GEOLOGICA CARPATHICA, 51, 5, BRATISLAVA, OCTOBER 2000
VOLCANIC EVOLUTION AND STRATIGRAPHY
OF THE MIOCENE BÖRZSÖNY MOUNTAINS, HUNGARY:
AN INTEGRATED STUDY
, EMÕ MÁRTON
, SZABOLCS HARANGI
, SÁNDOR JÓZSA
, ZOLTÁN PÉCSKAY
, SÁNDOR KOVÁCSVÖLGYI
and ALFRÉD DULAI
Eötvös University, Department of Physical Geography, 1083 Budapest, Ludovika tér 2, Hungary
Eötvös Loránd Geophysical Institute, Paleomagnetic Laboratory, 1145 Budapest, Columbus u. 1723, Hungary
Eötvös University, Department of Petrology and Geochemistry, 1088 Budapest, Múzeum krt. 4/a, Hungary
Institute of Nuclear Research, Hungarian Academy of Sciences, 4001 Debrecen, Bem József tér 18/c, Hungary
Eötvös Loránd Geophysical Institute of Hungary, 1145 Budapest, Columbus u. 1723, Hungary
Hungarian Natural History Museum, Department of Geology and Paleontology, H-1431 Budapest, Múzeum krt. 1416, Hungary
(Manuscript received December 17, 1999; accepted in revised form June 20, 2000)
Abstract: The Middle Miocene volcanic evolution of the Börzsöny Mountains, North Hungary, is presented, correlat-
ing new volcanological, petrological, geochemical, geophysical and paleontological data and establishing a detailed
stratigraphy on the basis of additional K/Ar radiometric and paleomagnetic measurements. For the earliest volcanic
activity, previous biostratigraphy showing an Early Badenian age has been confirmed and precisely defined by paleo-
magnetic investigations. The first-stage volcanic formations (16.516.0 Ma), deposited in a shallow marine environ-
ment, include resedimented, syn-eruptive, garnet-bearing dacitic volcaniclastics (originating mostly from small-scale
ignimbrite eruptions) and coeval, garnet-bearing dacitic lava domes, sometimes with their volcaniclastic aprons. As
the eruptions filled the marine basin, subaerial dacitic-andesitic volcaniclastics, comprising minor ignimbrites and
different types of debris-flow deposits were also deposited. A part of the latter may have been related to the formation
of two or three medium-sized calderas. The second stage (16.014.5 Ma) was characterized by andesitic lava dome
activity terminated by a hydrothermal event. During the first half of this stage, a ca. 30° CCW rotation occurred. The
third stage produced the most voluminous, moderately explosive, andesitic basaltic andesitic High Börzsöny sub-
aerial lava dome complex erupting up to the Badenian/Sarmatian boundary (ca. 13.7 Ma). Correlation of K/Ar geo-
chronological and volcanological data shows that lava dome activity of the second and third stage may have been
coeval with marine sedimentation in the southern Börzsöny.
Key words: Miocene calc-alkaline volcanism, Börzsöny Mountains, volcanology, geochemistry, paleomagnetism, K/Ar
In the past years, a renewed scientific interest has resulted in
a number of publications on the geological history of the Mi-
ocene dacitic-andesitic volcanism of the Börzsöny Moun-
tains, North Hungary. However, a synthesis of different sci-
entific approaches to this very complex volcanic area has not
been presented, in spite of contributions to the relationship
between timing of volcanism and ore mineralization (Korpás
& Lang 1993), volcanological aspects related to structure
(Karátson 1995, 1997), stratigraphical problems (Korpás et
al. 1998) and dividing volcanic formations on maps (Korpás
& Csillag-Teplánszky 1999; Karátson et al. 1999a). In this
paper, on the basis of an integrated research including field
volcanology, paleomagnetic and radiometric measurements,
petrology, geochemistry, gravimetry and paleontology, we
summarize the volcanic evolution and stratigraphy of the
Börzsöny Mountains. Although some open questions have
remained, the complexity of our method may serve as an ex-
ample for studying highly degraded volcanic mountains, like
many in the Inner Carpathian calc-alkaline Volcanic Chain.
2. Geologic and geomorphic setting
The Börzsöny Mountains are among the westernmost and
oldest members of the Carpathian Neogene to Pleistocene
Volcanic Chain (Fig. 1). Xenoliths in the volcanics and partly
borehole data show that the basement consists of carbonate
rocks related to the Transdanubian Mountains to the S and
crystalline schists of the Veporids to the N, separated by the
Diósjenõ line (Balla 1977). These rocks are overlain mostly
by Oligocene and Lower to Middle Miocene sedimentary
formations (predominantly clay, sandstone and gravel; e.g.
Korpás et al. 1998). Underlying the subsequent Middle Mi-
ocene volcanics, these formations crop out mostly along the
eastern margin of the Börzsöny Mountains. The volcanic
rocks are also covered by Middle Miocene (Badenian) lime-
stone and clay marl mainly along the western margin and in
the southern part of the mountains (e.g. Báldi & Kókay 1970;
Korpás et al. 1998).
From the geomorphic point of view (Fig. 2), the Börzsöny
Mountains are characterized by the contrast of the northern
and southern hilly terrain (400600 m) and the central High
326 KARÁTSON et al.
Fig. 2. Shaded relief map of the Börzsöny Mts. Computer-generat-
ed image is based on the digitized 1:50,000 topographic map of
Hungary. Note the well-preserved cone remnant of the High
Börzsöny with its prominent, deeply eroded central depression,
the circular shape of Kemence valley, the radial ridges and valleys
in the SW Börzsöny, the depression bordered by the Pap, Nagy-Kõ
and Nagy-Kõszikla hills in the SE, and the rectangular, NW-SE
and NE-SW-trending valley network mainly in the S.
Fig. 1. Geological setting of the Börzsöny Mts.
Börzsöny, the latter having a medium height (700900 m;
highest point Mt. Csóványos 939 m) and significant relief
energy. The northern hills are bordered by the Ipoly (Ipe¾)
River; the eastern, elevated margin towards the Nógrád Ba-
sin is sharply indicated by a NNE-SSW fault (Czakó & Nagy
1976; Balla 1977); the western part merges in the terraced,
alluvial plain of the Ipoly; and the extended southern hills
and small intermontane basins are terminated by the distinct
Szt. Mihály mountain group (not seen in Fig. 2) facing the
Pleistocene Danube Bend.
3. The volcanic formations and their environment
3.1 Deposits underlying the volcanic successions: pale-
The deposits directly underlying the volcanics belong to
the Karpatian-Lower Badenian (Lower-Middle Miocene)
Egyházasgerge Formation and Nagyoroszi Pebble Formation
of North Hungary. These sedimentary deposits are shallow
submarine (littoral-sublittoral) successions consisting of
sandstone, schlier and minor gravel beds (Császár 1997; Kor-
pás et al. 1998). Gravel intercalations in the Nagyoroszi Peb-
ble suggest that in the NE, dry land was in the close neigh-
bourhood (Korpás et al. 1998).
As for the character of the initial Middle Miocene volcanic
eruptions, the broader paleoenvironment is of great impor-
tance. Prior to the volcanism and during the early phase, a lit-
toral-sublittoral bay to the N, a swamp environment to the S
and a delta front to the E have been distinguished by Korpás
& Lang (1993) and Korpás et al. (1998). On the basis of pre-
viously described and newly found surface outcrops, howev-
er, we cannot see evidence for other than a shallow marine
environment (cf. Báldi & Kókay 1970; Borza 1973):
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 327
(1) In the S, a rich sublittoral marine fauna (53 taxa) was
described from the basal layers of the Kismaros Tuff by Bál-
di & Kókay (1970). The re-examination of the Kismaros fos-
sil material shows that the preferred water depth is known for
20 species: 85 % points to the infra- and circalittoral depth
range, while 15 % can be found in only the infralittoral zone.
The preferred Early Badenian age of the fauna was con-
firmed by Báldi-Beke (1980) on the basis of nannoplankton
(NN5) studies. Similar nannoplankton assemblages were
also mentioned by Báldi-Beke (1980) from the boreholes
Drégelypalánk-2 and Kemence-1. Near Kismaros, at Mári-
anosztra village, Jankovich (1974) described sublittoral ma-
rine fauna (molluscs, echinoids, foraminifers) between the
initial volcaniclastic layers.
(2) In the E, Badenian tuffitic sand and sandstone were de-
scribed in borehole Diósjenõ-2 between 10 and 39.5 metres
(Marczel 1977). The mollusc fauna of the borehole was
briefly mentioned by Báldi-Beke et al. (1980). There is a di-
verse fauna between 10 and 14 metres, similar to the fauna of
the Kismaros Tuff (Chlamys, Glycymeris, Fusus, Anadara,
Tellina). The Early Badenian, sublittoral fauna contains ex-
clusively normal marine species. W of Diósjenõ village (on
the E slope of Boros Hill), we have found a Chlamys-bearing
coarse sand, overlain with undulating, disturbed contact by
pumiceous volcaniclastics. The sand seems to be identical
with the Karpatian Chlamys-bearing sand at Kismaros vil-
lage with no described signs of volcanism there (Báldi &
Kókay 1970). Therefore, the new exposure implies the possi-
ble Karpatian beginning of volcanic activity.
(3) At Szívszakasztó hillslope in the Nagy Valley (a in
Fig. 3: the best outcrop of the contact between the Nagy-
oroszi Pebble and earliest volcanics), the well-rounded peb-
bles mostly quartzite and metamorphics, occasionally
pumice clasts bear the marks of rock borer clams and are
intercalated by fine-grained quartzofeldspatic sand with Os-
trea fragments. This clearly indicates a littoral environment
and a rocky seashore in the vicinity. What is more, fossils,
most frequently, Balanus fragments and marine bivalves
(Isognomon and Venerupis) as well as marine gastropods
(Gibbula and Nassa), have been recovered from the initial,
pumiceous volcaniclastic sequences. Although poorly pre-
served, the identified mollusc and barnacle remains also indi-
cate a shallow, agitated marine environment (shallower, than
at Kismaros or Diósjenõ-7), without any signs of freshwater
In the northernmost Hont Gorge (see Fig. 3), a thick Karpatian
Lower Badenian sedimentary succession underlying the volcaniclas-
tics crops out (Vass & Marková 1966; Borza 1973). Nearby, beneath
the volcaniclastics of the Bába Hill (Fig. 3), pebbles of the underlying
conglomerate also show the marks of rock borer clams, and the em-
bedded mollusc fauna shows similarity to the Nagy Valley fauna
(Isognomon [=Perna], Ostrea, Anomia, Venus, Venerupis, Turritel-
la, Balanus, solitary corals: Borza 1973).
(4) The earliest volcanism also started in a shallow marine
environment in the neighbourhood of the Börzsöny Mts. In
the Burda (Helemba) range SW of the Börzsöny Mts.,
Koneèný & Lexa (1994) inferred a water depth of ca. 200 m.
To the S, in the coeval Visegrád Mts., an Ostrea bed has been
discovered recently beneath the first pumiceous volcaniclas-
tics of the Holdvilág Gorge (Badics et al. 2000).
Although the initial volcanism should therefore have been
submarine, the calculated shallow water had to be rapidly in-
filled, if the up to 200 m thickness of the fossiliferous volca-
niclastic deposits is considered. This implies that the marine
basin rapidly became a changeable coastal environment.
The volcanic formations are discussed below in two groups:
volcaniclastic successions and massive rocks (lavas and sub-
volcanic bodies). These categories largely fit with the early
andesitic-dacitic and the late andesitic petrographical cat-
egories of Csillag-Teplánszky & Korpás (1982) and Korpás &
Lang (1993), adding that among the earliest formations, there
are also massive rocks. Correlating all available data, a three-
stage volcanism has been proposed by the present first author
(Karátson 1995; Karátson et al. 1999a). As a clue for the fol-
lowing discussion, a simplified volcanological map with such
a division is presented (Fig. 3).
3.2 The volcaniclastic successions
The volcaniclastic deposits of the Börzsöny Mts. (without
the High Börzsöny breccias) cover roughly 2/3 of the area
(see Fig. 3). A general sedimentological feature of them is
the succession of stratified and/or graded beds with rapid
changes in particle size and type (e.g. juvenile/lithic clast ra-
tio), suggesting complex volcanic-sedimentary processes.
Complexity is accentuated by the varied lithology of clasts
ranging from andesite to dacite (see Appendix and Sec-
tion 4). Facies relations of the volcaniclastics in six selected
lithological logs are presented in Fig. 4. In the following, we
briefly present the proposed stratigraphical units of the vol-
caniclastic deposits, then discuss their volcanology and time-
Nagy Valley Volcanogenic Sandstone
Lithological logs ac show that the deposits overlying the
Karpatian and Lower Badenian sedimentary succession are
composed of stratified/cross-stratified sandstone and minor
conglomerate beds with normal- to reverse-graded volcanic
clasts and minor pumice content. The Szívszakasztó locality in
Nagy Valley (Fig. 4a) and the E slopes of Boros Hill, as men-
tioned previously, expose the nonvolcanic, fossil-bearing un-
derlying sandstone and gravel beds as well. On the basis of
these and other scattered outcrops from Kismaros through
Márianosztra and Diósjenõ to Hont villages, a continuous suc-
cession from non-volcanic to volcanogenic deposits can be in-
ferred in the entire Börzsöny. In this paper, the fine-grained
volcaniclastic deposits are collectively called Nagy Valley Vol-
canogenic Sandstone. In the marginal parts of the mountains,
it may be overlain and/or intercalated by the Kismaros and Ke-
mence Tuffs and the Nagy-Kõ Hill Volcaniclastic Breccia (see
below and logs ac), whereas in the central areas, especially in
the High Börzsöny, its existence is only inferred by boreholes
under thick, subsequent volcanic formations.
Kismaros and Kemence Tuffs
Logs df in Fig. 4 have been selected to represent surface
outcrops and boreholes that contain moderate to large amount
328 KARÁTSON et al.
Fig. 3. Simplified volcanological map of the Börzsöny Mts. with paleomagnetic division of volcanic rocks.
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 329
(2030 %) of pumices in successive beds (see Appendix).
These deposits tuffs and lapilli tuffs, 510 m thick in indi-
vidual field exposures and up to 200 m thick in deep bore-
holes are characterized by commonly graded/stratified,
rarely massive beds, poor sorting of small (mm-cm-sized),
subangular to angular pumices and moderate amount (10
20 %) of lithics, lack of thick, well-sorted horizons, presence
of cm- and rarely dm-sized prismatically jointed clasts and,
in Királyrét (in the heart of the mountains: Fig. 3), the exist-
ence of an embedded lag breccia. Welding has been reported
(with some uncertainty) only in boreholes (Gyarmati 1976).
On the basis of borehole data, the pumiceous volcaniclastic
deposits are the thickest and most widespread in the
Börzsöny. Characterized by an overall dacitic-rhyodacitic
composition (see Section 4 and Karátson & Németh in print),
these sequences are ranging in mineral assemblages from
garnet-bearing pyroxene biotite amphibole dacite (e.g. at the
southern exposures: Kismaros Tuff, named first by Báldi &
Kókay 1970) to garnet-bearing pyroxene amphibole ± biotite
dacite (e.g. in the Kemence Valley: Kemence Tuff). However,
given the poor exposure conditions and no widespread mark-
er horizons, they are not divided in Fig. 3.
Nagy-Kõ Hill Volcaniclastic Breccia
In and around a large number of marginal ridges, the pumi-
ceous deposits give place to, or are interbedded with volca-
nic breccias totalling 50100 m in thickness (see logs e and f
and upper sections of a, c and d in Fig. 4). These sedimento-
logically highly variable breccias (see Karátson & Németh in
print) consist of dm- to m-sized andesite and minor dacite
clasts (occasionally with garnet), are frequently bedded and
graded, and have a fine-grained, occasionally stratified/
cross-stratified/laminated matrix sometimes with pumice
fragments. Not distinguished or named in previous studies,
this breccia is collectively called Nagy-Kõ Hill Volcaniclastic
Breccia in this paper.
The three above mentioned formations are proposed to
have been deposited in close time-space relationship. As a
detailed lithofacies study has pointed out (Karátson &
Németh in print), they represent a rapid evolution of a num-
ber of small- to medium-sized silicic volcanic centres infill-
ing the shallow submarine environment with pyroclastic and
Features of the Kismaros and Kemence Tuffs suggest
small- to moderate-scale ignimbrite eruptions occurring in
the close vicinity. As for the depositional processes, howev-
er, the frequent intercalation of pumiceous and volcanogenic
sandy-clayey ±fossiliferous material in graded/stratified beds
argues for the existence of resedimented syn-eruptive volca-
niclastics (McPhie et al. 1993). These may have been depos-
ited mostly and initially subaqueously, partly subaerially by
gravity-driven and/or water-supported volcaniclastic mass
flows (Karátson & Németh in print).
How are these syn-eruptive volcaniclastics related to the proposed,
primary ignimbritic origin? In recent literature (e.g. Cas & Wright
1991; White & McPhie 1997), there are three criteria for identifying
ignimbrites: 1) presence of pyroclasts; 2) facies characteristics indi-
cating deposition from a density current; 3) evidence for gas-support-
ed (i.e. hot) transport of pyroclasts. Whereas the former two require-
ments (cf. Fig. 4) are met for the Börzsöny volcaniclastic deposits, the
third one (i.e. welding, segregation pipes, fiammes, high-temperature
crystallization structures, etc.) is not or is uncertain. Nevertheless, the
significant amount of fresh, angular pumice (fragment)s, the radially
jointed blocks in many places (see logs in Fig. 4 and the E margin of
the Börzsöny), as well as the presence of the mentioned lag breccia at
Királyrét, suggest primary (ignimbrite eruption-fed) origin and hot
conditions in situ or not too far away. Direct deposition from subaque-
ous (cf. Cas et al. 1998; Legros & Druitt 2000) and even subaerial py-
roclastic flows may also have occurred (e.g. Kismaros [upper section],
Királyrét, Magyarkút). Our interpretation is in accordance with the
submarine pumice flows proposed by Koneèný & Lexa (1994) for
the neighbouring Burda Mts., but refines it and also all the former
views that regarded the first-stage deposits mostly as pyroclastics (e.g.
Korpás & Lang 1993; Karátson 1995; Korpás et al. 1998)
Further away from the eruptive vents, syn- and inter-eruptive
resedimentation resulted in deposition of the Nagy Valley Vol-
canogenic Sandstone. In the N, northward from Kemence Val-
ley, and in the SE, southeastward from Nagy-Kõ Hill, field ob-
servations show that the fine-grained volcanogenic deposits
are progressively better stratified, with pumice decreasing in
size and quantity and the strata are thinner and more graded.
These sedimentological features fit in with the existence of
evolved volcaniclastic aprons. In other words, the Kismaros
and Kemence tuffs and the Nagy Valley sandstone may repre-
sent end-member formations of proximal/more primary and
distal/more reworked facies, respectively. This relation is
more ambiguous in the central part of the mountains where
subsequent massive rocks overlie them.
Proximity of the tuffs seems to be supported by certain
types of the Nagy-Kõ Hill breccia. This breccia is interpreted
as deposited from high-concentration mass flows (Karátson
& Németh in print), that is debris flows (both submarine/sub-
aerial), lahars and hyperconcentrated streamflows (see Fig.
4; cf. Pierson & Scott 1985; Smith & Lowe 1991). Debris av-
alanche deposits to the S have also been identified (Karátson
& Németh in print). The breccias may have been among the
final products of the emerging first-stage paleovolcanic com-
plex. For the primary origin of certain covering breccias in
an elevated position those with monolithological compo-
sition and abundant pumices in the matrix we propose ex-
plosive eruption-associated destructive processes (e.g. dome
or sector collapses) resulting in small- to medium-sized
calderas (such as at Mt. Pinatubo, the Philippines, 1991: Ne-
whall & Punongbayan 1996; also see Section 5). In other
breccia types, the Nagy Valley sandstone is interbedded up-
section by monolithological breccia and the matrix is free of
pumices but prismatically jointed blocks are frequent (e.g. at
Kámor and Gömbölyû-Kõ hills: see c in Fig. 4). These brec-
cias are interpreted as subaqueous volcaniclastic debris-flow
deposit originating from either minor dome collapses or hy-
aloclastite formation. A more detailed study may identify
many types of syn- and post-eruptive debris flows, although
their mapping and correlation are highly complicated by the
lack of exposures. A part of the breccias (of more basic lithol-
ogy) may have originated from, or mixed with, the material of
the emerging, subsequent High Börzsöny edifice, for example
those exposed on its upper southern slopes (see point 3.3).
330 KARÁTSON et al.
Fig. 4. Selected lithological logs of the first-stage volcaniclastics. Note scale differences. For locality, see Fig. 3.
Valley (simplified). A Non-volcanic sand and conglomerate (submarine). B Mixed volcaniclastic debris-flow deposits developing
from A. C Mollusc-bearing volcanogenic sandstone (submarine). D As C but without fossils. E Volcanogenic mudstone (subma-
rine). F Volcaniclastic debris-flow deposit (submarine? subaerial?). G Debris-flow scour fill. H Similar to E but probably subaerial.
K Resedimented syn-eruptive debris-flow deposit (subaerial?) with scour fill. L Sandy volcaniclastic bed originated probably from flu-
vial deposition. M Volcaniclastic debris-flow deposit (subaerial).
Lohanc, Nagy Valley. A Volcaniclastic debris-flow deposit (sub-
marine?). B Volcanogenic sandstone (submarine?). C Volcaniclastic conglomerate (submarine?). D Hyperconcentrated streamflow
deposit. E Fluvial deposit. G Hyperconcentrated streamflow deposit. F, H Volcaniclastic debris-flow deposit (subaerial).
Hill E slope. A, B, C Volcanogenic sandstone intercalated with units of submarine volcaniclastic mass-flow deposits. D Probably re-
sedimented syn-eruptive volcaniclastic debris-flow deposits (submarine). E Volcanogenic sandstone beds with minor volcaniclastic mass-
flow deposits (submarine). F Syn-eruptive volcaniclastic debris-flow deposits perhaps originated from block-and-ash flow deposits (sub-
marine? subaerial?). G Amphibole pyroxene andesite lava flow.
NW-SE gully W of Kismaros, the principal exposure of Kismaros
Tuff. A volcanogenic sandstone (submarine). B Flow units of pumiceous resedimented syn-eruptive volcaniclastics (submarine). C
Pumiceous resedimented syn-eruptive volcaniclastics (submarine). D Resedimented syn- or inter-eruptive volcaniclastics (submarine). E
Volcaniclastic debris-flow deposit (submarine?). F Volcanogenic sandstone (submarine). G Debris-flow deposit (submarine?). L
Pyroclastic-flow deposit or resedimented syn-eruptive volcaniclastics (subaerial?).
Cicõke hillslope in Kemence Valley. A Resedi-
mented syn-eruptive volcaniclastic mass-flow deposit (submarine). B Flow unit of pumiceous resedimented syn-eruptive volcaniclastic
deposit (submarine). C, D Volcaniclastic debris-flow deposits (submarine). E Volcaniclastic debris-flow deposit (submarine? subaeri-
al?). F Volcanogenic sandstone (lacustrine?). G Volcaniclastic debris-flow deposit, i.e. lahar (subaerial?). H Volcaniclastic debris-
flow deposit (subaerial).
Nagy-Kõ Hill (compiled and simplified from two exposures). A Volcaniclastic debris-flow deposit (subma-
rine). B, C Volcanogenic lapillistone and sandstone, respectively, redeposited perhaps from pyroclastic fall. D Pumiceous
resedimented syn-eruptive volcaniclastic mass-flow deposits (submarine? subaerial?). E Flow units of pumiceous resedimented syn-
eruptive volcaniclastics (submarine? subaerial?). F Volcaniclastic debris-flow deposit (subaerial?). G Volcaniclastic debris-flow de-
posit interbedded with fluvial streamflow deposit (subaerial). H Fluvial and hyperconcentrated streamflow deposit (subaerial).
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 331
3.3 Massive rocks
In this point, we distinguish between massive lava- and
subvolcanic rocks in the SW, S, SE and in part N Börzsöny
and principal constituents of the High Börzsöny.
Dacite and andesite lava domes
Although some early workers already proposed a surficial volcan-
ism for the S and W Börzsöny, resulting in lava domes (e.g. Papp
1933a,b), most later authors (e.g. Balla & Korpás 1980; Korpás &
Lang 1993) used the terms shallow intrusive body and vent core
(or vent infill).
More recently, Korpás et al. (1998) refined the classification de-
scribing subvolcanic bodies and extrusions. The rocks in question
crop out mostly in isolated hills, penetrating the volcaniclastics or the
underlying sedimentary deposits (see the same stratigraphy in the
nearby Burda Hills, Slovakia: Koneèný & Lexa 1994). The various al-
teration of these rocks in the W and central Börzsöny, related to the
Nagybörzsöny hydrothermal ore mineralization, is not dealt with here,
since detailed works are available (e.g. Csillag-Teplánszky et al. 1983;
Korpás & Lang 1993).
According to our thin section studies (see Appendix), all
these rocks have porphyric texture. They have either more or
less oriented crystals in a glassy groundmass (ca. 2/3 of the
samples) or the groundmass is microholocrystalline (ca. 1/3).
Intrusive rock types have not been found; rarely, holocrystal-
line matrix occurs. At the same time, there are field observa-
tions suggesting an extrusive origin. These observations in-
clude coeval volcaniclastic rocks at same topographic level
(e.g. below Gömbölyû-Kõ Hill, see before) and glassy ground-
mass in thin sections around some hills otherwise character-
ized by microholocrystalline rock texture (e.g. Nagy-Koppány
and Nagy-Sas hills). In Pap Hill, one of the samples has micro-
holocrystalline, another (at the top) glassy groundmass. All
these data suggest that the majority of these rocks belongs to
more or less eroded extrusions/lava domes (e.g. Nógrád Cas-
tle Hill, Nagy-Pogány Hill, Nagy-Sas Hill, Pap Hill lava
domes: see Fig. 3). Some holo- and microholocrystalline ma-
trix could, indeed, indicate shallow subvolcanic bodies, but
even in these cases we prefer the interpretation of exposed
roots of eroded lava domes consisting of the more crystallized,
resistant basal part. On the other hand, for the majority of the
domes, we have no data to give a paleogeographical evalua-
tion. However, as treated in point 3.2, the outer facies of some
first-stage domes are volcaniclastic and embedded in volcanic
sandstone (see c in Fig. 4). For at least these cases, the pale-
oenvironment may have been subaqueous.
The High Börzsöny andesitic lava dome complex
In the High Börzsöny, typical facies of a highly eroded
subaerial lava dome complex have been identified (Karátson
1995, 1997; Karátson et al. 1999a). These include (a) coarse
pyroclastic breccias being the roots of collapsed domes, (b)
proximal facies of block-and-ash flow deposits and (c) hori-
zontal or subhorizontal beds of lava flows. (It is worth men-
tioning that a and b have been termed pseudo-agglomerate
and agglomerate in some previous Hungarian literature.) The
unsorted vent breccias are apparently matrix-free and contain
large, rounded boulders due probably to hot, in situ fragmen-
tation (Karátson et al. 1999b). In contrast, the block-and-ash
flow deposits have more angular clasts and a significant
amount of fine-grained matrix, and show typical features of
block-and-ash flows (monolithological composition, slightly
vesiculated clasts, interbedded tuff layers, frequent prismatic
jointing of blocks, occasional stratification and reverse grad-
ing). Their best localities, preserved as small radial paleoval-
ley-filling, now exhumed rock towers, are found mostly on
and around the rim of the High Börzsöny central depression
In certain places of these radial exposures, (a) and (b) show obvious
transitions. Transition between the two facies can be detected well by
investigating clast orientation (Karátson et al. 1999b): there is a rapid
improvement in orientation from vent breccias to block-and-ash flow
The lava flows of the High Börzsöny have already been
recognized by Pantó (1970), Balla (1978), Balla & Korpás
(1980) and others. They are commonly platy jointed, occa-
sionally with well-developed flowage structures (e.g. at Plés-
ka ridge in the NE High Börzsöny), in accordance with a
subaerial emplacement. The lava flows crop out differently
in the W and S-E Börzsöny. In the former, all lava flows are
interbedded with block-and-ash flow deposits; in the latter,
the N-exposed slopes are covered mostly with lavas whose
dip is largely identical with that of the slope (cf. Pantó 1970).
This implies the northward tilting or faulting of the S-E
Börzsöny (also see Section 5). Along the S and E rim of the
High Börzsöny, alternating block-and-ash flow deposits and
lava flows crop out in similar topographic levels indicating
truncated, exposed paleovalley fillings.
Typical thin sections of the High Börzsöny rocks are de-
scribed in the Appendix. The absence of garnet and biotite as
well as their presence as xenoliths cearly indicates the subse-
quence of High Börzsöny relative to the biotite- and garnet-
bearing marginal (and probably underlying) volcanics. The
rather uniform rock type typically amphibole pyroxene
andesite has some differences in the W and S-E parts: in
the W High Börzsöny, the andesite samples are more am-
phibole rich and the hypersthene frequently occurs without
augite. This petrographic difference, some dip directions of
platy jointed lavas and the too large caldera diameter relative
to other simple erosion craters of the Carpathians (Karátson
1995, 1996) suggest more than one eruptive vent (also see
sections 5 and 6).
In addition to the above, there is a distinct type of massive
rocks in the High Börzsöny. These are up to 30 m narrow,
some tenssome hundreds metres long dykes (Balla 1978;
Csillag-Teplánszky et al. 1983; Korpás et al. 1998). Petro-
graphically, they are mostly amphibole andesites. The
groundmass in thin sections is more or less glassy so the ex-
posures may represent the near-surface parts of the dykes.
Repeated K/Ar dating of the Mt. Várbükk dyke (Fig. 3) has
given an older age (14.314.7 Ma) than the majority of the
High Börzsöny (Table 2). This older age may be explained
by an amphibole-rich early dyke magma later being com-
pleted by pyroxene during the High Börzsöny eruptions. This
explanation is in contrast to a previous concept that the am-
phibole andesite dykes are the final volcanic products (e.g.
Balla & Korpás 1980; Csillag-Teplánszky et al. 1983).
332 KARÁTSON et al.
4. Geochemistry and petrogenesis
So far, no detailed geochemical work has been published for the
Miocene volcanic rocks of the Börzsöny Mts. Downes et al. (1995)
used 4 samples from the Börzsöny Mts. in the discussion of the
geochemistry and petrogenesis of the Miocene volcanic rocks of the
Inner Carpathian arc. Among them, sample #113 from the Nógrád
Castle Hill has an unusual composition and deviates from the trends
shown by the Börzsöny volcanic rocks. A new sample from the same
locality (Table 1) has been analysed and we have got a different
geochemical composition than #113 (Table 2 in Downes et al. 1995)
that fits better with the geochemical trends of the Börzsöny volcanic
We have analysed 18 samples selected to cover the petrographical
and assumed temporal range of the volcanic activity and to be as fresh
as possible. The geochemical analyses were carried out in the
Geochemical Laboratories of the Royal Holloway University of Lon-
don (U.K.) and in the XRAL Laboratories, Toronto (Canada). The ma-
jor elements and some trace elements (Ni, Cr, V, Rb, Ba, Pb, Sr, Zr,
Nb, Y, Th) were determined by XRF spectrometry using fused discs
for major elements and pressed pellets for trace elements. The rare
earth elements were determined by neutron activation analysis (Cana-
da) and by ICP-AES (U.K.). The major element composition of glass-
es (pumices and glass shards) was analysed by a JEOL Superprobe
JXA-8600 WDS with an accelerating voltage of 15kV at a beam cur-
rent of 10 nA (beam diameter 5 mm) at the University of Florence (It-
aly). Data were corrected using the procedure of Bence & Albee
(1968). More details about the analytical conditions can be found in
Mason et al. (1996) and in Harangi (1999).
The studied samples are usually fresh as shown by the low
LOI. The more silicic volcanic rocks of the marginal parts,
however, usually have a higher LOI content partly because of
the abundance of hydrous minerals, such as biotite and am-
phibole and partly due to secondary alteration. Nevertheless,
their chemical compositions may reflect the original charac-
teristics since no correlation can be observed between the
LOI content and the mobile low-field strength element
(LFSE, e.g. Rb, Ba, Pb, Sr) concentrations.
Geochemical compositions of representative samples from
lava flows and pumiceous volcaniclastics in marginal parts
of the mountains and lava flows and clasts from block-and-
ash flows of the High Börzsöny are presented in Table 1. As
Table 1: Representative chemical composition of volcanic rocks from the Börzsöny Mts. Numbers of localities are displayed in Fig. 3.
= basaltic andesite,
= dacite, am = amphibole, py = pyroxene, bi = biotite.
(am = breccia
(biamhyp @ lava
(biam = breccia
(biam = lava
(py >= lava
= lava flow)
major elements (wt. %)
trace elements (ppm)
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 333
Fig. 5. Classification of the volcanic rocks of Börzsöny Mts. us-
ing the TAS scheme (Le Bas et al. 1986) and the SiO
gram (Gill 1981).
Fig. 6. Comparison of glass composition of pumices and glass
shard from the pumiceous volcaniclastic deposit of Nagy-Kõ Hill
with glass analyses from Early to Middle Miocene ignimbrites of
Hungary (a) and with the host volcanic compositions of the
Börzsöny Mountains (b). Glass compositions are normalized to 100
wt. % (data from Harangi unpublished). Symbols for b as in Fig. 5.
shown in Fig. 5, the volcanic rocks of the Börzsöny are pre-
dominantly andesites. Some garnet-bearing samples are clas-
sified as dacite and, in the High Börzsöny, there are a few
basaltic andesite occurrences. It is worth comparing these
rocks to the similar-aged volcanic rocks of the nearby Viseg-
rád Mts. that are thought to have undergone similar evolution
(cf. Korpás et al. 1998). The Börzsöny volcanics have higher
total alkali content at a given SiO
. This difference is more
pronounced using the SiO
O diagram (Fig. 5 below).
The Börzsöny volcanic rocks belong to the high-K calc-alka-
line series and have systematically higher potassium content
than those of the Visegrád Mts. The glass composition of
pumices and glass shard from the pumiceous volcaniclastic
deposit of Nagy-Kõ Hill shows higher SiO
tents compared to the bulk volcanic rocks (Fig. 6b). They re-
semble the glasses of the Holdvilág Valley ignimbrite in the
Visegrád Mts. On the other hand, they are less silicic than
glasses from Early to Middle Miocene ignimbrites of Cserhát
and Bükk Foreland regions (Fig. 6a; the latter regions are
named in Fig. 1).
Major and trace element variations with increasing SiO
trated in Fig. 7. CaO, Fe
and MgO are compatible throughout the
series. These trends are consistent with fractional crystallization of
olivine and clinopyroxene in the most basic magmas. Al
show a clear trend, it remains roughly constant suggesting that plagio-
clase did not fractionate, but accumulated heterogeneously in the
Börzsöny volcanics. Among the trace elements, the negative trend of Sc
and Y implies a strong control of amphibole and/or garnet during mag-
ma evolution. Some scatter can be observed in the distribution of Rb,
Ba and Sr. They show positive or constant trends indicating that they
behaved incompatible, that is no significant plagioclase and K-feld-
334 KARÁTSON et al.
spar fractionation occurred. The scatter in the variation of La and Th suggests that
other processes than fractional crystallization may have also operated, such as mag-
ma mixing and partial melting of heterogeneous source rocks.
N-MORB normalized trace element distribution of the Börzsöny volcanics is pre-
sented in Fig. 8. In general, they show fairly uniform trace element patterns charac-
terized by enrichment of LFS elements, negative Nb-anomaly and strong positive Pb
anomaly. These features are typical of subduction-related volcanic rocks. The only
significant difference can be observed in the Y-Yb-Lu range. The garnet-bearing vol-
Fig. 7. Variation of some major and trace elements with increasing SiO
Symbols as in Fig. 5.
canics show a strong depletion in these elements sug-
gesting early garnet fractionation or residual garnet dur-
ing partial melting. In contrast, the garnet-free volcanic
rocks have very similar trace element signatures.
The geochemical composition of the volcanic
rocks from the Börzsöny indicates only slight
differences between the volcanic products of the
petrographically/volcanologically different parts
of the mountains, but they are not cogenetic. The
rocks of the volcaniclastic successions as well as
the garnet-bearing marginal lava domes are
more silicic and characterized by lower Zr/Nb
ratio (1214.6) than the rocks of High Börzsöny
and some other domes (Zr/Nb=1619).
The Zr/Nb ratio suggests a moderately en-
riched mantle source for the primary magmas of
the Börzsöny volcanics. As far as subduction-re-
lated metasomatism is concerned, it is a subject
of debate whether the Middle Miocene volcanic
activity in the W segment of the Carpathian
Neogene Volcanic Chain took place due to the
active subduction of the European plate beneath
the ALCAPA microplate of the Pannonian Basin
(e.g. Balla 1981; Szabó et al. 1992; Downes et
al. 1995) or it was a response of the overall ex-
tension of the Pannonian Basin (e.g. Lexa &
Koneèný 1974; Lexa et al. 1993; Harangi et al.
1998; Harangi 1999). In the latter case, melt gen-
eration occurred in the metasomatized lithos-
pheric mantle due to the thinning of the lithos-
phere in the Middle Miocene syn-rift phase of
the Pannonian Basin.
5. Volcanic structure and landforms
In this section, we attempt to correlate volcanic struc-
ture and gravimetry. Although there have been two rele-
vant contributions to structural geology [a photo-tecton-
ic interpretation by Czakó & Nagy (1976) and a
deep-structure investigation on a geophysical basis by
Balla (1977)], no modern correlation of tectonic and
volcanic structure using the updated gravimetric data
base has been carried out. Unfortunately, large-scale
postvolcanic tectonic movements (faults, uplifting) of
the broad vicinity (e.g. Fodor et al. 1999) obviously
overprinted the original situation; in addition, interpre-
tation can hardly be supported in the field due to the
general lack of microtectonic features.
Gravitational anomalies of the Börzsöny and
its surroundings contain significant regional ef-
fects not related to the Miocene volcanic activi-
ty. This is due to the high-density Lower Paleo-
zoic crystalline rocks to the N and low-density
Mesozoic carbonate rocks to the S, separated by
the already mentioned Diósjenõ line as a gradi-
ent zone. Anomalies are also caused by different
depths of the basement.
The regional effects, however, can be moder-
ated by filtering. The filtered residual gravime-
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 335
try map (Fig. 9) is more free from regional effects. The
sources of the residual anomalies can be interpreted either as
small-scale horizontal changes in the basement (e.g. the
maximum of Naszály Hill), or as relatively large, high-density
andesite bodies (lavas or subvolcanic rocks). These two fac-
tors may conspire in the central and SW part of the Börzsöny
where it is obvious that igneous/subvolcanic masses intruded
into the host rocks. When investigating the latter area in detail,
not only a single but also several individual bodies can be dis-
tinguished. These bodies largely correspond to topographic el-
evations located mostly in the SW Börzsöny (Fig. 3: e.g. Sas,
Só, Koppány, Csák, Hegyeshegy hills, interpreted as more or
less exposed lava domes: Karátson 1995). The topographically
most prominent Pap Hill lava dome is located obliquely
(northward) above the gravimetric maximum W of Szokolya.
Some other, assumed centres in the N and E are not supported
The other large concentric maximum zone can be seen un-
der the High Börzsöny area. This maximum may be related to
subvolcanic bodies under the High Börzsöny edifice. The ed-
ifice itself with a somewhat rectangular-shaped depression
(Figs. 23) was identified first by Balla (1978) who proposed
a stratovolcanic cone with an erosion caldera. As demon-
strated in point 3.3, the cone can be termed rather as a lava
dome complex with a number of eruptive vents. The large-
scale identity of the cone, however, is well supported by
When reconstructing the original form in more detail, it is
important that the S-E part of the edifice seems to be tilted
northward (see earlier, and Balla 1978, Karátson 1995): the
initial, pumiceous volcaniclastics crop out 650700 m a.s.l.
on the S-SE slopes and ca. 300 m a.s.l. along the Kemence
Valley. Taking this into account, and according to (1) the cal-
culated 13001400 m absolute height (Karátson 1996, 1997),
(2) the ca. 800 m inferred relative height and (3) the 5.5 km
basal radius, the volume of the (simple) original cone could
be some 25 km
. Because the residual gravimetry map and
field observations show no trace of collapse, the erosional or-
igin of the central depression is very likely. On the basis of
the classification of Karátson et al. (1999c), we think the
present depression has formed from a number of distinct cra-
ters/scarps by long-term fluvial erosion, whereas the western
sector may have avalanched to the W (Karátson 1995, 1997).
The interpretation of the volcanic structure is much more difficult
when trying to infer larger-scale forms, that is calderas. Although sev-
eral caldera-forming mechanisms, such as downsag, plate- or piston-
like, chaotic, piecemeal, etc., have been distinguished (e.g. Walker
1984; Scandone 1990; Lipman 1997; Moore & Kokelaar 1998), the
resulting gravimetry is largely similar (e.g. Rymer & Brown 1986; De-
plus et al. 1995): characteristic minima zones inside and outside the
caldera rim, due to shattered rocks of the collapse and/or to the low-
density infill (e.g. pumiceous volcaniclastics, lake sediments). In the
Börzsöny Mts., Balla & Korpás (1980) and Korpás & Lang (1993)
proposed a large number of calderas, whereas Karátson (1995, 1997)
argued for a single depression coalesced from three smaller ones.
In the residual gravimetric map of the Börzsöny Mountains,
instead of unambiguous caldera structures, extended, quiet
minima areas can be seen outside the two maxima zones
which should correspond to low-density, pumiceous, volcano-
genic material. However, in the SE Börzsöny around
Fig. 8. N-MORB (Pearce & Parkinson 1993) normalized trace el-
ement distribution of representative samples from different stages
of the volcanism of Börzsöny. Symbols as in Fig. 5.
Table 2: K/Ar ages from the Börzsöny Mountains measured since the review paper by Korpás & Lang (1993). For locality, see Fig. 3. bi
= biotite; am = amphibole; py = pyroxene; w.r. = whole rock.
I Bajdázó quarry lava dome
bi am dacite
II Kopasz Hill lava dome
am py bi andesite
III Száraz-fa-bérc lava dome
am py andesite
IV Mt. Hollókõ pyroclastic breccia
am bi andesite
V Mt. Csóványos W pyroclastic breccia
am py andesite
VI Inóc quarry lava flow
am py andesite
VII Rózsa adit,
borehole 259 m
Pécskay & Nagy 1993
VIII Kemence Valley central part breccia
py am andesite
IX Mt. Lófarú dyke
X Nagy-Kõ Hill volcaniclastic breccia
bi am andesite
XI Magyarkút ignimbrite
dacite lapilli tuff
XII Kámor Hill volcaniclastic breccia
py am andesite
XIII Perõcsény E lava dome/flow
bi am andesite
XIV Kemence Valley W lava flow
am py andesite
336 KARÁTSON et al.
Szokolya village, a prominent local depression appears that
may be related to a fault-bounded caldera structure (Nagy-Kõ
Hill caldera: Karátson 1997; in part corresponding to the
Szokolyahuta centre of Ferenczi 1936 and the Börzsö-
nyliget stratocone of Balla & Korpás 1980). Although subse-
quently formed tectonic lines may have overprinted it, the
caldera structure seems to be supported by the presence of (a)
concentric, low topographic ridges in the E and S (see Fig 3:
Karátson 1995, 1997), (b) resistant, commonly monolithologi-
cal breccia cover on these ridges interpreted as primarily relat-
ed to caldera formation (Fig. 4f: Karátson 1997; Karátson &
Németh in print), (c) occurrence of lag breccia at Királyrét, (d)
radial flow directions of submarine pyroclastic flows at Kirá-
lyrét, Nagy-Kõ Hill and Kismaros obtained from magnetic
anisotropy measurements (see Fig. 3) and (e) postvolcanic
lake sediment infill in the Szokolya basin (e.g. Ferenczi 1936).
In the N, along the Kemence Valley, the arcuate, flat ridge
of the valley (Balla & Korpás 1980) and the presence of sim-
ilar breccia with pumiceous matrix (Fig. 4e: Karátson 1997)
also suggest an (eroded, retreated) caldera rim section. In the
SW, in relation to the gravimetric maximum zone, geomor-
phic features (i.e. radial ridges and valleys, presence of
planèzes) as well as pumiceous deposits in radial valleys
have been proposed to represent an outer caldera slope
(Karátson 1997). These latter (N and SW) caldera rims, how-
ever, are not seen or are very uncertain in the gravimetric
map, suggesting that dome/sector collapse rather than
caldera collapse occurred.
6. Geochronology of the volcanism
6.1 K/Ar datings and their evaluation
During the last decades, over 100 K/Ar determinations have been
carried out on whole rock samples and mineral fractions from the
Börzsöny Mts. The main purpose of K/Ar dating was to obtain abso-
Fig. 9. Residual gravimetry map of the Börzsöny. Gravity anomalies of the Börzsöny Mts. have been filtered to intensify the short wave-
lenght anomalies (less than 10 km) and to suppress those having longer wavelength (i.e. larger-scale regional effects). This procedure
does not mean that only anomalies smaller than 10 km (horizontally) are displayed, because the anomalies of the individual bodies are
not sinusoid in character (every anomaly has a Fourier spectrum), therefore the regional anomalies may also have harmonics (i.e. short
wavelenght components) that are not filtered. Density for correction: 2400 kg/m
. Symbols as in Fig. 3.
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 337
lute ages for the time and duration of volcanic activity and to date hy-
drothermal processes and ore mineralization. Dating was carried out
mainly in the Institute of Nuclear Research of the Hungarian Academy
of Sciences (ATOMKI), Debrecen, and partly in the Geological Sur-
vey of Israel (GSI), Jerusalem. The first determinations from ATOMKI
were published by Balla et al. (1981); a summary with detailed results
of both laboratories was presented by Korpás & Lang (1993).
All the available K/Ar data, arranged in three volcanic
stages proposed by Karátson (1995), are shown in Fig. 10.
The K/Ar ages obtained since the publication of Korpás &
Lang (1993) are given in Table 2. Details of instruments and
methodology as well as calibration were published by
Balogh (1985). Atomic constants suggested by Steiger &
Jäger (1977) have been used for calibrating age.
Stratigraphic considerations (e.g. Báldi & Kókay 1970)
and paleomagnetic investigations (next point) suggest a
Lower Badenian time for the beginning of volcanism. Many
K/Ar ages also show a Lower Badenian earliest activity.
Among those obtained on unaltered rocks are worth mention-
ing the 16.5 Ma average age of garnet-bearing dacite tuff in
Kóspallag-11 borehole (Hámor et al. 1980), fitting with the
16.4 Ma age of the Middle Rhyolite Tuff (Tar Dacite Tuff) of
the Pannonian Basin (Hámor et al. 1979) and the 16.4 Ma
age of the garnet-bearing volcanic breccia of Kámor Hill em-
bedded in the underlying volcanogenic sandstone (this paper:
Table 2, Fig. 4c).
However, a significant part of the K/Ar ages are younger.
On one hand, a younger age has been confirmed for the well-
known hydrothermal event: 15.1 ± 0.5 Ma was given in the
GSI by averaging K/Ar data on hydrothermally altered rocks
and 14.6 ± 0.5 Ma in ATOMKI by using hydromuscovite
formed in the process (Pécskay & Nagy 1993). On the other
hand, the High Börzsöny andesite and basaltic andesite rocks,
which are not affected by hydrothermal alteration (cf. Korpás
& Lang 1993), systematically give K/Ar ages around or below
14.0 Ma. This Badenian/Sarmatian age poses a major strati-
graphic problem because it has been shown that the South
Börzsöny cover deposits are also Lower Badenian in age (e.g.
Báldi & Kókay 1970; Báldi-Beke et al. 1980; Dulai 1996).
When determining the exact geological age and time span of volca-
nism, we have to consider the following difficulties of K/Ar dating:
1. The given analytical errors of K and radiogenic Ar determinations
are necessarily uncertain, in the best case their values characterize the
average accuracy of the laboratory; the real error of individual measure-
ments can be different.
2. K/Ar and real geological ages may be biased. The radiometric
age is older, when (i) the rock contains undegassed xenoliths, (ii) the
rock contains excess argon, for example in amphibole that crystallized
in a magma chamber; in addition, (iii) during hydrothermal processes,
radiogenic argon can be incorporated. The radiometric age will be
younger, if (i) a long time after rock formation, radiogenic argon is
lost in the course of alteration or heat effect, or (ii) when K is incorpo-
rated in the rock also a long time after its formation.
Some of the rocks from the assumed first volcanic stage con-
tain excess Ar and show strong rejuvenation (e.g. ages on sam-
ples from borehole Perõcsény-26 range from 16.8 Ma to 12.1
Ma). Thus, the mean age of the first-stage rocks cannot be used
for estimating the beginning of volcanic activity; there is no in-
dependent criterion for eliminating overprinted ages from the set
of data. In the next section, however, we present paleomagnetic
measurements that can better resolve this problem.
In contrast, unbiased radiometric age can be calculated
for the end of volcanic activity by averaging K/Ar data from
rocks of the High Börzsöny, unaffected by hydrothermal al-
teration (Table 3). On the basis of volcanological and petro-
graphical considerations (see before), samples from the W
and S-E part of High Börzsöny have been treated separately.
Six ages are available from the W High Börzsöny, all ob-
tained in the ATOMKI. The 14.08 ± 0.81 Ma mean age
shows that the estimated average age in any further measure-
ments would fall within the ± 0.81 Ma range in a 95 % confi-
dence interval. On the other hand, the weighted mean and its
standard deviation is 14.41± 0.71 Ma. This can be regarded
as the geological age for the W High Börzsöny.
Seventeen ages are available from the S-E part of High
Börzsöny, of which 3 were dated in the GSI. Of the 14 ages of
ATOMKI, 2 data are omitted: the amphibole from Godóvár
Fig. 10. Summary histogram of K/Ar ages measured on rocks from
the Börzsöny Mts. 1, 2, 3: volcanic stages corresponding to Fig. 3.
Table 3: Statistical evaluation of K/Ar ages from the High
Börzsöny. Values in Ma. In brackets the number of performed dat-
Mean age (± confidence
weighted average: 13.75±0.32
of the weighted mean
338 KARÁTSON et al.
(17.8 Ma) contains excess Ar; the age of Mt. Magas-Tax
(15.7 Ma) is too old both on the basis of petrography and lo-
cality of the sample and statistical considerations. Error as-
sessment is different in the two laboratories, but the 2 mean
ages and the weighted means agree acceptably well. The (un-
weighted) mean of the weighted means, 13.71 ± 0.24 Ma, can
be accepted as the time when volcanic activity most likely
ceased in the Börzsöny Mts. Can we distinguish geochrono-
logically between older and younger parts of the High
Börzsöny? The several hundred ka lifetime of a dome complex
is realistic (cf. Davidson & de Silva 2000) and, as we present-
ed, there are both petrographical and volcanological differenc-
es. However, the mean age errors overlap, and the Kolmogor-
ov-Smirnov tests show that the two data populations are
statistically not different, so further radiometric datings are
It is more certain that the K/Ar ages support the view that
volcanic activity did not finish in the Early Badenian. It oc-
curred more or less continuously and lasted up to the Bade-
nian/Sarmatian boundary. We think that this duration of vol-
canic activity does not necessarily contradict the existence of
the above-mentioned Lower and Middle Badenian cover sed-
iments in the S and W Börzsöny. As we presented in point
3.3, the volcanic activity of both the S and High Börzsöny
was characterized by lava dome extrusions. Their effusive or
low-explosivity eruptions at around 1514 Ma, producing no
widespread pyroclastics, may have allowed a coeval, non-
volcanic shallow-marine sedimentation in the Badenian ar-
6.2 Paleomagnetic measurements and their evaluation
The paleomagnetic results we are using in this synthesis come from
different sources. A substantial amount of data were obtained in the six-
ties and finally published by Andó et al. (1977). Another set was mea-
sured in the late seventies and published by Balla & Márton (1978,
1980). In the latter papers, the earlier data were also tabulated and the
interpretation of all the available data led to the following conclusions:
(1) The overall mean paleomagnetic direction (based on the site-
mean directions) indicates that, at the time of volcanic activity, the
Börzsöny Mts. were at the present latitude and in the same orientation
(2) The Börzsöny Mts. are the product of brief volcanic activity that
took place during three magnetic polarity intervals (normalreverse
In recent years, systematic paleomagnetic studies of the ig-
nimbrites of the Bükk Foreland, the area N of the Mátra Mts.
and the Salgótarján Basin have revealed that rotations must
have occurred in the named areas after the emplacement of
ignimbrites (Márton & Márton 1995).
On the basis of volcanological and K/Ar radiometric consid-
erations, the pumiceous volcaniclastic deposits in the
Börzsöny, overlying the Karpatian-Lower Badenian sedimen-
tary formations, have been assumed to belong to the Middle
Rhyolite Tuff or Tar Dacite Tuff (more or less equivalent in
age to the Upper Ignimbrite of the Bükk Foreland: Márton &
Pécskay 1998). This volcanic complex exhibits about 30°
counterclockwise (CCW) rotation. For this reason, and be-
cause K/Ar ages have been found inappropriate to define the
exact beginning of volcanism, we performed additional paleo-
magnetic measurements on the volcaniclastic deposits (Kis-
maros and Kemence tuffs and Nagy Valley sandstone).
Although the number of suitable new outcrops of the volca-
niclastic deposits is limited, the results obtained on them are
significant (Fig. 11). All but one site exhibit moderately CCW
rotated declinations with normal polarity. These sites, along
with previous data also showing CCW rotation and normal po-
larity, are indicated with red circle in the volcanological map
(Fig. 3). One site (Magyarkút) is characterized by easternly
declination and reversed polarity. The former rock group of
sites is interpreted as deposited mostly subaqueously, but that
of Magyarkút subaerially, thus the latter must be younger. This
is supported by K/Ar dating (Table 2).
The new results call for the reconsideration of the earlier
interpretation concerning conclusion (1): the volcaniclastics
were affected by a CCW rotation which necessarily post-
dates their deposition, so the rotation should occur before or
during the emplacement of the subsequent massive andesitic
rocks. Indeed, some of the individual site means of massive
rocks of the previous measurements possess westernly, while
others have easternly declinations; declinations close to the
present N also occur. The earlier treatment of data, that is the
combination of all data (Balla & Márton 1980), averaged out
the differences. This treatment implied that the differences in
individual site mean directions are all due to secular varia-
tions of the Earth magnetic field, since there was no reason
to split the data into different groups.
In both groups that have westernly and easternly declina-
tions, there are sites with normal and with reversed polarity.
Fig. 11. New paleomagnetic site-mean directions with confidence
circles for volcaniclastic rocks of the Börzsöny Mts. For localities,
see Fig. 3. Coeval paleomagnetic direction in a stable European
framework is shown for comparison. Stereographic projection.
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 339
Thus four paleomagnetic groups can be defined (Fig. 12):
1. CCW declination + normal polarity (including new re-
sults), 2. CCW declination + reversed polarity, 3. CW dec-
lination + reversed polarity, 4. CW declination + normal
polarity. Since the garnet-bearing subaqueous volcaniclas-
tics overlying the Karpatian-Lower Badenian sedimentary
deposits are definitely among the oldest volcanics, group 1
exhibiting CCW rotation + normal polarity must be the
oldest. What is more, group 1 also includes massive garnet-
bearing dacitic rocks suggesting that this rock composition
is indicative for age. Groups 2, 3, 4 are proposed to be suc-
cessively younger. Although such a subdivision of the pale-
omagnetic directions may seem arbitrary, volcanology and
K/Ar data broadly agree with the above succession of the
Naturally, there are a few problematic sites. The problems, from the
paleomagnetic side, arise from the fact that some of the magmatic bod-
ies, for example narrow dykes and thin lava flows, may be the point
readings of the magnetic field, thus their directions are influenced more
by the secular variation of the Earth magnetic field than the post-cool-
ing large-scale tectonic movement of the area itself. Another problem
may be that the sampled site is not strictly in situ. Both mechanisms
may be invoked to explain the paleomagnetic outliners in the W High
Börzsöny, where apart from non-rotated samples, two earlier data sug-
gested CCW rotation. Repeated experiment, that is sampling and
measuring four new sites in a limited area near Mt. Magyar (see Fig. 3),
has shown that two of the sites exhibit no rotation, one is characterized
by about 30° westernly declination, the last (this one with very poor sta-
tistical parameters) with easternly rotated declination: that is the aver-
age rotation is zero. Moreover, the last one has reversed polarity, while
the others have normal polarity, indicating that in this part of the
Börzsöny, we have to calculate with a number of small fast-cooling
magmatic bodies (lava flows), each of them being point readings of the
magnetic field rather than individually useful for tectonic interpretation.
Another problematic area is the Kemence Valley. Similarly to other
N Börzsöny locations, the base of Cicõke locality (Fig. 3) shows a
CCW rotation. This is in accordance with the pumiceous matrix of the
proposed Kemence caldera wall and the 15.9 Ma K/Ar age of a caldera
rim lava flow (see Fig. 4e and Table 2). On the other hand, we ob-
tained a 12.5 Ma exceptionally young K/Ar age from an embedded
andesite block at Cicõke and, in addition, some of the cover breccias
and lava flows along the Kemence Valley are pyroxene andesites simi-
lar to the S-E High Börzsöny lavas. If the young age is right, it can
only be explained by assuming an old caldera in the N that subse-
quently became buried by the High Börzsöny distal products; later on,
the caldera rim could be exhumed by postvolcanic tectonism and nor-
mal erosion that have revealed the original structure again.
To summarize, the recently obtained and previously published
paleomagnetic results suggest that the older suites of the
Börzsöny (group 1 and 2) were formed prior to the CCW rota-
Fig. 12. Paleomagnetic groups in the Börzsöny Mts. (displaying all existing data) with CCW rotated declinations + normal polarity (I),
CCW rotated declinations + reversed polarity (II), CW rotated declinations + reversed polarity (III) and CW rotated declinations + nor-
mal polarity (IV). Symbols as in Fig. 3. Stereographic projections.
340 KARÁTSON et al.
tion that affected the Middle Rhyolite Tuff Complex in the north-
ern part of the Pannonian Basin. The rotation must have oc-
curred during a reversed polarity interval (= end of group 2).
Prior to this reversed polarity zone, the CCW rotated sites
of the Börzsöny Mts. are of normal polarity. What is their ex-
act age? The Upper Ignimbrites of the Bükk Mts. (K/Ar ages
17.516.0 Ma), which are the products of more than one vol-
canic pulse (Szakács et al. 1998), are all reversely magne-
tized. Thus, they cannot be strictly of the same age as the ini-
tial volcaniclastics of the Börzsöny Mts. We think that the
latter are younger and so must have formed during the domi-
nantly normal polarity zone ending at 16 Ma (Fig. 13). The
reversely magnetized paleomagnetic groups 2 and 3 may
have erupted between ca. 16 and 14.5 Ma. The age of group 4
(with normal polarity) cannot be placed with certainty in any
of the younger than 15 Ma polarity zones, thus the termination
of the volcanism is more constrained by K/Ar than paleomag-
7. Summary: volcanic evolution and stratigraphy
The volcanic evolution and stratigraphy of the Börzsöny
Mountains can be well established correlating volcanology, pe-
trology and geochemistry with K/Ar and paleomagnetic results
(Fig. 13). Whereas for the beginning of the eruptive activity, pa-
leomagnetism could yield reliable information, the termination
of volcanism could only be dated by the K/Ar method.
Products of the first volcanic stage, garnet-bearing pre-
dominantly dacitic rocks, were deposited in a shallow sub-
marine environment ca. 16.516.0 Ma ago, as part of the
Middle Rhyolite Tuff (Tar Dacite Tuff) of the Pannonian Ba-
sin. This stage was dominated (1) by explosive dacitic erup-
tions originating probably from a small number of silicic,
medium-sized paleovolcanoes located in, and emerging
from, the coeval archipelago, and (2) by shallow submarine
lava-dome extrusions. As for the volcanic-sedimentary pro-
cesses, resedimented syn-eruptive volcaniclastic mass-flow
deposits, in part shallow submarine pyroclastic-flow depos-
its, have been identified. We propose to divide the resulting
volcaniclastics into more proximal and more distal facies.
The near-source origin of the former deposits (directly relat-
ed to explosive eruptions) is suggested by (a) abundant lithic
content, (b) dm-sized clasts showing frequent prismatic
jointing, (c) a lag breccia occurrence and (d) quasi-radial
flow directions (relative to supposed calderas) obtained from
magnetic anisotropy measurements.
Later on, when the submarine basin gradually infilled, sed-
imentation became more complex resulting also in a small
volume of subaerial ignimbrites and different genetic types
of mostly subaerial debris-flow deposits. One type of the lat-
ter may have been related primarily to destructional process-
es resulting in small- to medium-sized calderas.
In the second stage (ca. 16.014.5 Ma based on paleomag-
netism and K/Ar datings), lava dome formation was going
on. During the first half of this period, a major ca. 30° CCW
rotation event occurred. The commonly andesitic rocks
formed at this stage have varied lithology, but do not contain
garnet. At present, due to selective erosion, the more crystal-
lized basal parts of lava domes are often exposed forming in-
verse morphology. The K/Ar age and paleomagnetic data of
the Magyarkút ignimbrite exposure imply that explosive ac-
tivity may have been rejuvenated and taken place on land.
The source of this ignimbrite, due to limited ocurrence, is un-
certain. A major, late event of the second stage was a hydro-
thermal polymetallic ore mineralization in the W that result-
ed in intense alteration of the coeval and older rocks.
The final, third stage was the build-up of the High
Börzsöny andesitic edifice erupted during a normal polarity
zone up to the Badenian/Sarmatian boundary (ca. 13.7 Ma
based on K/Ar data). In fact, the reverse polarity of a single
site at Mt. Magyar indicates that the normal polarity zone was
interrupted by a reverse zone (see Fig. 13). The High
Börzsöny volcano was a subaerial dome complex producing
block-and-ash flows and lava flows probably from a few cra-
ters. At present, however, due to intense erosion, the exposed
near-vent facies of collapsed domes are more common than
real block-and-ash flow deposits. A proposed distinction be-
tween the older W and the younger S-E parts of the edifice is
supported by volcanology, petrography and in part K/Ar geo-
chronology. Mostly in the S and W Börzsöny, nonvolcanic
marine sedimentation (that may have been continuous in plac-
es) was simultaneous with the final stage.
Due to tectonic movements and erosion, the majority of
the original volcanic successions and primary landforms
Fig. 13. Proposed stratigraphy of the volcanic activity of the
Börzsöny Mts. with the magnetic polarity time scale (adopted from
Cande & Kent 1995) displaying suggested positions of the four pa-
leomagnetic groups in Fig. 12 (symbols the same as in Fig. 3), se-
lected, representative K/Ar data and major volcanic events.
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 341
have degraded (Karátson 1995, 1997). Given their resistant
texture, the massive rocks and breccias are those forming the
majority of cover strata at present. The fine matrix of debris-
flow and block-and-ash flow deposits prevents significant
erosion, so they frequently form inverse morphology (Karát-
son 1999). In contrast, the massive rocks, especially the hy-
drothermally altered subvolcanic levels and the platy-jointed
lava rocks, have experienced more intense weathering and
frost shattering. The probable pyroclastic-fall deposits of the
first stage, except for local accumulation pods, have been de-
Acknowledgements: Research work was supported by Hun-
garian Scientific Research Funds (OTKA) F015629, F014122,
F25556, T032774 and FKFP-0175/2000. The geochemical
work belongs to a research project funded by the OTKA
T025833. Additional K/Ar datings and part of the geochemical
measurements were supported by DKs Magyary Postdoctoral
Scholarship. New paleomagnetic measurements on volcani-
clastics were carried out by G. Imre, A. Stankóczi and Á.
Tichy in the Paleomagnetic Laboratory of Eötvös Loránd Geo-
physical Institute under the supervision of E. Márton. Discus-
sions with P. Müller helped to clarify important stratigraphical
and paleoenvironmental problems. We appreciate that L.
Korpás gave us much information prior to the publication of
his map, and we thank K. Németh, G. Redl, B. Székely and B.
van Wyk de Vries for valuable comments on an early draft of
the manuscript. Thorough reviews by J. Lexa, L. Korpás and
D. Vass improved the paper. Additional computer work by J.
Lazányi and field assistance by the Nagymaros Forest Compa-
ny of the Ipoly Forest Ltd. are gratefully acknowledged.
Andó J., Kis K., Márton E. & Márton P. 1977: Palaeomagnetism of
the Börzsöny Mountains, Hungary. Pure Appl. Geophys. 115,
Badics B., Bendõ Zs., Gméling K., Ízing I. & Harangi Sz. 2000: An
ignimbrite deposit from the Holdvilág Creek, Visegrád Mts,
Hungary. Suppl. Acta Mineral. Petrogr. Minerals of the Car-
pathians Int. Conf., XL, Szeged, 8.
Báldi T. & Kókay J. 1970: The fauna of the Kismaros tuffite and the
age of the Börzsöny andesitic volcanism. Bull. Hung. Geol.
Soc. 100, 274283 (in Hungarian with German abstract).
Báldi-Beke M. 1980: The nannoplankton of the Oligocene-Miocene
sediments underlying the Börzsöny Mts. (Northern Hungary)
andesites. Bull. Hung. Geol. Inst. 110, 159179 (in Hungarian
with English abstract).
Báldi-Beke M., Bohn-Havas M., Korecz-Laky I., Nagy-Gellai Á. &
Nagy E. 1980: Recent paleontological and stratigraphical re-
sults on the Oligocene and Miocene of the Börzsöny Moun-
tains and its surroundings. Discuss. Palaeont. 26, 61103.
Balla Z. 1977: Complex interpretation of data on geology and min-
eralization of the Börzsöny Mts. Ann. Rep. Hung. Geophys.
Inst., 2037 (in Hungarian).
Balla Z. 1978: Reconstruction of the High Börzsöny paleovolcano.
Bull. Hung. Geol. Soc. 108, 119136 (in Hungarian).
Balla Z. 1981: Neogene Volcanism of the Carpatho-Pannonian Re-
gion. Earth Evol. Sci. 34, 240248.
Balla Z., Csongrádi J., Havas L. & Korpás L. 1981: Age of the
Börzsöny volcanics and accuracy of K-Ar dating. Bull. Hung.
Geol. Soc. 111, 307324 (in Hungarian with English abstract).
Balla Z. & Korpás L. 1980: Volcano-tectonics and evolution of the
Börzsöny Mts. Ann. Rep. Hung. Geol. Inst. on the year 1978,
75101 (in Hungarian with English abstract).
Balla Z. & Márton E. 1978: The palaeomagnetic sequence in the
Börzsöny volcanic area. Geophys. Hungarica 19/2, 5159, 19/
3, 114120 (in Hungarian).
Balla Z. & Márton E. 1980: Magnetostratigraphy of the Börzsöny
and Dunazug Mts. Rep. Hung. Geophys. Inst. 26, 5777 (in
Hungarian with English abstract).
Balogh K. 1985: K/Ar dating of Neogene volcanic activity in Hun-
gary. Experimental technique, experiences and methods of
chronological studies. ATOMKI Rep. D/1, 277288.
Bence A.E. & Albee A.L. 1968: Empirical correction factors for the
electron microanalysis of silicates and oxides. J. Geol. 76,
Borza T. 1973: Stratigraphical and paleontological investigations in
the vicinity of Hont (northern Börzsöny Mountains). Bull.
Hung. Geol. Soc. 103, 2740.
Cande S.C. & Kent D.V. 1995: Revised calibration of magnetic po-
larity timescale for the Late Cretaceous and Cenozoic. J. Geo-
phys. Res. 100, B4, 60936095.
Cas R.A.F. & Wright J.V. 1991: Subaqueous pyroclastic flows and
ignimbrites: an assessment. Bull.Volcanol. 53, 357380.
Cas R.A.F., Monagham J.J. & Kos A. 1998: Simulating the entry of
pyroclastic flows into the sea. Particulate Gravity Currents
conference, University of Leeds, UK, Abstract Volume, 22.
Császár G. (Ed.) 1997: Basic lithostratigraphic units of Hungary.
Hung. Geol. Inst. 114.
Csillag-Teplánszky E., Csongrádi J., Korpás L., Pentelényi L. &
Vetõ-Ákos É. 1983: Geology and mineralization of the central
area in the Börzsöny Mts. Ann. Rep. Hung. Geol. Inst. on the
year 1981, 77127 (in Hungarian with English abstract).
Csillag-Teplánszky E. & Korpás L. 1982: Explanations to the geo-
logical maps of Börzsöny and Dunazug Mts. I-II. Archives of
the Hung. Geol. Inst. (in Hungarian).
Czakó T. & Nagy B. 1976: Correlation between the data of photo-
tectonic map and prospecting for ore deposits in the Börzsöny
Mts. (N Hungary). Ann. Rep. Hung. Geol. Inst. on the year
1974, 4760 (in Hungarian with English abstract).
Davidson J. & de Silva S. 2000: Composite volcanoes. In: Sigurds-
son H. (Ed.): Encyclopedia of Volcanoes. Academic Press, San
Deplus C., Bonvalot S., Dahrin D., Diament M., Harjono H. &
Dubois J. 1995: Inner structure of the Krakatau volcanic com-
plex (Indonesia) from gravity and bathymetry data. J. Volcanol.
Geotherm. Res. 64, 2352.
Downes H., Pantó Gy., Póka T., Mattey D.P. & Greenwood P.B.
1995: Calc-alkaline volcanics of the Inner Carpathian arc,
Northern Hungary: new geochemical and oxygen isotopic re-
sults. In: Downes H. & Vaselli O. (Eds.): Neogene and related
magmatism in the Carpatho-Pannonian Region. Acta Vulcanol.
Dulai A. 1996: Taxonomic composition and palaeoecological fea-
tures of the Early Badenian (Middle Miocene) bivalve fauna of
Szob (Börzsöny Mts., Hungary). Ann. Hist. Nat. Mus. Nat.
Hung. 88, 3156.
Ferenczi I. 1936: Contributions to the geology of Börzsöny Mts.
Ann. Rep. Hung. Geol. Inst. on the year 192528, 131142 (in
Hungarian with German abstract).
Fodor L., Csontos L., Bada G., Györfi I. & Benkovics L. 1999: Ter-
tiary tectonic evolution of the Pannonian Basin system and
neighbouring orogens: a new synthesis of palaeostress data. In:
Durand B., Jolivet L., Horváth F. & Séranne M. (Eds.): The
Mediterranean Basins: Tertiary extension within the Alpine
Orogen. Geol. Soc. London, Spec. Publ. 156, 295334.
342 KARÁTSON et al.
Gill J.B. 1981: Orogenic andesites and plate tectonics. Springer
Verlag, Berlin-Heidelberg-New York, 1390.
Gyarmati P. 1976: Volcanological history and petrogenesis in the
Börzsöny Mts. Ann. Rep. Inst. Geol. Hung. from the year 1973,
5762 (in Hungarian with English abstract).
Hámor G., Ravasz-Baranyai L., Balogh K. & Árva-Sós E. 1979: K/
Ar dating of Miocene pyroclastic rocks in Hungary. Ann. Géol.
Pays Hellén. 2, 491500.
Harangi Sz. 1999: Geochemistry and petrogenesis of the volcanic
rocks of Csód Hill, Visegrád Mts., Northern Hungary. Topogr.
Miner. Hungariae VI, 5985 (in Hungarian with English ab-
Harangi Sz., Downes H. & Thirlwall M. 1998: Geochemistry and
petrogenesis of Miocene volcanic rocks in the Northern Pan-
nonian Basin and Western Carpathians. Carpatho-Balkan
Geol. Assoc. XVI Congress, Abstract Volume, 203.
Jankovich I. 1974: Contributions to the stratigraphy of the SE mar-
gin of Börzsöny Mts. Ann. Rep. Inst. Geol. Hung. from the year
1972, 3337 (in Hungarian with English abstract).
Karátson D. 1995: Ignimbrite formation, resurgent doming and
dome collapse activity in the Miocene Börzsöny Mountains,
North Hungary. Acta Vulcanol. 7, 2, 107117.
Karátson D. 1996: Rates and factors of stratovolcano degradation in
a continental climate: a complex morphometric analysis for
nineteen Neogene/Quaternary crater remnants in the Car-
pathians. J. Volcanol. Geotherm. Res. 73, 6578.
Karátson D. 1997: Volcanic activity of the Börzsöny Mts., and its
relationship to the caldera problem. Bull. Hung. Geogr. Soc.
CXXI (XLV), 34, 151172 (in Hungarian with English ab-
Karátson D. 1999: Erosion of primary volcanic depressions in the
Inner Carpathian Volcanic Chain. Z. Geomorphol., Suppl.-Bd.
Karátson D., Harangi Sz., Szakmány Gy., Pécskay Z., Márton E.,
Balogh K., Józsa S. & Kovácsvölgyi S. 1999a: 1:50,000 scale
volcanological map of the Börzsöny Mts, North Hungary. 10
EUG Congress, Strasbourg, Abstract Volume, Session D07,
Karátson D., Telbisz T. & Sztanó O. 1999b: Change in particle ori-
entation and shape in volcanic breccias as a function of trans-
port distance: a photo-statistical method. International Union
of Geodesy and Geophysics 22
General Assembly, Birming-
ham, Abstract Volume, B, 168.
Karátson D., Thouret J.-C., Moriya I. & Lomoschitz A. 1999c: Ero-
sion calderas: origins, processes, structural and climatic con-
trol. Bull. Volcanol. 61/3, 174193.
Karátson D. & Németh K. in print: Lithofacies associations of an
emerging volcaniclastic apron in a Miocene volcanic complex:
an example from the Börzsöny Mountains, Hungary. Int. J.
Koneèný V. & Lexa J. 1994: Processes and products of shallow sub-
marine andesite volcanic activity in southern Slovakia. IAVCEI
Congress, Ankara, Abstract Volume, Session 4.
Korpás L. & Lang B. 1993: Timing of volcanism and metallogenesis
in the Börzsöny Mountains, Northern Hungary. Ore Geol. Rev.
Korpás L., Csillag-Teplánszky E., Hámor G., Ódor L., Horváth I.,
Fügedi U. & Harangi Sz. 1998: Explanations for the geological
map of the Börzsöny-Visegrád Mts. Hung. Geol. Inst., 1 216.
Korpás L. & Csillag-Teplánszky 1999: Geological map of the
Börzsöny-Visegrád Mts. and their surroundings. Hung. Geol.
Le Bas M.J., Le Maitre R.W., Streckeisen A. & Zanettin B. 1986: A
chemical classification of volcanic rocks based on the total al-
kali-silica diagram. J. Petrology 27, 745750.
Legros F. & Druitt T. 2000: On the emplacement of ignimbrite in
shallow-marine environments. J. Volcanol. Geotherm. Res. 95,
Lexa J. & Koneèný V. 1974: The Carpathian Volcanic Arc: a discus-
sion. Acta Geol. Acad. Sci. Hung. 18, 279294.
Lexa J., Koneèný V., Koneèný M. & Hojstrièová V. 1993: Distri-
bution of volcanics of the Carpatho-Pannonian region. In:
Rakús M. & Vozár J. (Eds.): Geodynamical model and deep
structure of the Western Carpathians. GÚD, Bratislava, 57
69 (in Slovak).
Lipman P.W. 1997: Subsidence of ash-flow calderas: relation to
caldera size and magma-chamber geometry. Bull. Volcanol. 59,
Marczel F. (Ed.) 1977: Deep borehole data of Hungary from 1974.
Hung. Geol. Inst. 1860 (in Hungarian).
Márton E. & Márton P. 1995: Large scale rotations in North Hunga-
ry during the Neogene as indicated by palaeomagnetic data. In:
Morris A. & Tarling D.H. (Eds.): Palaeomagnetism and Tec-
tonics of the Mediterranean Region. Geol. Soc. London, Spec.
Publ. 105, 153173.
Márton E. & Pécskay Z. 1998: Correlation and dating of the Mi-
ocene ignimbritic volcanics in the Bükk foreland, Hungary:
complex evaluation of paleomagnetic and K/Ar isotope data.
Acta Geol. Hung. 41, 4, 467476.
Mason P.R.D., Downes H., Thirlwall M., Seghedi I., Szakács A.,
Lowry D. & Mattey D. 1996: Crustal assimilation as a major
petrogenetic process in East Carpathian Neogene to Quaterna-
ry continental margin arc magmas. J. Petrology 37, 927959.
McPhie J., Doyle M. & Allen R. 1993: Volcanic textures: a guide to
the interpretation of textures in volcanic rocks. University of
Tasmania, CODES Hobart, 1196.
Moore I. & Kokelaar P. 1998: Tectonically controlled piecemeal
caldera collapse: A case study of Glencoe volcano, Scotland.
Bull. Geol. Soc. Amer. 110, 11, 14481466.
Newhall C.G. & Punongbayan R.S. (Eds.) 1996: Fire and Mud.
Eruptions and lahars of Mount Pinatubo, Philippines. Philip-
pine Institute of Volcanology and Seismology, University of
Washington Press, Quezon City/Seattle, 11126.
Pantó Gy. 1970: Tertiary volcanism of the northern part of the
Börzsöny Mts. In: Kubovics I. & Pantó Gy. (Eds.): Volcano-
logical studies in the Mátra and Börzsöny Mountains.
Akadémiai Kiadó, Budapest, 163302 (in Hungarian with En-
Papp F. 1933a: On the petrological and geological buildup of the vi-
cinity of Márianosztra and Nagyirtás. Bull. Hung. Geol. Soc.
63, 6295 (in German with Hungarian abstract).
Papp F. 1933b: Petrographical and geological observations near Ki-
sirtás and Bányapuszta. Bull. Hung. Geol. Soc. 63, 201215 (in
German with Hungarian abstract).
Pearce J.A. & Parkinson I.J. 1993: Trace element models for mantle
melting: application to volcanic arc petrogenesis. In: Prichard
H.M., Alabaster T., Harris N.B.W. & Neary C.R. (Eds.): Mag-
matic Processes and Plate Tectonics. Geol. Soc. London, Spec.
Publ. 76, 373403.
Pécskay Z. & Nagy B. 1993: New K-Ar data for hydrothermal activ-
ity in the Neogene volcanic region of Nagybörzsöny, NE Hun-
gary. Ann. Rep. Hung. Geol. Surv. on the year 1991, II,
Pierson T.C. & Scott K.M. 1985: Downstream dilution of a lahar:
transition from debris flow to hyperconcentrated streamflow.
Water Resour. Res. 21/10, 15111524.
Rymer H. & Brown G.C. 1986: Gravity fields and the interpretation
of volcanic structures: Geological discrimination and temporal
evolution. J. Volcanol. Geotherm. Res. 27, 229254.
Scandone R. 1990: Chaotic collapse of calderas. J. Volcanol. Geo-
therm. Res. 42, 285302.
Smith G.A. & Lowe D.R. 1991: Lahars: volcano-hydrologic events
VOLCANIC EVOLUTION AND STRATIGRAPHY OF THE BÖRZSÖNY MTS. 343
Petrography of rocks from the Börzsöny Mts.
stage: pumiceous volcaniclastics
Emphasizing the limits of petrography in classifying volcani-
clastic rocks, 25 selected thin sections of pumiceous volcaniclas-
tics are summarized below. Approximate rock compositions are
shown in Fig. 3.
The rock texture is moderately to highly clastic. Mm-sized pum-
ices and pumice fragments have a proportion up to 2030 %. Their
margins are not sharp, their shape is commonly irregular, isomet-
ric, subangular or subrounded. Pore size is constant. In some sam-
ples, the matrix also has a pumiceous character. Lithics, similar to
pumices in size, include porphyric andesite and, rarely, dacite
clasts. Groundmass of the lithics contains usually more than 50 %
glass. Infrequently, holocrystalline rocks also occur. The crystals
include, first of all, small, thin, sometimes oriented plagioclases.
Amphibole (brown or green), sometimes with hypersthene and pla-
gioclase transformation zone, is frequently opacitized. Typically
large crystals of fresh biotite are also abundant, often adjoining
fragments or phenocrysts of garnet. The latter mineral, that is
found in ca. 1/2 of samples but due to large crystal size, the occur-
rence is naturally accidental, can be regarded common. Hyper-
sthene is more frequent than augite that is missing from many
stage: massive rocks
A summary of 20 selected, representative thin sections is given
below. Rock compositions are shown in Fig. 3.
The rock texture is porphyric. 1/3 of the samples have micro-
holocrystalline groundmass (e.g. Nógrád Castle, Pap, Kis-Sas,
Széles, Kopasz hills), the remaining 2/3 (e.g. Nagy-Pogány, Nagy-
Sas, Ruzsás, Galla hills, Rustok saddle) has 580 % glass in the
matrix. Crystal size, 12 mm in average, is predominatly uniform.
Definite crystal orientation is limited (e.g. Széles Hill, Rustok sad-
dle, E of Perõcsény quarry). Mineral assemblages and mineral dis-
tribution in the matrix vary to a great extent. The most abundant
mineral is plagioclase: its single or compound crystals are cycli-
cally zoned. Mostly opacitized, amphibole (hornblende) is also
abundant; unaltered crystals occur mostly in garnet-bearing sam-
ples. Commonly unaltered biotite of significant quantity (1030
%) is missing from only the Börzsönyliget and Széles Hill types.
In turn, augite occurs in only these latter biotite-free types. Hyper-
sthene is always associated with augite, but also occurs alone; it is
missing from Nagy-Pogány and Pap hill types and also from some
samples of the combined Nagy-Sas Hill group. Various endogenic
xenoliths can be found, mostly in garnet-bearing rocks.
and deposition in the debris flow hyperconcentrated flow
continuum. SEPM Spec. Publ. 45, 5969.
Steiger R.H. & Jager E. 1977: Subcommission on Geochronology:
Convention on the use of decay constants in geology and geo-
chronology. Earth Planet. Sci. Lett. 36, 3, 359362.
Szabó Cs., Harangi Sz. & Csontos L. 1992: Review of Neogene and
Quaternary volcanism of the Carpathian-Pannonian region. In:
Ziegler P.A. (Ed.): Geodynamics of rifting. Vol. I. Case studies
on rifts: Europe and Asia. Tectonophysics 208, 243256.
Szakács A., Seghedi I., Zelenka T., Márton E., Pécskay Z. & Póka T.
1998: Miocene acidic explosive volcanism in the Bükk Fore-
land, Hungary: identifying eruptive sequences and searching
for source location. Acta Geol. Hung. 41, 4, 413435.
Vass D. & Marková E. 1966: Comments on delimiting the lower
boundary of the South Slovakian and North Hungarian Torto-
nian formations. Bull. Hung. Geol. Soc. 96, 4, 414420 (in
Hungarian with German abstract).
Walker G.P.L. 1984: Downsag calderas, ring faults, caldera sizes,
and incremental caldera growth. J. Geophys. Res. 89, B10,
White M.J. & McPhie J. 1997: A submarine welded ignimbrite-
crystal-rich sandstone facies association in the Cambrian Tyn-
dall Group, western Tasmania, Australia. J. Volcanol.
Geotherm. Res. 76, 277295.
Representative thin section descriptions:
1st stage. Bajdázó quarry: garnet-bearing bi am
dacite classification based on geochemistry). Groundmass is mi-
croholocrystalline and contains apatite. Crystals of 2 mm in aver-
age size are relatively scattered. They include plagioclase (45 %),
biotite (33 %), amphibole (22 %) and garnet (few crystals). Am-
phibole is opacitized (25 %). An endogenous xenolith with holoc-
rystalline texture containing plagioclase and totally opacitized am-
2nd stage. Kis-Sas Hill: bi am
. Fine-grained groundmass con-
tains ca. 70 % glass, frequent zeolithic vesicles and banded apatite.
Moderately abundant crystals, 1 mm in average size, include plagio-
clase (60 %), opacitized amphibole and biotite (together 40 %).
stage: lava rocks and clasts from block-and-ash flow
deposits of the High Börzsöny dome complex
Summary of 17 selected, representative thin sections is given
below. Rock compositions are shown in Fig. 3.
The rock texture is porphyric, the groundmass contains 40100 %
glass, except for one sample (10 %). Abundant crystals have 12
mm average size. Some samples have two size populations. Rough-
ly 1/2 of the samples shows strong crystal orientation. The most
abundant mineral is plagioclase, then mostly opacitized am-
phibole (hornblende; more than 10 %). In the W High Börzsöny
samples, except for one, amphibole content well exceeds pyroxene
content and may equal plagioclase content. Hypersthene occurs in
all samples but one; augite is missing from 1/2 of the samples. Hy-
persthene is frequent without augite in the W High Börzsöny, and
occurs both alone and with augite in the E and S part. It is common
that hypersthene has augite overgrowth. Biotite is present in only
xenoliths in vent breccias of collapsed domes (W part: Mt. Hollókõ,
along with garnet; SE part: Vilma-pihenõ rock tower). (The xenolith
origin of the Hollókõ biotite, however, is questionable and the rock
type needs a more detailed study.) Garnet has also been found as xe-
nolith in lava flows (Inóc quarry).
Representative thin section descriptions:
Mt. Kövirózsás S (W part, platy jointed lava flow): hyp am
Groundmass is totally glassy and contains some large opaque miner-
als. Crystals, 1.5 mm in average size, are slightly oriented and include
plagioclase (60 %) brown amphibole (25 %), and hypersthene (15 %).
Mt. Csóványos W (SE part, clast from proximal facies of block-
and-ash flow deposit): am py
. Fine-grained groundmass contains
ca. 70 % glass. Smaller crystals are densely, while larger (1 mm in av-
erage size) are rarely spaced. They are plagioclase (55 %), hyper-
sthene (30 %), totally opacitized amphibole (10 %) and augite (5 %).