnar) prove the past mining and iron industry. It is self-understanding that such activities have had an impact on the karst land and its surroundings. In the middle of the 19th Century the ironwork in Bohinj only had 57 placers where the ore was excavated. Digging to the a depth of 30 m was usual, but there are cases of digging at the bottom of 62 m deep vertical shaft (BELAR, 1889); according to some unveri fied information (BRENCJC, 1991) miners penetrated even to the depth of300 m. The iron ore was lifted up from the shafts in wooden buckets, using wooden winches. Not only placers but also dolines were used as "ore pools" At the end of the 18th and in the beginning of 19th Century ironworks at Dvor (Dolenjsko) got the iron ore from an area with a radius of30 km. In Karavanke Mts. there were real mine galleries; Savske Jame mine, for example, has 7.6 km of underground rails. In 1855, I 0350 tones ofiron ore were excavated or gathered altogether in Kranjska. Two thirds of this quantity came from only two mines in Karavanke Mts. (OcEPEK & REDAKCUA, l996a; 1996b ).

PAGE 125

122 From 1580 up to the 18th Century the iron industry was in crisis in the alpine parts of Austria In Kranjska other difficul ties occurred also. First, Bohinj went short both in iron ore and fuel-charcoal. At the end of the 18th century, the owner of ironworks Z. Zois tried to use soft (spruce tree) instead of hard (beech tree) charcoal (RIAZANCEV, 1963b). In 1874 the ironworks in Bohinj have excavated 841 t of ore while the pro duction decresead to only 422 tin 1875 and 487 tin 1876 (RIAZANCEV, 1962a) In 1738 there were about 1000 ironwork workers in Bohinj, while in 1811 only 386 remained (RIAZANCEV, 1963a). Not only lacking of primary material but also the competition led to the termination of the iron industry on the karst of Kranjska. In 1835, the first modem ironworks of Rosthom were built in the Slovene part of (Carinthia), in Prevalje, and the old type of ironworks were no more com petitive (LESKOVEC, 1989). In Slovenia, during the 19th Century (Kropa 1880, Stara Fuzina in Bohinj 1890 and Bohinjska Bistrica 1891) and at the beginning of the 20th Century (Zelezniki 1902 1909) all the old types of ironworks have ceased to function (SM!TEK, 1992; Vmrc, 1998; VoJVODA, 1987). The modern iron or steel industry, which has remained in Sloveni a is neither based on domestic iron ore, nor it extracts iron but processes imported base-products or semimanufactures. The impact of the past industry on karst was both direct and indirect. The direct impact consists of traces of digging on the References AGRICOLA, G. ( 1950) De re melallica. I-XXXI, Dover Puhlications, New York, 638p (translatedfrom the firs/ Latin edition of 1556). BAT, M. & I. VRANJCAR (2001) Zelezniki Enciklopedija Slovenije. 15, pp. 311312. BELAR, A. ( 1889) Einc Grotte im Wocheincr Bohnerz -Rcvicrc. Millei lungen derSeclionfiir Naturkunde des OTC, I, 9, pp. 71-72 BJ7.JAK, 1., lAMNIK, 1'., & OsF.RSTAR, A. (2001) prvinc v jamnh Suite krajinc. Na .fe jame, 43, pp. 118-127. (Archaeological clements in karst caves ofSuha krnjinn, in Slovene' 13RENCJC-KuKJ, M. ( 1991) Rudniki na krasu. /Jilten JK Zeleznicar, p 26. (Mines on karst, in Slovene) DuLAR, 1. & BoZic, D. ( 1999) Zclczna doba. In: Zakladi tisocletij, Modrijan Ljubljana, pp. 98-183 (The Iron Age. In: Treasures of Millennia, in Slovenian) GosrooARJC, R. & PoJJAR, 1. ( 1966) svojstva rud. Zelezar T e hnicna p riloga, VIII, I pp. 7 -25. (Geological properties of iron ore deposits, in Slov e n e). HoRVAT, J & SASEt-Kos, M (1999) Rimska doba. In: Zakladi Modrijan, Ljubljana, pp. 184-291. (The Roman Era. In: Treasures of Millennia, in Slovenian) A. Kranjc karst surface, in dolines, in the caves and shafts and disused mine shafts and galleries and ore pools" scattered across the landscape On the karst plateaux ofPokljuka and JeloviCa only, there are remains of thousands of caves, hollows and shafts where the broad beari ore and limonite were excavated. On the plateau ofPokljuka, in the place called Rudno Polje [=Metal Ore Field] which is a sort of a very shallow and flat uvala, morphological changes can still be seen due to large stock of iron ore stored there. On this plateau the digging lasted sev eral tens of years, the ore was stored nearby and it was trans ported into the valley during the winter A direct impact is also the big quantity of charcoal consumed (the so-called Slovene furnace needed 50-60% more charcoal than ore) and the pol lution of water used for the washing of the crushed ore Not only on the Dinaric part ofthe karst but also on the karst of the Alps there were much Jess forests than nowadays An early photography of the ironworks at Bohinjska Bistrica shows bare mountains -white rocks -in the background (MOLLNER, 1909) Nowadays coniferous forests cover the slopes. The indirect impact consisted mainly in the high concentration of indus trial plants and inhabitants During the 18th and 19th centuries at Zelezniki there were two smelting furnaces, seven ironworks and II 0 blacksmith's fires, altogether giving work for about 2000 people Today although Zelezniki is an administrative centre there are only 31 00 i nhabitants altogether. In the middle of the 191h century Kropa had two smelting furnaces and 240 nailmakers In 1869 there were 1119 inhabitants while now there are only 875 (BAT & VRANICAR, 2001; SMITEK, 1992). KRANJC, A ( 1979) Kras v povirju Ljubije Geografski veslnik, 51, pp 31-42. (Karst in the waterhead, in Slovenian) LAMUT, J., ( 1988) trna metalurgija. Encikl op edija Slovenije, 2, pp 155-159. (Black Metallurgy, in Slovene). LEsKOVEC, A., 1989: Enciklopedija Slovenije, 3, pp. 170-172. (Ironworks, in Slovenian). MOLLNER, A (1909) Gesch i chle des Eisens in Krain, GfJrz und lslrien Halm und Goldmann, Wicn & Leipzig, 763 p. OcnrEI>, D & RF.DAK<.JJA, ( J996a) Rudnrstvo. Enciklopedija Slovenije, 10, Ljubljann, pp. 306-309 (Mining, in Slove ne). OcEPI!K, D. & RllDAKCJJA, ( 1996b) Rudnik. Enciklopedija Slovenije, 10, Ljubljana pp. 311-313, (Mine, in Slovene). RJAZANCEV, A. ( 1962a) Bobovci Julijskih Alp. Zelezar Tehnicna priloga, IV, 2, pp. 1-13, (Bean ore of Julian Alps, in Slovene). RJAZANCEV, A ( 1962b) Poskusno taljcnje bobovca v vetrni y Studorju Zelezar Tehnicn a priloga, IV, 2, pp. 14-24. (Experi mental melting of bean -o re in "wind furnac e" at Studor, i n Slovenian ) RJAZANCI!V, A. ( 1963 a) Po poti fclezarskega Bohinja. Zelezar Telrnicna priloga, V, 2, pp. 43-67 (Following the iron Bohinj, in Slovene)

PAGE 126

The history of karst reso11rces exploitation RIAZANCEV, A. ( 1963b) Vesti o bobovcih iz Julijskih Alp. 2elezar Tehnicnapri/oga, V, 2, pp. 67-70 (News related to bean-ore in Julian Alps, in Slovene). RIAZANCEv, A. (1963c) Bobovci Julijskih Alp. Zelezar Tehnicna priloga, V, 2, pp. 71-85. (Bean-ore of Julian Alps, inS/ovenian). RIAZANCEV, A. (1963d) Drugo poskusno taljenje bobovca na mejci nad vasjo Studor. 2elezar Tehnicna priloga, V, 2, pp. 85-97 (The second experimental melting of bean-ore at village near Studor, in S/ovenian). RIAZANCEV, A. (1963c) Od bobovca do tcleza ielezar Tehnicna priloga, V, 2, pp. 97-99 (From bean-ore to iron, in Slovenian). RIAZANCEV, A. ( 1966) Kemizem bobovcev in njih primerjava ter pogoji anti
PAGE 128

Theoretical and Appliid 15 (2002), pp. 125-132 Rapid communication Recent paleontological investigations in some caves of the Crimean mountain-range (SE Ukraine) Matei M. & Bogdan Ridush2 I University. Dept. ofGeology-Palf:ontology. str. Kogiilniceanu /, 3400 Cluj-Napoca, Romania. l Yuri Fedkovych Chernivtsi National University, Dept. of Geography. Kotsubynskogo 2, Cernivtsi 580/2, Ukraine. Abstract Preliminary results of the most recent speleo-palaeontological investigations made in the Crimean high-mountains arc presented. Several caves were investigated in the Chatyrdag massif; also new information regarding fossil remains from caves in several other plateaus (Aj Petri, Yaltinskaya, Knrnbi) arc added. The most interesting results come from the Em inc Bair Khozar cave, which significantly contribute to highlight the Late Pleistocene and Holocene vertebrate faunas existing in the remote mountains of the Crimea. Key words: vertebrate palaeontology, Late Pleistocene/Holocene, Crimea, Ukraine. Recherches pa/eonto/ogiques recentes dans quelques grottes de Ia zone montagneuse de Crimee (sud-est de /'Ukraine) Resume On prcsente les rcsultats des recherches sprUco-palcontologiques les plus rccents faites dans Ia haute montagne de Ia Crimee. Plusieurs grottes du massif de Chatyrdag ont ete fouil/ees; en outre, on a ajoute de nouvelles informations sur les restes fossiles des grottes situees sur d'autres plateaux (Aj-Petri, Yaltinskaya, Karabi). Les plus interessants resultas sont ceux obtenus dans Ia grotte d'Emine Bair Khozar parce qu 'i/s contrihuent a Ia connaissance des faunes de Vertebres du Pleistocene superieur et du Holocene des montagnes lointa!ns de Ia Crimee. Mots-cles: pa/eontologie des Vertebres, Pleistocene superieur/Holocene, Crimee, Ukraine. Introduction Since the beginning of the XXth Century, many archaeological explorations and diggings were carried out in Crimea, especially in the caves and rock-shelters developed in the middle range of the Crimean Mountains. Some of them are worldwide known, as the Kiik-Koba Mousterian site providing Neander thalian remains (STEPANCHUK, 2002), as well as the Kizil-Koba cave, famous for its culture-layers. Paleontological studies on Late Pleistocene faunas were related especially to these archaeological sites. Explorations on the main mountain range were rare; in fact until the 1960'h no paleontological investigations were carried out. Even afterwards, very few information was corresponding author. E-mail add1ess: ID 2002, Editura Academiei Rom line. All rights reserved. available on the palaeontological sites of the high-plateaus of Crimea (BACHYNSKY, 1970; DusLYANSKY & LOMAEV, 1980; LYSENKO, 1998). Following the 1999 and 200 I speleo-paleontological investi gations by us in three caves of the Chatyrdag plateau, several preliminary results were published (VREMIR, 2000) and a large-scale excavation project emerged. Spring 2002 was the starting moment for new excavations and extensive field investigations in the southern mountain range of Crimea, carried out by a joint team University in Cluj (Dept. of Geology) and Chemivtsi Univ. (Dept. of Geography). A second campaign was organised in the autumn of 2002; this time it focused on the Emine Bair Khosar cave, as well as on several new cave-sites. Two new

PAGE 129

126 sites have been studied in the Emine Bair Khosar cave; also an extensive excavation has been started in one of the previ ously known locations (Skull Chamber). New information re. garding "bone-bearing caves" located in other high-plateaus (AJ-Petri, Yaltinskaya, Karabi) were gathered. All of these strongly recommend the Crimean high-mountain range for future paleontological investigations. Investigated areas and sites The Chatyrdag Plateau The Chatyrdag Plateau is situated 30 km S from Simferopol, in the middle part of the main mountain range, being the sec ond highest mountain in Crimea (Ecliziburun Peak-1527 m a.s.l.). Like most of the Crimean yaylas ("yay Ia" is a local Tatarian term, meaning a flat mountain-top sur face, usually strongly karstified, characterised by steppe flora), Chatyrdag has a flat-top relief (Fig. I). It two levels named the Upper ( 1400 m a.s.l.) and the Lower ( 1000-11 00 m a.s.l.) Plateau, respectively. From a geological perspective, a flysh-type marine sequence, represented by the Tavricheskaya Series (Upper Triassic-Lower Jurassic) is unconformably overlapped by a very thick, up to I 000 m, carbonatic series, consisting of Oxfordian conglom erates and massive, Lower Tithonian, limestones. The limestone plateaus and cliffs are extensively developed between the elevations of700 and 1500 m, and more than I 00 potholes and caves are registered. M. M. Vremir & B. Ridush At the present time, on the Chatyrdag lower plateau (1000-II 00 m a.s.l.), five caves present interest from a paleontologi cal point of view: the Emine Bair Khosar, Emine BatrKoba, Krapivnyj, Cherepa and Mramornaya caves. In the Chatyrdag upper plateau (1400 m a.s.l.) the Angar-Burun pothole was investigated and a bone accumulation was identified. Mramornaya Cave .. Mramomaya is an approximately 2 km-long, partially show cave, developed in Late Jurassic limestones. It includes large chambers and passages, as well as some pits which connect the two main, inactive, levels. During the expedition in 1999, almost the whole cave was searched, and a very few ve11ebrate remains were founded in the lowest level of the cave, as well as in the main fossil pas sage. In the lowest level (site MrA), the most interesting discovery was a partial cave lion skeleton (Panthera leo spelaea). The preservation of the bones is rather poor, the skeleton being disarticulated and partially destroyed by exposure (Bahrens meyer's weathering stages 2-5). Some of the bones show ro dent gnawing marks, which are partially covered by calcite crust. In the same chamber, three fox ( Vztlpes vulpes (corsac?)) and a hear (Lepus timidus) complete skeletons were discov ered, preserving the outlines of the body and internal organs. Some rodents (Cricetidae) remains as well as a few goat teeth (Capra sp. cf. C. ibex) were also found. In the main level of the cave (site MrB), we find several poorly preserved skull fragments belonging to a deer ( Cervus elaphus). Fig. I. Chatyrdag-upper plateau. Chatyrdag -le haul plateau.

PAGE 130

Recent paleontological investigations in Crimean mountain range A hear (Lepus timidus) and a young wild boar (Sus scrofa) skeletons were discovered in a side-passage (site MrC), which also preserve the outlines of the body and soft tissues Except for the lion, the goat and the deer remains, all materi als are recent in age. From a biostratinomical point of view, the majority of the specimens (especially the hears and the wild boar) were trapped in the cave following accidental pen etrations, and show no further disturbance. Bone accumula tions or potentially rich paleontological deposits are absent in this cave, due to its particular morphology. The only impor tant find is the cave-lion skeleton (rarely preserved as a whole), a species which was identified in other Crimean caves as well: Krystalnaya (Aj-Petri Plateau) (BACHYNSKY 1970) and Emine Bair Khosar (Chatyrdag Plateau). Emine Bair Khosar Cave Emine Bair Khosar is the most interesting paleontological cave in the plateau. It is located on the northern edge of the Lower Chatyrdag plateau, very close to the Mramornaya cave com plex and to the same karstogenetic cycle. With a total Fig. 2. Emine Bair Khosar-the base of the entrance shaft mine Bair Khosar -Ia base du puits d 'en tr ee 127 length of 1460 m and a depth of-125m, it is one ofthe big gest cavities in the plateau. The collapse-pit entrance (Fig. 2) is situated on the northern slope of Chatyrdag, but all known passages are oriented not towards the closest erosion base level, but inside the mountain massif(DuBLYANSKY & LoMAEV, 1980, p. 41). This peculiar configuration is probably related to the hydrothermal origin of the cave, which presents two main levels: the upper one is composed by large chambers and passages which progressively descend to -50 m, while the lower level, connected by pits, is and develops at a depth of-125m. Fossil bones were identified within both levels, however the most important accumulations are in the upper one. In 1960, at the beginning of the speleological exploration, the cave provided 180 bones, which were found in a small cham ber near the Access Passage, under a flowstone crust. This osteological material belongs to 24 individuals of cave be a r ( Ursus spelaeus), wolf(Canis lupus L.) corsac ( Vulpes cor sac L.), cave lion (Felis spelaea GoLDF.), northern lynx (Lynx lynx L.), horse (Equus cabal/us L.), red deer (Cervus elaphus L.), hare (Lepus sp.). DUBLYANSKY & LOMAE V (1980) consider that this bone accumulation was formed not far from the former entrance (subsequently closed by a breakdown), following a long inhabitation by large carnivores (op.cit., pp 136-1 37). Due to recent construction-works, this site is no more avail able for studies. In the 19901h more bones were collected from different sites inside the cave but unfortunately there are very few information regarding the specimens found and their origi nal locations. We have searched several fossiliferous sites within the main (upper) passage. In the entrance area (sites BA I and BA2) (Fig 3a, b) we could recognise a complex taphocenosis con taining faunal elements which belong to different bio topes, accumulated here in a long period of tim e. The mixed bone material is abundant in a huge debris -c one deposit accu mu lated in the lower part of the large, pitch-type, e ntrance (site BA I). The site BA2 (Museum Chamber) is located several tens of meters away from the natural entrance, at a depth of28.4 m. This represents the edge of a sliding cone pressed in a lateral passage, which also contains a large amount of bone material. This site was partially excavated for fitting purpos es; later it was transformed in a small Museum Chamber. Generally, all the bones were collected as isolated elements. The most interestin g specimens are represented by lar ge her bivores such as the mammoth (Mammuthus primigenius), wholly rhino (Coelodonta antiquitatis), some bovines (Bas primigenius; Bison priscus), red deer (Cervus e/aphus), roe deer (Capr e olus capreolus) and equids (Equus caba l/us). Sev eral other mammals are also present such as the wild hog (Sus scrofa), cave lion (Panthera l e o spela ea ), mustelidae and ro dents (a...complete taxonomic identification is still required) The second newly investigat ed, site (BB) is represented by a very narrow passa ge with a bone-breccia typ e infill which is

PAGE 131

128 M. M. Vremir & B. Ridush c. 5 4 /los I Hisonl Ccrvus elaphu.f 3c Sait:a tala rica 3b Salgn tatarica .. 3a 2a 0 .5m Fig. 3. Locations of the fossiliferous sites in the Emine Bair Khosar cave (a: plan; b: long profile). c: stratigraphic section on the BC she (Skull Chamber). Situation des site vfossiliftres dans Ia grotte d'Emine Bair Khosar (a: plan; b : ,profil). c : coupe stratigraphique dans le -site BC (Salle du Crane)

PAGE 132

Recent paleontological investigations in Crimean mountain range developed at the depth of -34 m, under the debris-cone. The bones are very well preserved, sometimes covered by calcite crusts. From this site we recovered for identification some cranial and postcranial elements belonging to wholly rhino (Coelo donta antiquitatis), possible wisent (?Bison priscus), horses (Equus cf. mosbachensis, Equus hemionus), and red deer (Cervus elaphus). Most of the bones are cemented in flowstone and bone-breccia, or between large boulders. The taphonomic characteristics suggest the arrival of isolated skeletal elements (especially long bones and vertebras, but skulls and other elements are also present), after a short transport with or within the collapse material accumulated here. It is very likely that the finer sediments were subsequently washed away (even nowadays, during the wet season a small stream cascades through this hole). This site represents one of the "undisturbed" and promising accumulations, which however is not acces sible at the present-time because of the very narrow entrance passage. A chamber infill with clayey and stony sediments represents the third location (site BC, The Skull Chamber) (Fig. 3). The bones are also very abundant, and we could recognise a similar kind of faunal assemblage as in the previous sites (large herbivores including mammoth (Fig. 4), wholly rhino, bovides, 129 cervides, equids as well as antelopes, carnivores, rodents and birds). However the taphonomical characteristics are rather different. The ascendant passage is above the Skull Chamber at the depth of -34 m and somehow below the main debris-cone (last levelling place this site a few meters under the Museum Chamber), representing in fact the lower section of the Museum Chamber profile. Besides the large mammals with the dominancy ofbovides, cervides and antelopes (Saiga tatarica) the carnivores (Panthera leo spelaea, Vulpes) and human remains are also present. Complete skeletons, skulls and other skeletal parts in anatomical connection were recov ered, suggesting the initial presence of more or less complete .carcasses mixed with isolated bones or disarticulated skeletons. The fourth newly searched (BD) is located at the junction of Dublyanskogo Chamber and Skull Passage: a few bones collected from here belong to a fox (Vulpes vulpes), horse (Equus sp.) and a red deer (Cervus). The cave fitting works, recently disturbed this site. The fifth site (BE) is located close to the Skull Chamber at the depth of -48 m, near the tourist path. In a small niche partially covered by flowstone, rodents (Cricetidae) and small carnivores {Mustelidae) remains were accumulated (study in progress). )) Fig. 4. Excavation survey: partial mammoth skeleton in BC site (September 2002). Carte de lafouil/e: squelette partie/ du mammouth dans le site BC (septembre 2002)

PAGE 133

130 From a taphonomical point of view, the pitch-type entrance of the cave (above site BA I and BA2) functioned for a long time as a natural trap. Several thousands of bones (disarticulated skeletons) can be accumulated inside the debris cone, More detailed information will be available following the statistical ., I / ... Y;: l / I /' / '--... J. ( .. ... -v _,.., t ,,/ r r ... -: _, .... ;'. (;': .. / ),.: \. ; \. I ," . J :-: ...... ., \. ' ', . .. ::. ... I .. _.. . ,., '.(_I : .J/ / /' ' I .' \<) / ) -'_ ::-_.,/ .... i ,. : c \ ;.. : ', Fig. 5 Mammoth skeleton in BC site (April/September 2002) (top) and reconstruction of the original position of the carcass (bottom). Squelelle du iHammouth dans le site BC (av ril/septembre 2002) (en haul) et reconstitution de Ia position originate de Ia carcasse ( e n bas ) M. M. Vremir & B. Ridush data processing and interpretation. The most important results come from the site BC (Skull Chamber) which preserves par tial carcasses or complete skeletons in anatomical connection, being accumulated very close one to another probably in a short time. From a few square meters of digging, more than 1000 bones were recovered, belonging to several large herbi vores (Mammuthus primigenius, Coelodonta antiquitatis, Bos primigenius, ?Bison priscus, Cervus elaphus, Rangifer tarandus, Equus-probably two species), as well as saiga an telope (Saiga tatarica), carnivores (Pantheta leo spelaea, Vulpes vulpes), Iagomorphes (Lepus timidus) rodents, birds (PyrrocC?rax graculus, Columba cf. oenas, a.s o.) and finally a human vertebra We have to notice the presence of a mammoth skeleton (young specimen) (Fig. 5, 6), one auorachs skeleton, three deer skel etons, partial horse and bovid carcasses, a saiga skeleton, hear, fox and other small mammal partial skeletons or isolated bones. The preservation and spatial distribution of the bone material, the stratigraphic and microambiental data as well as the pas sage configuration suggest a very peculiar taphofacies. At the present time, an underground glacier with a steep slide is the most adequate model for this kind of layered accumulation; all carcasses have been pressed in a relatively narrow passage at the bottom of slide and the decomposition processes took place in situ. The paleontological potential of this cave is huge. We can recognise faunal assemblages which belongs to the last glacial stage (cold steppe elements, data supported by pollen analysis as well), also to interstadials and postglacial (mostly related to grassland and forested landscapes in a warmer climate. The Fig. 6. The mammoth skeleton retrieved from BC site Emine Bai r Khosar The scale bar indicated by an arrow is 15 em long Squelette drrmammouth recuper e d u site BC de Ia gro tt e d'Em ine Bair Khosar. L 'eclzelle e st indiquee par Iafle c he de 1 de longueur.

PAGE 134

Recent paleontological in Crimean mountain range large amount of bones, the state of preservation, the tapho cenosis and the affluence of taxa, as well as the complex as semblages accumulated in a long period of time, grant a spe cial status to this cave, which can solve questions concerning the palaeo-biogeographic, palaeoecologic and palaeoclimato: logical problems of this region. Krapivnyj Grotto It is located 0.7 km south from the Emine Bair Khosar cave, on the edge of the lower plateau. From a morphological point of view, it is a sub-horizontal cave, developed a few meters below the surface, and opened by erosion processes. The entrance area is formed by a chamber of7 x 15 x 7 m (site Kr. l ), which continues with a narrow passage, I 0 m in length (site Kr 2), recently excavated by the cavers from Simferopol. The entrance chamber (Kr.l) represents a new archaeological site. The preliminary search revealed the presence of a large amount of pottery belonging to different cultures, from the Late Bronze/Early Iron Age till late Middle Age, as well as bone materials (domestic and wild species) The bones are mostly fragmented, but on the basis of a few teeth and man-. dibular fragments the cervids (Cervus elaphus), bovids (8os), equids (Asinus, Equus) as well as small carnivores were docu mented. The most interesting find is a skullcap belonging to a mustellid, which was excavated from the site Kr.2. The fossil is well pre served, being cemented in a very thick (over I m) tlowstone fom1ation, which closed the inner passage. The age of this bone is as yet unknown. In the process of speleological digging both sites were super ficially disturbed, and no stratigraphical or taphonomical data are available. Chercpa Cave The preliminary investigation in this relatively small cave, with a narrow potholetype entrance, situated close to the Emine Bair Khosar cave, revealed a subfossil bone accumulation, which contains numerous horse, ox, deer and goat remains Cavers from Simferopol made some diggings (explorative purposes), which disturbed the site, and the original tafofacies. Further excavations are needed. The Karabi Plateau The only skull of musk-ox (Ovibos moschatus) comes from Lzmnava Cave from Karabi Plateau (material deposited at the Geographical Dept. of Vernadsky Tavrichesky University, Simferopol). The finds of Ovibos in the Eastern Europe are numerous, but only north from so parallel. They are mostly related to Late Paleolithic sites (GROMOVA, 1965, p. 104). In the Western Europe the most musk-ox remains belong to the last glacial stage; the South they were found as far as Hungary and the SW of France At that time Ovibos have reached also Dobrogea in Romania (GROMOVA, 1965, p. I 02), 131 but it was not numerous. V ERESCHAGIN & BARISHNIKOV ( 1980) considered that the musk-ox probably did not penetrate into Crimea. They have noted as well the absence of Bos primigenius in the Palaeolithic. It seems that this is the first Ovibos record in Crimea and also the southernmost in Eastern Europe. Another shaft-cave from this plateau (Mammoth Cave), pro vided fossil bone materials, which were attributed mainly to the wholly 111ammoth (Mammuthus primigenius). The A)-Petri Plateau On the western side of Aj-Petri, in the shaft of Crystal Cave, an almost complete cave lion and a fox skeletons were found (BACHYNSKY, 1970). Radiocarbon dating of the cave-lion (Dr. Doris Nagel, Wien, pers. comm. ) indicates 12000 BP. Preliminary results and conclusions Information regarding speleal bone-accumulations or bone caves in the Crimean mountains were scarce until now At least ten situated on four high-plateaus were re corded, which potentially could yield information on Late Pleistocene and Holocene mammalian faunas of the remote Crimean highlands. The best-documented area is the Chatyrdag Plateau, where six caves/potholes were searched, two of them (Mramornaya and Emine Bair Khozar caves) being properly investigated. Other caves also provide some infonnation, which led to the following conclusions : Some of the large shaft-caves and potholes, functioned for a long time as natural traps Bone accumulations are frequently mixed and/or dis turbed. Very few layered and/or undisturbed bone-deposits are known. The identified faunal elements, belong to different assemblages habitats, and periods, suggesting a complex taphocenosis (at least in some of the sites). The Last Glacial Stage is well documented in almost all investigated caves, being characterised by a cold-steppe fauna with mammoth, wholly-rhino, musk-ox, wisent, reindeer, a s .o. Warmer steppe elements are represented by horse, Asiatic wild asses, aurochs and antelopes. The forested landscape is indicated by roe deer, red deer, wild hog, wolf, fox, lynx, cave lion and bear More recent faunal elements are also recorded. The presence of different faunal elements (including fossil man) together in the same deposit, suggest a complex taphonomic history, but also the possibility of co existing faunas due to the geomorphologic conditions.

PAGE 135

132 The first complete mammoth (Mammuthus primi genius) skeleton in Crimea has been found. The first documented aurochs (Bos primigenius) and the first Saiga antelope skeleton and skulls in Crimea have been found. The first musk-ox (Ovibos moschatus) record in the Crimea. Extensive field investigations are needed in ali of the high plateaus (Aj-Petri, Yaltinskaya, Dolgonukovskaya, Babugan, Chatyrdag, Demergi, Karabi), since obviously many of the large shaft-caves or potholes identified until now, could have functioned as natural traps. Bone accumulations were recorded in more than ten caves (four plateaus), but the potential seems to be much higher. It is very likely that the undisturbed-lay ered bone deposits will yield valuable information regarding the vertebrate faunas of the Crimean high-mountains, during the Last Glacial Stage and the Holocene time. References BACHYNSKY, G A. ( 1970) The taphonomic characteristics of fossil vertebrate sites in karst caves of Ukraine. Fizyclma geographija ta geomorphologia, 4, pp. 153-159 (in Ukrainian). DuBLYANSKY, V.N. & LoMAEV, A.A. (1980) Karst Caves of Ukraine. Naukova Dumka, Kiev, 180 p. (in Russian). GROMOVA, V. ( 1965) The short review of Quaternary Mammals of Europe (the experience of comparison). Moskow, 144 p. (in Russian). LYSENKO, N.l. (1998) Mammoth in the Crimea. Priroda, 3-4, pp. 8-12. (in Russian). M M Vremir & B. Ridush Acknowledgements We are grateful for the help during the field investigations to the following students and cavers: Alexandra (Omega Cluj); PQp Lucian; Balogh Mikl6s (CSA Cluj); Kovacs AI par, Tomas R6bert, Burian Szabolcs, T6th Attila and J6zsa Bolyai" University in Cluj, Dept. of Geol ogy) and the cavers from Simferopol and Yalta (Crimea). We thank for the help on taxonomic identification to: Gal ErikaCluj (birds), Dr. Doris Nagel-:Wien (carnivores) and Gasparik Mihaly-Budapest (micromammals). The present research was partially supported by a scholarship awarded to MV by Domus Hungarica Scientiarium et Artium. Also we thank the follow ing sponsors for their help: "Romgaz" "Oncos" Cluj; "ATTA'' Cluj; "Orbital" Cluj well as "Onyx Tour" and Alexandr Kozlov, Simferopol. STEPANCHUK, V.N. (2002) Late Neanderthals ofthe Crimea: Kiik Koba site. Kiev, 216 p. (in Russian). VERESCHAGtN, N.K. & BARYSHNtKov, GF. (1980) Mammals ofNorth Crimean foothills in Palaeolithic time. Proc. of the Zoological Institute Leningrad, 93, pp. 26-49. (in Russian). VREMtR, M. (2000) Palaeontological investigations in some caves of the Chatirdag Plateau (Crimea; Eastern Ukraine): A preliminary report. In: Rea/iziiri perpective in Studiul Cuaternarului din Roinonia (ONAC, B.P. & T., Eds.), Presa Universitar!l Clujcanil, Cluj-Napoca. pp. 35-37. received: 29 October 2002 revised: 22 November 2002 accepted: 26 November 2002

PAGE 136

Theoretical and Applied 15 (2(Xl2) Monografia carstului din Muntii Padurea Craiului Gheorghe Racovitl, Oana Moldovan, & Bogdan P. Onac, Editors Ed. Presa Universitara Clujeana. Cluj-Napoca, 2002. 264 ( +xii) p., 3 b&w photographs, format 18 x 24 em. ISBN: 973-610-101-0 "The Karst of Padurea Craiului Mountains. Monographic study" (in Romanian, sum mary and figures captions in English). price: USD 10 plus postage; on sale at the "Emil Racovita" Institute, Cluj Branch, str. Clinicilor 5, 3400 Cluj-Napoca, Roma nia. email: Book Reviews A characteristic of the Romanian karst, where the limestones outcrop on only 2% of the territory, is the diversity and scattering of karst units. P:tdurea Craiului Mountains, one of the most important karst regions of Romania, i s an exception. Here, the karst plateaus are especially unitary and well-devel oped. The karst is hosted by a low-altitude, peneplaned massif, located in between two large depressions; its surface exceeds 450 km2 and includes Triasic, Jurassic and Cretaceous limestones. Moreover, at the limit between Tithonian a,.d Neocomian, a bauxite-bearing paleokarst is located. The po sition of the karst ; the permanent water input coming from the imperv i ous deposits, the stable base'levels and its pre-Quaternary evolution allowed the formation of extensive subterranean networks (among which the 45 km long Vntului, the longest in Romania), as well as that of an impressive exokarst. For its importance, the karst of P:tdurea Craiului was a permanent attraction for cavers and a generous research topic for karstologists. Apart from nu merous articles and studies dealing with a broad range of subjects, two mono graphs and a hydrogeological synthesis have been published within the last 15 years. The first book, Carstul din Munfii Padurea Craiului (Rusu, I 988) is a regional karstology study that includes extensive morpho-hydrographic and cartographic works ofthe most important researcher of this area. Taking into account the exhaustive interest of the author for this region and the care for even smallest details, one should hardly imagine, a decade ago, tha t a new approach to P:tdurea Craiu lui will hapPen too soon. However, in I 991 the Hydrogeological Map of Padurea Craiului Mountains appeared in Theo retical and Applied Karstology I 991 ) and, this year the "Emil Racovit:l" Institute's branch in Cluj has published the volume Monografia carstului din Munlii Padurea Craiului dedicated to the memory ofTheodor Rusu. I have mentioned these historical details in the introduction ofthis review in order to emphasize that the diversity and the evolution of kars t ology con cepts over a short tim e -span may be integra te d into an attempt to continue and, especially unifY various researches This is the first impression one will have when skimming through the book and it will defini t ely become clear after reading i t. The volume is a collection of contributions often authors (D. Borda, R. Bucur, A. Fekete, C S_.)epure, 0. Moldovan, B.P. Onac G. Rajka, G. Racovi\A, M. Venczel) making 13 chapters that include both new research and updates of studies already published. All these complete the "big pic t ure" of the karst ofPlldurea Craiului and finally yield a unitary sy s t ematic and complex image of the region.

PAGE 137

134 The first two chapter outline the subject by presenting a history of karst research (G. Racovitii) and the geologic and geographic background (B.P. Onac). The latter chapter includes a synthesis on the exokarst features, sys tematically presenting the repartition of karren, sinkholes, ouvalas, karst depressions, karst valleys and plateaus. The analysis of caves is much broader (B P Onac and G Rajka). I t starts with the hydrogeological setting (swallets and springs, underground drain ages, water chemistry, hydro-karstic systems), then the caves and their speleogenetic models are dealt with accordingly. Important sections are dedicated to the mineralogy and geochronology of subterranean deposits. This chapter concludes with a presentation of the caves Litophagus Vantului, Ciur Izbuc, Ciur Ponor, which are considered as representative for Piidurea Craiului. The underground climatology is dealt with in a separate chapter (G Racovi{ii) that includes analyses ofthe topoclimate from Vfintului cave, cu Apii din Valea P. de Ia Fata Apei, and P de Ia Vadu The results ofbiospelcological researches arc described in more detail. The chapter dedicated to the aquatic fauna (S. lepure and 0. Moldovan) outlines the diversity ofstygobiontic and endemic fauna, which includes twelve groups with 87 taxa and 255 species, among which 19 are endemic. The cave 6f Vadu (59 species and subspecies) was selected for a complete presentation of the species and their biotopes. The terrestrial fauna is described within two other chapters: snbterranean Coleoptera (0. Moldovan, G. Racovitii and R. Sueur) and Chiroptera (D. B.orda) They do not include only fauna lists but also interesting zoogeo graphic considerations, criteria for the speciation study (Drimeotus), nu meric keys for the detennination of the species (Parapholeon), problems related to behavior or genetic variability (leptodirines) genus or subgenus revisions. A special note for the presentation of bats populations from five caves These chapters are completed by another one dedicated to the study of environmental conditions and fauna variability in two caves from north-em Piidurea Craiului. The next two chapters deal with paleontological (M Venczel) and archaeo logical (C. studies carried out by the Muzeul Tiirii in Oradea. Karst paleontology studies were especially focused on the impor tant fauna ofTriassic reptiles and Cretaceous dinosaurs found in the bauxite mines but also to newer (Neogene) remains The archaeological chapter pre sents the stratigraphy and the artifacts discovered in Ungurului and caves. The volume ends with a short chapter dealing with the subterranean envi ronment protection (0. Moldovan) followed by an extensive bibliographic list, which is representative especially for the papers published in Transylvania and Hungary. Cristian Goran Book Review.f

PAGE 138

Information for contributors to Theoretical and Applied Contributions to TAK may be submitted for publication as Articles (usually up to 8000 words) or as Short Notes (usually up to 2000 words). Larger articles, such as Reviews may also be published but their subject and size should be discussed first with the Editors. Manuscripts must be submitted in English or French, in electronic format accompanied by two hard (paper) -copies. The order of preference for elec-. Ironic submission is: (1) e-mail attachment, (2) floppy-disk(s), (3) CD, (4) ZIP Disk (100 or 250Mb). For fast submission of large files, FTP transfer may be arranged. We currently use IBM-compatible PCs and can read most popular wordprocessor, spreadsheet and graphic file formats; if you are using OS/2 or AppleMac, please format your disk as a DOS disk. Also, please consult us if your graphic files were saved under an uncommon format (e. g : the .tpw format of FigP .sta format of Statistica and alike) or if you used special character sets (such as East European, Turkish, etc. ) for diacritical marks. In any case, the hard-copy is essential for reference as we may unknowingly lose parts of your text or formatting in the conversion process. Also note that after acceptance, all the corrections or revisions to your paper will be made only on hard-copies in order to reduce the possibility of error. Papers should be headed by the title and the author(s) full name(s). This will be followed by the complete address(es) of all the authors. If more than one author, please specify the name and address of the person to whom the correspondence and proof should be sent. An abstract (up to 300 words) in the pai>er's language should be placed directly after the title and author(s)' name(s), followed by the keywords. If possible, these will be followed by the translation of the title, abstract and key-words in the 'secondary' language of the paper (English if the paper's main language is French and vice-versa); else the translation will be made by the Editors at no charge for you. References cited in the text (and only these) should be listed, in alphabetical order, on (a) separate sheet(s), after the heading References/ Blbllographle,taking for example an older issue of TAK. References should be cited in the text in parentheses, in small capitals using the 'name and year' ('Harvard') system, e.g. '(AHRENs, 1962)' except when the author's name is part of the sentence, e.g. 'AHRENS (1962) has shown that .. .'. When reference is made more than once to same author and year, a, b, c, etc. shpuld be added to date in the text and reference list. If more than two authors, it is in order to put'et al.' behind the first name, e g '(KLEIN et e/. 1988)' within the text, but a// authors should be listed in the Reference list. In general, one should follow the American Psychological Association (APA) Method of referencing. If in doubt, please consult us. Mathematical equations should be numbered in consecutive order throughout the text, using Arabic numerals. Figures (maps or line drawings in black ink on transparent paper) should be prepared in anticipation of reduction; small symbols or fine details may poorly. If. possible, the original figures shall not exceed 31 x 20 em. The number of folded plates being limited, only one figure exceeding this size will be, as an exception, accepted. Only black-and-white photo graphs may be published, therefore, if you send color photographs make sure that the important details will legible when converted to grayscale. Whenever possible, please send us the illustration in both electronic and hard-copy format, but, as a rule, the original will always be preferred over the electronic version. Plates, figures and tables should be numbered consecutively, with Arabic numerals, in order of appearance. The top of the figure should be indicated whenever exists any possibility of ambiguity. Please indicate in the manuscript the caption-figure or group of figures is to be inserted. The captions and explanations of the figures and tables will be doubledtyped on separate sheets, in both volume's languages, if possible. Acknowledgements should be placed at the end of the text, after the heading Acknowledgements, before the References. Upoll acceptance, an edited proof of the contribution will be sent to the authors for final revision. Normally, the proof will be sent by email, as a PDF file. Authors are expected to print the file, make the final corrections (if necessary) sign the Agreement for Publishing and fax it, together with the revised proof, to the Editors Authors bear sole responsibility fO{ the content of their contribution: this includes obtaining any clearance for the use of copyrighted materials. As a general rule, NO manuscripts will be returned, regardless if the paper was accepted or not. Papers accepted by the Editorial Board and published into TAK will become the property of the Publisher. No article or any part thereof may be reproduced in whatever form without the written permission of the Publisher. A total of 25 reprints of each paper will be supplied to the authors free of charge. Authors will benefit of discount prices for additional reprints if ordered before publication.

PAGE 139

In this issue: I" P. Forti Speleology in the Third Millenium: Achievments and Challenges. p 7 B. P. Onac, R. Breban, J. Kearns, & T. Tama Unusual minerals related to phosphate deposits in Cioclovina Cave, $ureanu Mts. (Romania). p 27 D. Dumitra, $. Marincea, G. Diaconu, C. Constantina, & R. Pavel -X-Ray p owder data on some mineral species from Pel}tera Curate de Ia Nandru (Hateg Basin, Romania) p.35 J.-C. Ljubojevic & V. LjubojevicKarst of the Ridge .. Our simple philosophy is nothing more than a piece of soul mixed with some ideas and a lot of work. We don't have any incorporated anywhere but we assure you that we care about your comfort and safety; . outdoor m project 1:111 A. Tenu, avidescu, R. PetreJ, & L. Coarna-Environmental isotopes studies and the model of South Dobrogea (Romai;I'Ja). p, 6 D. Taborosi Biokarst on a tropii:al carbonate Guam, Mariana Islands. p '73 H. C. Kohler & J. Karfunkel-Th morphogenesis of the Lagoa Santa ropical kars Minas Gerais State, SE Brazil. p. 93 M. Erbajeva, F. Khenzykhenova, & N. Alexeeva-Small mammals of the cave sites in the Baikalian region. p 101 E. $tiuca & R. Arghir-Martes genus representatives in the Wurmian of Romania. p. 111 Dzevrinska Greda: fluvial influences, caves, and groundwater A. Kranjc-The history of karst resources exploitation : the circulation. p. 43 example of iron industry in Kranjska (Slovenia). p. 117 I. Jemcov, R. Pavlovic, & Z. Stevanovic-Morphotectonic M. M. Vremir & B. Ridush Recent paleontological analysis in hydrogeological research of karst terrains A caseinvestigations in some caves of the Crimean mountainstudy of SW Kucaj Massif, Eastern Serbia. p 51 range (SE Ukraine). p. 125 and [l] antech outdoor equipment for professionals solitaire backpacks sleeping bags ponchos gaiters bags caving sacks professional clothing accessories