GeolCarp_Vol56_No1_77

GEOLOGICA CARPATHICA, FEBRUARY 2005, 56, 1, 7790

Petrogenesis of convergent-margin calc-alkaline rocks and

the significance of the low oxygen isotope ratios: the Rodna-

Bârgãu Neogene subvolcanic area (Eastern Carpathians)

DELIA CRISTINA PAPP

, IONEL URECHE

IOAN SEGHEDI

, HILARY DOWNES

and LUIGI DALLAI

Geological Institute of Romania, Cluj-Napoca Branch, P.O. Box 181, 400750 Cluj-Napoca, Romania; deliapapp@pcnet.ro

Institute of Geodynamics, Str. Jean-Luis Calderon 1921, 70201 Bucharest, Romania

Birkbeck/UCL Research School of Geological and Geophysical Sciences, Birkbeck College, Malet St., London WC1E 7HX,

United Kingdom

4**

Istituto di Geologia Ambientale e Geoingegneria IGAG, c/o Dipartimento Scienze della Terra, Università di Roma La Sapienza,

P.le Aldo Moro 5, 00185 Rome, Italy

Present address: CNRIstituto di Geoscienze e Georisorse, Area di Ricerca di Pisa, Via Moruzzi 1, 56127 Pisa, Italy

(Manuscript received March 28, 2003; accepted in revised form March 16, 2004)

Abstract: Neogene calc-alkaline magmatites (from basaltic andesites to rhyolites including mafic cognate enclaves) of

the Rodna-Bârgãu subvolcanic area (East Carpathian arc) are evaluated on the basis of new mineral compositional data,

major and trace elements, as well as Sr and O isotope data. Two different series of rocks have been separated. The

magmas of the medium-K series had a rapid ascent toward the surface, as proven by the presence of primary garnet

bearing rocks, or by the sporadic occurrence of mafic cognate enclaves. The δ

O values of amphiboles vary from 4.2 to

5.4 (SMOW). The δ

O value measured on garnet is 4.3 . The range of

Sr/

Sr ratios is from 0.70588 to 0.70887.

The decrease of the δ

O values as

Sr/

Sr ratios and SiO

increase is interpreted as a progressive contamination of a

mantle derived magma with a contaminant depleted in δ

O and enriched in

Sr/

Sr (i.e. hydrothermally altered lower

crustal rocks). Within the high-K series the presence of intermediate magma chambers where assimilation-fractional

crystallization processes took place is considered. The δ

O values measured on clinopyroxenes vary from 4.6 to 5.7

and on amphiboles from 3.8 to 6.7 . The range of

Sr/

Sr ratios is from 0.70605 to 0.70950. The covariation of the

O values and

Sr/

Sr ratios is scattered. The highest δ

O values correspond to the highest

Sr/

Sr ratios and are

consistent with assimilation of the local upper-crustal rocks. The lower δ

O values and the observed oxygen isotope

disequilibrium between coexisting pyroxenes and amphiboles are explained by interaction with heated meteoric water.

Key words: Eastern Carpathians, subvolcanic intrusions, calc-alkaline magmas, crustal interaction, enclaves, low δ

Introduction

Neogene magmatic rocks of the Inner Carpathian arc have

been the subject of many recent studies (e.g. Salters 1988;

Downes et al. 1995; Mason et al. 1996, 1998; Seghedi et al.

1995, 2001). However, there are as yet few published data on

the subvolcanic zone in the East Carpathian arc (e.g. Ureche et

al. 1995; Papp 1999; Niþoi et al. 2002). Here we report the first

detailed geochemical and isotopic study on Neogene magmat-

ic rocks from the Rodna-Bârgãu Mountains (Romania), which

form the main area of the subvolcanic zone. We evaluate the

calc-alkaline magmatism on the basis of new mineral compo-

sitional data, major and trace elements, as well as Sr and O

isotope data. The role of the mantle source, crustal assimila-

tion and magma mixing is addressed. Parental magmas show-

ing a depleted oxygen isotope signature relative to average

MORB are a special feature of the magmatic rocks from the

Rodna-Bârgãu Mountains. They are interpreted in terms of as-

similation of hydrothermally altered crustal rocks and interac-

tion between intrusive bodies and heated meteoric water.

Up to now, similar low δ

O rocks have not been reported

within the Carpathian Neogene volcanic arc. More extended

occurrences of rocks (e.g. alkaline basalts, granulite enclaves)

depleted in oxygen isotope have been described in the Pan-

nonian Basin (e.g. Kempton et al. 1997; Dobosi et al. 2003),

but those are related to the distensional efforts behind the Car-

pathian arc.

Worldwide, significantly

O-depleted primary magmas

have been recovered from plume derived basalts erupted in

ocean island settings (e.g. Iceland, Hawaii, Canary Islands)

(Eiler 2001 and literature therein). The origin of this low δ

signature has been attributed to either interaction of basaltic

magmas within the hydrothermally-altered lower oceanic crust

or as a primary feature of the plume itself. For the more

evolved magmas from Canary Islands, Thirlwall et al. (1997)

favour an interpretation that their

O-depleted character

might have been acquired during contamination by hydrother-

mally altered rocks in the current lithosphere.

Our study brings new arguments in the debate of the origin

of the

O-depleted magmas enforcing the importance of con-

www.geologicacarpathica.sk

*Corresponding author: E-mail: deliapapp@pcnet.ro; Tel./Fax:

+40264-429430

78 PAPP, URECHE, SEGHEDI, DOWNES and DALLAI

tamination by hydrothermally altered crustal rocks in conti-

nental arc-related magmatism. It also provides new data,

which might contribute to improving the models of geo-tec-

tonic evolution of the Inter-Carpathian area.

Geological setting of the Rodna-Bârgãu Mountains

Magmatic rocks in the Inner Carpathian arc are considered

by most authors (e.g. Csontos 1995; Balintoni 1996) to belong

to an active continental margin, connected to a subduction

zone located at the southwestern border of the Eurasian plate.

Although subduction was related to Alpine tectonic activity,

which started during the Cretaceous, the major convergent

event occurred in Miocene times (Sãndulescu 1984). Neogene

magmatism was associated with the consumption of a small

piece of ocean crust, attached to the European plate, beneath

the ALCAPA (Alpine-Carpathian-Pannonian) and Tisia-Getia

continental blocks (Rãdulescu & Sãndulescu 1976; Seghedi et

al. 1998).

The subvolcanic zone is located on the Tisia-Getia block

close to the boundary with both the ALCAPA block and the

Eastern European Plate, between the two volcanic segments:

Oaº-Gutâi in the north-west and Cãlimani-GurghiuHarghita

in the south-east (Fig. 1, insertion). Within the subvolcanic

zone numerous intrusive bodies are located in three areas:

Þibleº, Toroioaga and Rodna-Bârgãu.

In the Rodna-Bârgãu area special geological-structural con-

ditions resulted from the tectonic contact of the Rodna meta-

morphic massif with the Transcarpathian Flysch Zone, delin-

eated by the Someº Fault system (Fig. 1). The host-rocks of

the magmatites are crystalline schists of the Middle Dacides

and sedimentary deposits of the Transcarpathian flysch.

The intrusive bodies from the Rodna Mountains, hosted by

metamorphic rocks, are mostly large massive bodies (stocks,

and laccoliths associated with sills and dykes). More acidic fa-

cies are represented by: Parva rhyolites, Cormaia rhyodacites

and Valea Vinului quartz andesites (porphyritic micrograno-

diorites). In contrast, sedimentary flysch deposits form the

host-rocks of the intrusive bodies in the Bârgãu Mountains.

The main intrusive units are: Bucnitori and Sturzii (dacites),

Pleºii-Mal (quartz garnet andesites), Cornii (andesites),

Chicera and Arsente (Mãguri) (microdiorites) and Heniu,

Oala, Iliuþa, Colibiþa (South Bârgãu) (andesites). The intru-

sions vary in volume and have a surface exposure from 1 km

(Pleºii-Mal) to 20 km

(Cornii, Heniu).

Within the Rodna-Bârgãu area the age of the magmatites is

119 Ma, corresponding to Pannonian, similar to other volca-

nic activity in the Eastern Carpathians (Pécskay et al. 1995).

The distribution of ages among the delimitated intrusive units

is as follows: Sturzii 10.6±0.7 Ma, Runc 10.4±0.8 Ma, Valea

Vinului 9.0±0.5 Ma, Cornii 9.8±0.8 Ma, Mãgura Arsente

8.8±0.5 Ma, and Mãgura Rodnei 8.6±0.4 Ma.

Petrography of the magmatites

The main petrographic types of the Neogene magmatites are

basaltic andesites, microdiorites, quartz biotite amphibole

andesites, quartz-garnet andesites, dacites, rhyodacites, and

rhyolites. Transitional textures occur between subvolcanic and

plutonic facies, and between hypabyssal and volcanic-like fa-

cies. There is a relatively high degree of crystallization and

most of the rocks are porphyritic.

The main minerals are: plagioclase feldspars, amphiboles,

pyroxenes, biotite, quartz and subordinate potassic feldspars.

The accessory minerals are: apatite, magnetite, zircon, garnet

and the secondary minerals are: clay minerals, sericite, chlo-

rite, calcite, and epidote.

Magmatic cognate enclaves are relatively frequent in andes-

ites, microdiorites and diorites (Niþoi et al. 1995). They are

subangular to rounded in shape and vary from 23 cm up to

2025 cm in size. The cognate enclaves have a holocrystalline

hipidiomorphic structure, frequently showing a poikilitic char-

acter due to the presence of large amphibole crystals (mega-

crystals), which may include other minerals such as pyroxenes

and feldspars. The texture is massive. Amphiboles ± feldspars

represent the mineralogical composition of most cognate en-

claves, forming a hornblendite-like composition. Pyroxene ±

amphibole ± feldspars (pyroxenite-like) are also present.

Within the Valea Vinului quartz andesites, cog-

nate enclaves containing biotite + amphibole + feldspar oc-

cur. The contact between cognate enclaves and host-rocks is

either sharp or can show evidence of partial dissemination in

the host melt. Detailed petrographic descriptions of cognate

enclaves and their relationship with the host-rocks can be

found in Niþoi et al. (2002).

The magmas that generated the host-rocks in the Rodna-

Bârgãu Mountains had a calc-alkaline character and show a

complete differentiation trend, while the cognate enclaves are

much richer in FeO and MgO and display tholeiitic features

(Ureche 2000).

Sampling and analytical techniques

We have studied most of the important intrusive rocks in the

Rodna-Bârgãu sector. Samples of basaltic andesites, mi-

crodiorites, quartz biotite amphibole andesites, quartz garnet

andesites, dacites, rhyodacites and rhyolites, as well as mafic

cognate enclaves, were selected for whole rock and mineral

separates analysis.

For most of the whole-rock samples, major elements were

determined by wet chemical methods at the Prospecþiuni S.A.

(Bucharest) Laboratory. Trace elements and REE have been

determined by neutron activation method at the Geological In-

stitute of Romania. In addition, 23 pressed powder pellets

from the same whole-rock samples were made for determining

Rb, Sr, Y, Zr, Nb content by XRF analyses. For four samples

(I4/A2, I4/A3, I4px0, I4px4) complete chemical analysis (ma-

jor and trace elements) was performed by XRF. XRF analyses

have been carried out at the Department of Earth Science of

the La Sapienza University of Rome, using a PHILIPS 1480

Spectrometer equipped with a Rh tube running at 30 Kv

60 Ma for major elements and 50 Kv50 Ma for trace ele-

ments. Ba, La, Ce, Cr, and V have been performed by a W

tube running at 50 Kv50 Ma conditions. International stan-

dards were used for calibration and the precision and accuracy

80 PAPP, URECHE, SEGHEDI, DOWNES and DALLAI

were analysed in bulk, and the rocks were decomposed with a

mixture of ultrapure HF and HNO

in a teflon vessel at 70 °C

for 48 h. The resulting solution was evaporated and taken up

in 6.2 N ultrapure HCl. The Sr was separated in a 3 ml AG

50W-X8 resin column. Isotopic analyses were carried out us-

ing both Finnigan Mat 262RPQ multicollector and VG 54E

single collector mass spectrometers. For the latter machine,

the procedures of Ludwig (1994) were applied for data acqui-

sition and reduction. Internal precision (within-run precision)

of a single analytical result is given as two standard error of

the mean. Repeated analyses of standard NBS987 gave aver-

ages and errors expressed as 2 standard deviation as follows

Sr/

Sr=0.71024 +/2 (n=20) and the isotope ratios were

normalized to

Sr/

Sr to 0.1194.

Mineral separates of amphibole, pyroxene and garnet have

been analysed for O isotope ratios. Mineral separates were

made by hand picking under a microscope. In some cases sup-

plementary purification of the separates was performed, by

washing in an ultrasonic bath and/or acetone. Oxygen isotope

analyses were carried out at the CNR Centro di Studio per

il Quaternario e lEvolutione Ambientale, Rome, using a laser

fluorination system attached to a Finnigan MAT DELTA

plus mass spectrometer. Some samples have been replicated

or newly performed at the Royal Holloway University of Lon-

don using a laser fluorination system following the method de-

Table 1: Representative microprobe analyses of amphiboles of host rocks and enclaves. Abbreviations: Hen Heniu; V.Vin V.Vinului;

Chi Chicera; Ars Arsente; ro rock; en enclave; r rim; c core; pt potassian; tsch tschermakite; mghas mag-

nesiohastingsite; parg pargasite.

scribed by Mattey & Macpherson (1993). The determinations

made in Italy and the UK are comparable. Two or more ex-

tractions were made on each sample; the average reproducibil-

ity of isotopic analyses is ±0.2 or better. The analytical data

are reported in the δ-notation referenced to SMOW as the

mean of two or more replicate analyses.

Results

Main mineral species

Amphiboles are the main mafic minerals within the magmat-

ic rocks in the Rodna-Bârgãu area. All are Ca- rich and have

been classified according to IMA (Leake et al. 1997) nomen-

clature (Table 1). The magnesium number [Mg#=MgO/

(MgO+FeO)] varies from 0.53 (Valea Vinului cognate en-

clave) up to 0.87 (amphibole megacrystalArsente cognate en-

clave). Most amphibole phenocrysts, especially those found in

andesites and microdiorites are magnesiohastingsites. Am-

phiboles from more acidic facies are tschermakites. Amphib-

oles in cognate enclaves are also relatively heterogeneous, dis-

playing magnesiohastingsite (ChiceraArsente), pargasite

(Cornii) and tschermakite (Heniu) compositions. Zoning is

present within most of the amphibole crystals, both in host-

Sample

Anls.

Loc.

I15

ro(r)

Sturzii

I15

ro(c)

Sturzii

I31

Pleºii

Hen

I18

Iliuþa

I11

V.Vin

I11

V.Vin

Chi

Ars

en(r)

Ars

en(c)

Ars

I4mega

Ars

I14

Cornii

I14

Cornii

SiO

44.53

41.98 41.90 43.06 42.72 43.37

40.95

40.72

42.44

42.70

42.08

41.52

42.64

42.73

42.95 43.27

TiO

0.91

1.00

0.69

1.05

1.86

2.18

1.47

1.83

1.30

1.21

1.67

1.87

1.46

1.19

1.95

2.15

11.48

13.98 13.40 11.74 12.18 12.49

12.02

12.69

13.18

13.04

13.13

12.98

13.27

13.12

12.25 11.70

0.00

0.05

0.07

0.03

0.06

0.00

0.08

0.04

0.06

0.09

0.10

0.08

0.00

0.10

FeO*

15.96

16.82 17.99 17.51 13.83 14.32

18.69

16.25

9.54

8.86

11.52

13.51

9.52

10.56

12.11 11.50

MnO

0.62

0.51

0.60

0.23

0.25

0.56

0.35

0.10

0.16

0.23

0.32

0.15

0.28

0.20

MgO

12.18

10.50 10.33 11.37 12.61 12.75

9.25

10.69

15.28

15.79

14.13

12.68

15.58

15.16

12.77 13.85

CaO

9.78

10.37

9.88

9.56 11.12 10.19

11.35

11.21

11.68

11.73

11.74

11.56

11.92

11.51

11.37 11.92

1.40

1.62

1.86

1.70

1.86

2.24

1.46

1.87

2.26

2.22

2.37

2.24

2.49

2.29

2.20

1.96

0.23

0.32

0.38

0.24

0.48

0.38

1.38

0.84

0.75

0.71

0.80

0.69

0.92

0.74

BaO

0.00

0.01

0.00

0.01

0.00

0.02

0.03

0.01

0.00

0.02

0.01

0.02

0.08

0.13

0.14

0.12

0.00

0.11

0.14

0.21

0.24

0.14

0.08

0.06

0.21

Total

97.17

97.27 97.15 96.98 97.08 97.17

97.34

97.19

96.91

96.85

97.84

97.57

98.01

97.47

97.01 97.62

6.51

6.18

6.21

6.35

6.29

6.21

6.15

6.18

6.20

6.13

6.12

6.14

6.18

6.35

6.34

1.49

1.82

1.79

1.65

1.71

1.79

1.85

1.82

1.80

1.87

1.88

1.86

1.82

1.65

1.66

T sites

8.00

0.50

0.61

0.55

0.39

0.40

0.37

0.36

0.40

0.44

0.43

0.38

0.37

0.39

0.41

0.49

0.36

0.10

0.11

0.08

0.12

0.21

0.24

0.17

0.21

0.14

0.13

0.18

0.21

0.16

0.13

0.22

0.24

0.00

0.01

0.00

0.01

0.00

0.01

0.00

0.01

0.75

0.84

0.94

0.96

0.53

0.65

0.55

0.44

0.48

0.52

0.48

0.50

0.46

0.62

0.20

0.26

2.65

2.31

2.28

2.50

2.77

2.79

2.09

2.40

3.32

3.42

3.07

2.79

3.34

3.27

2.82

3.02

1.00

1.12

1.15

1.02

1.09

0.95

1.82

1.54

0.61

0.49

0.88

1.12

0.64

0.57

1.28

1.11

C sites

5.00

0.20

0.11

0.14

0.17

0.09

0.16

0.01

0.05

0.07

0.06

0.05

0.04

0.05

0.08

0.02

0.04

0.08

0.06

0.07

0.03

0.07

0.04

0.01

0.02

0.03

0.04

0.02

0.03

0.02

1.52

1.64

1.57

1.51

1.75

1.60

1.84

1.79

1.82

1.83

1.84

1.78

1.80

1.87

0.20

0.19

0.23

0.24

0.13

0.21

0.08

0.12

0.09

0.08

0.12

0.14

0.07

B sites

2.00

0.09

2.00

0.20

0.27

0.31

0.25

0.40

0.43

0.35

0.44

0.54

0.53

0.58

0.55

0.62

0.53

0.49

0.04

0.06

0.07

0.05

0.09

0.07

0.27

0.16

0.14

0.13

0.15

0.13

0.17

0.14

A sites

0.24

0.33

0.38

0.30

0.49

0.50

0.62

0.70

0.67

0.72

0.68

0.77

0.65

0.66

0.63

1.96

2.63

2.78

1.94

2.00

2.69

2.60

1.90

1.98

1.94

1.96

1.97

1.90

Mg#

0.69

0.65

0.64

0.68

0.70

0.72

0.53

0.60

0.83

0.86

0.77

0.70

0.83

0.87

0.68

0.73

Modif.

ferro

pts

Name

tsch

mghas

parg

PETROGENESIS OF CALC-ALKALINE ROCKS: NEOGENE SUBVOLCANIC AREA (EASTERN CARPATHIANS) 81

rocks and cognate enclaves. Some amphiboles (e.g. Arsente

microdiorite) display Mg-richer core and Fe-richer rim, while

other amphiboles (e.g. Sturzii dacite) display reverse zonation

with Mg-richer rim and Fe-richer core.

Pyroxenes are present only in the more basic petrographic

types, where they are subordinated in abundance to the am-

phiboles. They are also found in some mafic magmatic cog-

nate enclaves. In some cognate enclaves pyroxene is the only

mineralogical component. Only clinopyroxene occurs in the

studied rocks (Table 2). Within the cognate enclaves mostly

diopside is present, while augite occurs mainly in the host-

rocks. Occasionally, clinoenstatite was found in the pyroxen-

ite enclaves within the Arsente microdiorite.

Biotite is present in the more acidic petrographical types ei-

ther as the only mafic mineral (Parva rhyolite) or in associa-

tion with amphiboles (Ilva rhyodacites and dacites, as well as

Valea Vinului quartz biotite andesites). It is also present in the

mafic magmatic cognate enclaves within the Valea Vinului In-

trusive Unit, where it has a slightly different chemical compo-

sition compared with the host-rock biotite. This corresponds

to higher Al and Fe, and to lower Mg content.

Plagioclase feldspars are the main components of all the

petrographic types; they form both phenocrysts and microlites

in the matrix. Anorthite content varies according to the petro-

Sample

Analysis

Location

Rock

Chicera

rock

Arsente

I4/A2

Arsente

I4px0

Arsente

I4px0

Arsente

I4px1

Arsente

I4px2

Arsente

I4px4

Arsente

I14

Cornii

en(cor)

Runc

en(rim)

Runc

SiO

52.13

52.07

51.12

53.65

55.14

52.54

52.83

52.25

50.67

51.84

51.59

1.56

1.28

3.18

1.77

1.83

1.85

1.52

2.25

3.84

2.08

2.14

TiO

0.63

0.28

0.56

0.24

0.13

0.20

0.11

0.72

0.62

0.58

0.10

0.07

0.05

0.52

0.34

0.14

0.10

0.00

FeO*

8.17

9.29

6.30

3.92

11.15

5.27

5.46

6.40

6.70

9.44

9.41

MnO

0.48

0.46

0.24

0.20

0.25

0.13

0.24

0.27

0.24

0.37

0.39

MgO

14.69

14.14

15.62

17.03

30.08

15.51

15.21

15.59

13.70

15.22

14.94

CaO

20.86

21.59

21.87

22.50

0.67

24.18

24.00

22.39

23.69

19.97

20.33

0.04

0.03

0.01

0.00

0.02

0.03

0.02

0.50

0.25

0.31

0.21

0.02

0.13

0.30

0.41

0.29

0.27

BaO

0.00

0.03

0.00

0.02

0.00

0.02

0.00

0.11

0.24

0.00

0.03

0.00

0.14

0.20

0.31

0.00

0.23

Total

99.26

99.72

99.27

100.00

99.60

100.09

99.90

100.06

99.86

99.91

43.41

44.17

44.90

45.54

1.30

48.39

48.37

45.40

49.19

40.91

41.68

42.53

40.26

44.62

47.95

81.40

43.18

42.66

44.02

39.58

43.39

42.62

14.06

15.57

10.48

6.51

17.30

8.43

8.97

10.58

11.23

15.70

graphic type: between 2025 % (oligoclase) in acid rocks, and

between 5576 % (labradorbytownite) in most microdiorites

and basaltic andesites (Table 3). The plagioclases in Chicera

and Heniu Magmatic Units (microdiorites, amphibole andes-

ites) are even more basic, with an anorthite content of 70

80 %. Plagioclase feldspars frequently show normal and oscil-

latory zoning, which is best shown in samples of the Arsente,

Valea Vinului, Cornii and Sturzii Units, indicating modifica-

tion of crystallization conditions (i.e. magma chamber refilling

and/or rapid cooling during the emplacement of the intrusive

body).

Potassic feldspars are present in very small amounts (0

4 wt. %), mostly in the cognate enclaves or in the form of mi-

crolites in the matrix of the more acidic petrographic types

(Valea Vinului). They form totally subordinated phenocrysts

in porphyritic microgranodiorites. They are mostly orthoclase.

In many cases they are replaced by sericite, kaolinite and cal-

cite.

Garnets are present only in the Pleºii-Mal quartz andesites

and Sturzii dacite, as 12 wt. % of the rock volume. They

form phenocrysts with subhedral or euhedral morphologies,

0.52.5 mm in size, and have almandine-rich compositions

(over 55 %) (Table 4). Garnets are fresh, without inclusions

and reaction zones. Most are found as inclusions in plagio-

Sample

Analysis

Location

I15

rock

core

Sturzii

I15

rock

middle

Sturzii

I15

rock

rim

Sturzii

I31

rock

Pleºii

rock

Heniu

I8/hb1

Heniu

I8/hb2

Heniu

I4/A2

core

Arsente

I4/A2

middle

Arsente

I4/A2

rim

Arsente

I4mega

Arsente

rock

Runc

SiO

55.42

57.43

56.25

57.15

50.97

52.04

64.54

55.40

54.46

56.33

65.03

51.66

27.78

26.38

27.18

26.40

30.56

29.57

18.18

27.23

28.00

26.62

18.31

29.66

CaO

10.29

8.59

9.59

8.71

13.67

12.81

0.17

9.81

10.77

9.31

0.11

13.08

MnO

0.11

0.00

0.11

0.04

0.03

0.00

0.04

FeO*

0.14

0.07

0.09

0.25

0.24

0.16

0.28

0.29

0.06

0.63

BaO

0.00

0.05

0.04

0.01

0.00

0.03

0.11

0.00

0.06

0.03

0.10

0.00

5.70

6.65

6.07

6.43

3.75

4.26

0.66

5.93

5.14

6.13

0.39

4.00

0.16

0.15

0.14

0.17

0.07

0.09

15.38

0.38

0.32

0.33

15.99

0.20

Total

99.60

99.33

99.07

99.34

99.05

99.19

99.03

99.04

100.00

99.30

49.59

57.83

52.97

56.60

33.00

37.40

6.00

51.12

45.49

53.32

3.60

35.20

49.49

41.29

46.25

42.40

66.50

62.10

0.80

46.71

52.67

44.77

0.00

63.70

0.92

0.88

0.78

1.00

0.50

93.10

2.17

1.84

1.91

96.40

1.10

Table 3: Representative microprobe analyses of feldspars of host rocks and enclaves. Abbreviation: en enclave.

Table 2: Representative microprobe analyses of pyroxenes of host rocks and enclaves. Abbreviation: en enclave.

82 PAPP, URECHE, SEGHEDI, DOWNES and DALLAI

Table 4: Representative microprobe analyses of garnets. Abbreviations: plz plagioclase; hbl hornblende.

Sample

I9 Mal

Pleºii

Analysis

in plz

in plz (from rim to core)

in hbl

rim

middle

core

rim

middle

core

SiO

37.69

37.90

37.52

37.79

37.70

37.67

37.52

37.55

37.42

37.63

37.71

TiO

0.19

0.16

0.06

0.23

0.09

0.13

0.21

0.25

0.28

0.29

0.21

0.18

20.78

20.82

20.87

20.60

20.85

20.58

20.59

20.55

20.81

20.29 20.5

20.75

0.08

0.00

0.08

0.00

MgO

5.81

5.57

5.68

5.60

5.45

5.25

5.05 4.6

5.35

5.42

5.45

CaO

5.22

5.37

4.51

4.74

5.35

4.66

5.65

4.35

5.11

4.95

4.58

5.13

MnO

3.85

3.45

4.15

4.03

3.59

4.52

3.75

5.22

4.11

3.55

3.99

3.65

FeO

27.35

27.51

27.72

28.14

27.87

28.05

28.03

27.69

27.73

28.36

28.56

28.07

Total

100.97

100.78

100.89

100.86

100.99

100.89

100.95

100.26

100.94

100.20

100.89

100.94

End

members

0.24

0.00

0.24

0.00

Adr

0.53

0.45

0.17

0.63

0.25

0.37

0.59

0.72

0.79

0.80

0.59

0.51

Grs

13.29

13.95

11.91

11.95

14.00

12.08

14.47

11.17

12.86

12.46

11.60

13.17

Alm

56.49

57.53

57.96

58.29

57.97

58.49

58.31

59.09

57.80

59.29

59.34

58.42

Sps

8.06

7.31

8.79

8.46

7.56

9.55

7.90

11.28

8.68

7.53

8.40

7.69

Prp

21.39

20.77

21.17

20.68

20.22

19.51

18.73

17.50

19.87

19.92

20.07

20.21

Sum

100.00

clase and rarely in hornblende phenocrysts. Garnets occur ei-

ther in the core of their host crystals or they can have a periph-

eral position. The optical and chemical features of the plagio-

clases hosting garnet crystals are similar to other plagioclases

in the rocks.

The presence of the assemblage amphibole-biotitequartz

plagioclaseK-feldsparmagnetite in lithologies like Sturzii

dacite and Valea Vinului andesites recommend that the Al-

content in hornblende geobarometer of Johnson & Rutherford

(1989) may be applied for pressure estimations. Assuming

temperatures close to the range over which this barometer has

been calibrated (740780 °C) pressures of 550800 MPa have

resulted. The same barometer has been tentatively applied for

others intrusive units (Pleºii, Cornii, Chicera, Arsente, Heniu,

Iliuþa) resulting in pressures within the same range, both for

host-rocks and cognate enclaves. Slightly higher pressure val-

ues have been obtained in the case of cognate enclaves com-

pared with their host-rocks. The highest pressure has been

found for the Sturzii dacite. Significant differences are record-

ed between the pressure values corresponding to the core of

the crystals from Sturzii dacite (~780 MPa) and the pressure

value corresponding to the rim (~490 MPa). This finding im-

plies that decompression occurred during crystallization of

amphiboles. The pressure estimates for host-rocks and cognate

enclaves suggest mid-crustal depths of approximately 15

25 km for amphibole crystallization.

Major and trace element geochemistry

Major and trace elements compositions of the host-rocks

and cognate enclaves are presented in Table 5. SiO

ranges be-

tween 39.81 wt. % and 49.87 wt. % in the mafic magmatic

cognate enclaves, and from 52.42 wt. % (Colibiþa andesite) to

75.39 wt. % (Parva rhyolite) in the host-rocks. This variation

corresponds to basalts, basaltic andesites, andesite, dacites and

rhyolite. Two series of rocks can be seen in the K

O-SiO

dia-

Fig. 2. Si O

vs. Na

O+K

O diagrams for Rodna-Bârgãu igneous

rocks showing the separation of two series of rocks.

gram (LeMaitre 1989) (Fig. 2). The first series comprises

andesites and microdiorites from South Bârgãu (Heniu, Oala),

the quartz garnet andesites (Pleºii-Mal), Sturzii-Bucnitori dac-

ites and Parva rhyolites and lies in the medium- to low-K do-

main. The second series contains some of the basaltic andes-

ites, Mãguri and Cornii andesites and microdiorites, and Valea

Vinului biotite quartz andesites. It trends towards the high-K

field. It is worth noting that within the medium-K series am-

phiboles are the only mafic minerals, while within the high-K

series clinopyroxenes, amphiboles and biotites all occur. The

cognate enclaves fall mostly in the domain of basalts, with

SiO

less than 50 wt. %, and K

O less than 1 wt. %, except

those from the Valea Vinului and Arsente which are enriched

in K

O due to high biotite content.

84 PAPP, URECHE, SEGHEDI, DOWNES and DALLAI

Table 5: Major and trace element data for Rodna-Bârgãu igneous rocks. Major elements are given in percent (%) and trace elements in ppm.

Abbreviations: ba basaltic andesite; pa pyroxene andesite; aa amphibole andesite; qba quartz biotite andesite; qga quartz

garnet andesite; md microdiorites; d dacite; r rhyolite; nd not determined. Total iron is expressed as FeO.

Sample

Location

Rock type

Parva

I15

Sturzii

P21

Sturzii

P22

Bucnitori

P23

Bucnitori

I10

V.Vin.

qba

I11

V.Vin.

qba

P13

V.Vin.

qba

P14

V.Vin.

Plesii

qga

Mal

qga

I14

Cornii

P82

Cornii

P108

Cornii

Arsente

P62

Arsente

Chicera

P71

Chicera

SiO

75.39 74.17 74.52 65.30 67.22 66.53 67.27 60.33 61.42 63.95 63.22 64.92 58.60 58.23 59.36 58.17 54.50 55.93 53.12 54.23

TiO

0.17 0.05 0.12 0.34 0.33 0.20 0.52 0.49 0.58 0.54 0.64 0.45 0.66 0.64 0.51 0.61 1.04 0.92 0.47 0.84

14.53 15.18 15.26 17.30 16.16 16.14 16.98 15.01 16.69 16.07 16.14 16.98 18.97 18.4 17.74 16.53 17.77 18.14 19.96 17.69

FeO*

1.41 1.25 1.65 3.87 3.40 3.13 3.33 6.83 5.70 6.45 5.49 4.15 5.07 8.43 6.31 7.23 8.24 6.23 5.43 7.54

MnO

0.07 0.06 0.04 0.11 0.10 0.09 0.09 0.24 0.13 0.11 0.13 0.09 0.16 0.15 0.14 0.17 0.24 0.14 0.05 0.14

MgO

0.20 0.30 0.20 1.44 1.38 1.26 0.86 3.68 2.73 2.43 2.45 2.27 1.98 2.62 2.7 3.52 4.28 4.14 4.62 4.78

CaO

1.49 1.41 1.98 5.17 5.06 5.58 3.87 5.88 4.68 0.07 5.41 4.71 6.83 6.01 6.47 6.41 7.18 7.25 9.45 8.84

2.10 2.25 3.23 3.59 3.89 4.11 4.05 2.8 3.13 3.83 3.31 4.05 3.4 3.14 4.15 3.64 3.32 3.58 3.05 3.05

3.50 3.47 2.32 0.96 1.39 1.86 1.81 2.88 3.24 3.24 2.95 1.15 0.93 1.81 2.35 2.66 2.07 1.89 2.75 1.56

0.05 0.04 0.16 0.26 0.09 0.08 0.10 0.23 0.3 0.60 0.27 0.09 0.02 0.16 0.20 0.21 0.23 0.18 0.06 0.18

2.36 2.17 1.28 0.95 1.14 1.41 1.30 0.89 1.11 1.97 0.66 1.68 2.9 0.18 0.90 1.00 1.63 1.77 1.12 1.41

Total

100.27 100.35 100.76 99.29 100.16 100.39 100.18 99.26 99.71 99.26 100.67 100.54 99.52 99.77 100.83 100.15 100.50 100.17 100.08 100.28

189

nd 128

39 61 71

120

111

118 115

66 81 71

173 207 225

287 386 325

351

420

393

288 345

368

287

435

482 358

386

235

333

207

14 11

nd 15

nd 10

22 23

24 30 20

nd 17

39 44 67

111 137 101

103

168

156

150 188

123

110

145

162 138

109

nd 96

nd 11

nd 10

nd 4

nd 5

21 42 13

3 4

24 14

11 15 13

nd 11

15 15

nd 12

19 21

16 16 16

nd 52 47

45 42

nd 45

120

125 73

71 114

nd 59

4 5 7

6 7

15 10

16 10 12

1 5 3

4 5 5

11 9

7 14

325 601 378

220 376 403

516

1023

1010

746 993

307

300

844

772 821

426

327

246

5 6 10

11 10 10

84 84

36 32 21

253

236

14 23 18

45 66

nd 36

122

112

108 136

153

122 132

179

107

150

226

nd 8 10 6

12 15

23 12 28

65 13 20

22 10

39 31

34 38 35

nd 2.03

nd 2.06 1.14 2.00 2.06 2.09 nd 2.50 2.00 2.02 nd

23.77 24.89 nd 25.00 30.49 17.25 33.03 36.00 31.00 43.27 62.70 17.91 12.00 39.00 53.65 32.56 14.60 nd 3.80

26.70 nd

nd 31.00 nd 31.45

nd 23.03 18.00 29.00 29.36 nd 13.30 nd 4.80

3.35 nd

nd 4.33 nd 4.36

nd 2.76 4.40 4.20 4.25 nd 3.8

nd 1.30

0.85 nd

nd 0.43 nd 0.44

nd 0.82 0.75 0.37 0.37 nd 0.84 nd 0.51

0.45 0.93 nd 0.46 nd 0.16

nd 0.55 0.52 0.69 0.37 nd 0.38 nd 0.43

0.61 nd

nd 1.20 1.63 0.88 1.45 2.40

nd 2.47 2.61 1.35 1.58 2.38 1.55 2.54 0.67

nd 0.79

nd 0.27 nd 0.27

nd 0.17 0.24 0.12 0.12 nd

9.42 nd

nd 0.51

nd 0.31 nd 2.80 nd

nd 1.20 nd

10.16 nd

nd 4.40 nd 4.46

nd 11.00 11.10 11.64 11.49 1.64 1.00 4.00 4.05 nd 4.30 nd

1.60

1.99 nd

nd 2.60 nd

nd 1.54 1.50 2.50 nd

2.41 nd

nd 1.20 nd

nd 1.02 0.50 1.40 nd

K vs. high-K series) criteria, several groups of δ

O values

have been delimited. Descriptive statistics of the δ

O values

is presented in Table 7. For comparison, the variation of the

O values in mantle clinopyroxenes (amphiboles present) is

also shown (data from Mattey et al. 1994). To test for signifi-

cant differences between means, one-way analysis of variance

test (ANOVA) has been performed giving p=0.0026 (differ-

ences are significant at p<0.05). The pyroxenes from cognate

enclaves show the closest values to the mantle. Comparable

values were found in Iliuþa basaltic andesite. The lowest val-

ues are shown by the Mãguri andesites (high-K series). The

highest δ

O values were found in the Cornii (6.6 ) and

Colibiþa andesites (6.7 ). The most scattered values are

shown by amphiboles from the cognate enclaves. In some of

the analysed enclaves (I4px1, I4px2, I4px4, I8/hb2) isotopic

disequilibria between amphiboles and coexisting pyroxenes

can be observed.

The δ

O value for garnet from the Pleºii quartz garnet

andesite (4.3 ) lies within the range of the δ-values exhibit-

ed by the amphiboles from the medium-K series. Much higher

O value (7.3 ) was obtained for the metamorphic garnet

(alm 61 %; and 23 %; sps 4 %; prp 12 %) from the staurolite

garnet micaschists within the Rebra series. The δ

O value of

the Pleºii garnet is lower than the values for garnet reported in

the literature (Harangi et al. 2001 and literature there in). For

the Carpatho-Pannonian region the δ

O values of the igneous

almandine garnet range from 6.1 to 10.5 (Mason et al.

1996; Harangi et al. 2001).

Age correction of the measured

Sr/

Sr ratio values (using

data from Pécskay et al. 1995) has been performed in order to

PETROGENESIS OF CALC-ALKALINE ROCKS: NEOGENE SUBVOLCANIC AREA (EASTERN CARPATHIANS) 85

Table 5: Continuing.

Sample

Location

Rock type

P95

Chicera

Runc

P11

Heniu

Oala

I16

Colibita

I18

Iliuþa

Cognate

enclave

I10/en

V.Vin.

I14/en

Cornii

I4/A2

Arsente

I4/A3

Arsente

I4 px0

Arsente

I4 px 1

Arsente

I4 px 4

Arsente

I5/en

Chicera

I8/hb2

Heniu

I8/hb1

Heniu

I16/en

Colibita

SiO

56.93 52.46 52.83 53.08 58.96 52.42 53.78 SiO

47.91 41.68 48.62 42.52 49.87 46.97 47.05 39.81 43.31 42.33 47.62

TiO

0.65 0.93 1.07 1.12 0.67 0.72 0.95 TiO

1.73 1.86 1.29 1.17 0.35 0.73 0.69 2.00 1.67 1.40 1.23

17.51 17.02 18.78 18.15 17.82 18.94 18.76 Al

16.65 13.98 14.05 13.25 2.42 7.00 6.35 16.66 12.25 13.70 17.75

FeO*

6.27 7.98 10.60 8.17 5.86 9.21 7.69 FeO* 13.47 14.15 11.13 10.43 6.63 9.06 8.85 16.20 11.34 11.95 10.66

MnO

0.10 0.17 0.17 0.26 0.15 0.18 0.17 MnO 0.29 0.31 0.21 0.18 0.12 0.16 0.15 0.21 0.19 0.14 0.17

MgO

4.15 6.31 3.27 3.49 2.69 3.56 4.93 MgO 6.55 11.50 10.23 14.89 19.56 16.42 16.98 9.09 13.86 14.17 7.34

CaO

8.15 10.81 8.57 7.69 6.17 6.88 6.90 CaO 5.19 10.16 9.23 11.53 19.83 17.31 17.81 11.34 12.98 11.52 9.51

3.42 2.76 3.56 3.81 3.88 3.91 3.89 Na

O 2.45 1.99 1.88 2.33 0.31 0.93 0.82 1.98 1.60 2.08 2.77

1.69 1.37 1.00 1.34 1.18 1.12 1.06 K

3.3 1.17 2.31 0.68 0.04 0.33 0.28 0.39 0.87 0.76 0.81

0.14 0.12 0.13 0.14 0.19 0.2 0.17 P

0.12 0.19 0.06 0.1 0

0.09 0.02 0.11 0.05

0.06

1.67 0.96 0.75 2.10 2.11 2.06 1.07 H

O 2.56 2.25 1.00 2.26 0.89 1.00 1.00 1.88 1.77 2.11 1.94

Total

100.68 100.89 100.73 99.35 99.68 99.20 99.37 Total 100.22 99.24 99.34 73.39 100.02 100.00 100.00 99.67 99.89 100.16 99.89

50 29 16 69

157 11

319

454

472

256 210 442

327

220 129

195

154

163

127

190

40 23 17 22

39 28

102

181

116 73 143

146

58 48

120

nd 6

10 9 4 6

16 4

4 14 2 3

17 16

19 16 18 17

32 26

nd 77

nd 57

nd 73

217

nd 42

14 24 2.5 8

17 19

109

22 22 9 10

65 21

432

460

231

310 550 270

240

1000 0

584

210

nd 0

115

48 320 44 19

130

6 70

111

378

370

nd 80

237

340

191

320 120 160

240

300 28

411

334

300

273

267

320

450

330

440

28 25 11.7 15

90 15

nd 22

nd 0

280

210

32 35 32 43

110

210 50

nd 43

nd 63

310

nd 21 2 2.5

nd 3

5.5 3

nd 0

nd 11.41 11.43 31.31 13.85 30.00 La

<30

nd <30

<30

nd 10.40 10.42 29.71 11.68

nd 37

nd 2.61 2.65 5.82 3.00

nd 0.73 0.73 1.32 0.78

nd 0.55 0.55 0.57 0.45

nd 3.00 2.61 1.85 0.64 1.55 3.00 Yb

7 3

nd 5

nd 0.13 0.16

nd 1.91 nd 2

nd 4.30 2.71 2.73 12.21 4.00 10.00 Th

nd 3

nd <2

nd 2.01 nd

nd 0.53 nd

obtain the initial

Sr/

Sr ratios (Table 6). The differences be-

tween corrected and uncorrected values are negligible. The

initial

Sr/

Sr ratios have been used in all diagrams. The total

range of

Sr/

Sr ratios is from 0.70588 to 0.70950. The low-

est value is from the Iliuþa basaltic andesite from the medium-

K series. Variation curves of the host-rocks from medium-K

and high-K series plot as distinctive straight lines on the

Sr/

Sr vs. 1/Sr diagram (Fig. 5), suggesting two-component

mixing. The increase of the

Sr/

Sr ratios matches the in-

crease of the Sr content in the high-K series. For the medium-

K series there is a slightly negative correlation between the

two parameters. The cognate enclaves display

Sr/

Sr ratios

from 0.70620 to 0.70902. On the

Sr/

Sr vs. 1/Sr diagram,

the pyroxenites and the hornblendites define distinct fields.

There is a larger variation of Sr content within the cognate en-

claves than in the host rock. The similar

Sr/

Sr signature of

the cognate enclaves as compared with their host rocks is con-

sistent with their co-magmatic origin. In some cases higher

values were found in cognate enclaves relative to their host

rock. This could be explained by their relatively low Sr con-

Fig . 5.

Sr/

Sr vs. 1/Sr for host-rocks and cognate enclaves in the

Rodna-Bârgãu Mountains.

tent (41 to 195 ppm) that made them more susceptible to

contamination.

86 PAPP, URECHE, SEGHEDI, DOWNES and DALLAI

Discussion

The results of the geochemical study based on the major,

minor, rare earth elements, O and Sr isotopes can be used to

form a framework for discussion of the petrogenetic processes

that influenced the magma evolution in the region.

Processes related to magma differentiation

Differentiation by fractional crystallization is suggested by

the existence of a complete series of rocks containing all the

petrographic types, from basalts to rhyolites. The presence of

a great variety of mineral species: plagioclases, amphiboles,

pyroxenes, quartz, biotite, garnets, iron oxides etc. corre-

sponds to a succession of fractional crystallization. The pres-

ence of obvious trends of linear correlation of the major ele-

ments with differentiation indices (SiO

) also pleads for

magma differentiation (Fig. 3a). Another argument for frac-

tional crystallization is offered by the REE pattern. The nega-

tive anomaly of Eu found in all lithologies indicates the im-

portance of plagioclase fractionation. The negative Eu

anomaly could also be generated by partial melting with a pla-

gioclase-rich residue. Scattered variations of minor elements

with SiO

, indicate diverse conditions of differentiation for

Series

Sample

Location

Rock type

Sr/

cpx

hbl

grt

medium-K

I15

Sturzii

dacite

0.70818

0.70812

4.3

Mal

qtz grt andesite

0.70704

0.70699

I31

Pleºii

qtz grt andesite

4.8

4.3

Oala

amph andesite

0.70671

0.70668

5.2

Heniu

amph andesite

0.70929

0.70925

4.8

I8/hb1

Heniu

hornblendite

0.70798

0.70796

5.1

I8/hb2

Heniu

hornblendite

0.70902

0.70887

4.2

I18

Iliuþa

basaltic andesite

0.70593

0.70588

5.4

high-K

I10

V.Vinului

qtz bi andesite

0.7090

0.70889

I11

V.Vinului

qtz bi andesite

0.70749

0.70738

Arsente

microdiorite

0.70797

0.70791

3.9

I4/A2

Arsente

hornblendite

0.70620

0.70605

3.8

I4mega

Arsente

hornblendite

0.70954

0.70950

4.6

I4px0

Arsente

pyroxenite

0.70862

0.70860

5.7

I4px1

Arsente

pyroxenite

0.70633

0.70632

5.2

4.2

I4px2

Arsente

pyroxenite

0.70879

0.70877

4.6

5.2

I4px4

Arsente

pyroxenite

0.70941

0.70939

4.6

4.0

Chicera

microdiorite

0.70706

0.70700

4.6

3.7

I5en

Chicera

hornblendite

5.4

I14

Cornii

amph andesite

0.7090

0.70891

6.6

I14en

Cornii

hornblendite

5.7

5.6

Runc

px andesite

0.70794

0.70790

5.3

I16

Colibiþa

amph andesite

0.70845

0.70839

6.7

I16en

Colibiþa

hornblendite

5.2

Rebra

5/96

Rodna

micaschist

7.3

Table 6: Sr-O isotope analyses for representative host-rocks and cognate enclaves from the Rodna-Bârgãu Mountains. Sr isotope data are

expressed as measured and age (10.68.6 Ma) corrected

Sr/

Sr ratios.

different magmatic units. Consequently, each intrusive unit

may have encountered specific differentiation processes and

possibly did not evolve from a single source.

Similarities in mineralogy of the main mineral species, in P-T

conditions of amphibole crystallization, in incompatible trace

element patterns, and in strontium and oxygen isotope compo-

sition between cognate enclaves and their host rocks, clearly

indicate that cognate enclaves formed from the same magmat-

ic source. They could be products of magma mixing processes

resulting from repeated feeding of magma chambers with pa-

rental mafic melts. This could also explain why cognate en-

claves predominantly occur within larger intrusive units with

intermediate composition (Heniu, Mãguri, Cornii, and Valea

Vinului). The maintenance of an intermediate composition

during crystallization and crustal rock assimilation involves

new inputs of mafic magma. Oscillatory zoning and corrosion

shown by plagioclase phenocrysts from Arsente, Valea Vinu-

lui and Cornii intrusions might also indicate refill of the mag-

matic chamber with less evolved melts. However, an alterna-

tive possible explanation is that the cognate enclaves represent

broken fragments of cumulate layers formed on the floor of

the intermediate magma chambers.

Evidence for crustal assimilation

In subduction-related magmatism possible contamination

mechanisms are: source contamination related to the descend-

ing slab and its sediments, and/or crustal contamination within

crustal magma chamber achieved by assimilation-fractional

crystallization processes (AFC). The existing geotectonic

models in the Carpatho-Pannonian region (e.g. Rãdulescu &

Sãndulescu 1976; Csontos 1995) consider that during the sub-

duction process oceanic crust, thinned continental crust, as

well as related sediments (present-time External Carpathian

Flysch strata) were consumed.

Mantle

Cognate

enclaves

Medium-K

series

High-K

series

Cornii/Colibiþa

cpx

amph

cpx amph

amph

Mean

5.52

5.30 4.53

4.92

4.95 3.80

6.65

Min.

5.25

4.6

3.8

4.3

4.6 3.7

6.6

Max.

5.75

5.7

5.6

5.4

5.3 3.9

6.7

Std. dev. 0.16

0.52 0.76

0.42

0.49 0.14

Table 7: Summary of the distribution of the δ

O values.

PETROGENESIS OF CALC-ALKALINE ROCKS: NEOGENE SUBVOLCANIC AREA (EASTERN CARPATHIANS) 87

High

Sr/

Sr ratios are usually interpreted in terms of

crustal assimilation. As the increase of the

Sr/

Sr ratios is

also linked to the enrichment of lithophile trace elements as

compared to primitive mantle and MORB, crustal assimilation

process is to be considered in the petrogenesis of the Rodna-

Bârgãu magmatites. Crustal assimilation is also proved by the

presence of crustal enclaves (metamorphic and sedimentary

rocks).

The Th/Nb vs. SiO

diagram (Fig. 6) clearly shows discrim-

ination of the two series of rocks. The high-K series is charac-

terized by higher Th/Nb ratio suggesting a more important

contribution of the upper crust and/or sediments in their gener-

ation, whereas the medium-K series display lower Th/Nb ra-

tios is, therefore, less affected by contamination with such

components.

For the medium-K series the most probably contaminant is

the lower crust. Its involvement is supported by the presence

of garnet-bearing andesites. Petrogenesis of primary garnet-

bearing igneous rocks requires special composition and P-T

conditions of the magma, as well as special tectonic setting for

facilitating rapid ascent of the magma to the surface because

of the limited stability field of the garnet. It is explained by

partial melting of the lower crust due to the uprising of man-

tle-derived magmas, producing small volumes of acid rocks

with a high content of Al

(e.g. Harangi et al. 2001). The

assemblage garnet+plagioclase+amphibole+quartz is stable

at pressures higher than 800 MPa and temperatures of 800

850 °C in a dacite melt with 5 % H

O (Day et al. 1992). Such

highly hydrated dacite magma can be generated from a basal-

tic precursor with 23 % water (Green 1992). For the garnet

bearing andesites (Pleºii-Mal) and dacites (Sturzii) high inter-

nal pressure of the magma due to a high H

O-rich fluid con-

tent, and the appearance of a decompression regime could fa-

cilitate relatively rapid ascent of magma and preservation of

garnets. The occurrence of the Rodna-Bârgãu garnet-bearing

rocks close to the Someº Fault and the lack of cognate en-

claves, which would indicate the existence of intermediate

chambers, are other arguments for a rapid ascent of magma. A

hydrous mantle source, which is another essential requirement

for the formation of primary garnet-bearing rocks, is consis-

tent with the elevated Ba/La ratio of the rocks from the medi-

um-K series and also with the exclusive occurrence of hydrous

mafic minerals within this series. Exploded fluid inclusions,

surrounded by other small fluid inclusions, found within

quartz grains from Bucnitori dacite, clearly suggest that de-

compression took place (Papp et al. 2003) facilitating rapid as-

cent of magma and preservation of garnets. Pressure decreas-

ing, inferred from the chemical composition of amphiboles in

Sturzii dacite, also suggests a decompression regime.

Significance of the low oxygen isotope ratios

Different processes could induce depletion in

O of the

magmatic rocks. Typically, low-

O rocks are interpreted in

terms of hydrothermal alteration in the upper crust by interac-

tion with cold meteoric water or seawater. Low temperature

(between 200 and 400 °C) hydrothermal systems with short

lifetimes (<10

years) produce large isotopic effects and im-

portant isotopic disequilibrium between rock-forming miner-

als. Other mineralogical and petrographic effects typically oc-

cur (Taylor 1974): partial or total alteration of primary mafic

minerals, micrographic intergrowths of alkali feldspar and

quartz, miarolitic cavities and veins filled with quartz, alkali

feldspar, chlorite, etc. Conversely, sub-solidus hydrothermal

exchange at very high temperatures (400800 °C) is compati-

ble with the general absence of hydrous alteration products in

the mineral assemblages, with the presence of clinopyroxenes,

and with a relatively uniform oxygen isotope composition in

all lithologies (Hoefs 1997 and literature therein). In certain

cases, low-δ

O magmas may be formed by remelting of hy-

drothermally altered country rocks or by large-scale assimila-

tion of such material.

Hydrothermal alteration products are poorly developed or

absent in most of the intrusive bodies under study. δD values

measured on whole-rocks (andesites, microdiorites, rhyolites)

or amphiboles vary from 55 to 75 (Papp 1999) and

are normal values for igneous rocks. Because of the high

diffusion rate of hydrogen in rocks, late low-temperature inter-

action with crustal fluids would have produced important

shifts towards low δD values. All these observations, as well

as the existence of a correlation between oxygen and stron-

tium isotopic ratios, lead us to discuss the low measured δ

values in terms of primary isotopic characteristics of the pa-

rental magmas. However, sub-solidus exchange with meteoric

underground water has to be taken into consideration as a

complementary process during the cooling history of the

rocks, also leading to depletion in

On the

Sr/

Sr vs. δ

O diagram (Fig. 7), the medium-K

series defines a significant negative correlation (correlation

coefficient, r=0.72). The trend starts from basaltic andesites

(Iliuþa) to andesites (Oala), quartz andesites (Pleºii-Mal) and

dacites (Sturzii). The decrease of the δ

O values as

Sr/

Sr ra-

tios and SiO

increase is interpreted as a progressive contamina-

tion of a mantle-derived magma with a contaminant depleted in

Fig. 6. Th/Nb vs. SiO

diagram for Rodna-Bârgãu magmatites.

Flysch sediments (SED) and upper crust (UC) from Mason et al.

(1996), lower crust (LC) from Kempton et al. (1997). Different

fields for cognate enclaves, medium-K and high-K series are

shown. Symbols as in Fig. 2.

88 PAPP, URECHE, SEGHEDI, DOWNES and DALLAI

O and enriched in

Sr/

Sr (i.e. hydrothermally altered crust-

al rocks). As shown in the previous paragraph lower crust could

be taken into consideration as the crustal component. A lower

crust characterized by low δ

O values (<4.3 ) and medium

to high

Sr/

Sr ratios (>0.710) has to be assumed in order to

explain the isotopic characteristic of these rocks. Kempton et al.

(1997) reported similar low δ

O values for the lower crust in

the Carpatho-Pannonian area, although they correlate with

much lower

Sr/

Sr ratios (see Fig. 7).

Iliuþa basaltic andesite is the least contaminated rock, show-

ing δ

O values and

Sr/

Sr ratios close to a mantle source.

Taking into account its volcanic-like texture and the small vol-

ume of the intrusion, we can assume that the magma solidified

rapidly near the surface, without important interaction with

meteoric water to affect the initial δ

O value.

The petrogenesis of primary garnet-bearing rocks (Pleºii-

Mal and Sturzii) imply a rapid ascent of magmas, without sig-

nificant contamination within the middle/upper crust. Garnet

is a highly refractory mineral to oxygen isotope exchange. In

the Pleºii andesite the garnet appears to be in isotopic equilib-

rium with the coexisting amphibole (4.8 ). Equilibrium oxy-

gen fractionation between garnet and amphibole is less than

0.4 in the temperature range of 8501000 °C (Zheng 1993a,b).

Moreover, the garnets of the Pleºii andesite display a signifi-

cantly different δ

O value compared with the garnets of the

Rebra series (7.3 ) and therefore a xenocrystic metamorphic

origin for the garnets from the quartz andesite could be ruled

out suggesting a primary magmatic origin. Taking all these

facts into consideration, the garnets of the Pleºii andesite

could be indicative for the oxygen isotope composition of the

magma from which it crystallized (i.e. 4.3 ).

In spite of its more basic composition relative to the Pleºii

quartz andesite and Sturzii dacite, Heniu amphibole andesite

exhibits higher

Sr/

Sr ratios and higher δ

O value. A pro-

cess of assimilation and equilibrium crystallization (AEC)

(Huppert & Sparks 1985) could be a possible explanation for

the increasing

Sr/

Sr. Hotter and voluminous mafic mag-

mas are able to assimilate more crustal material than cooler

less voluminous acid ones. The preservation of an intermedi-

ate composition during assimilation and crystallization im-

plies new inputs of mafic magma, which are supported by the

presence of cognate enclaves (Heniu is by far the largest intru-

sive unit).

Within the high-K series a correlation between the δ

O val-

ues,

Sr/

Sr ratios and SiO

content cannot be identified

(Fig. 7). For the Mãguri andesites the distribution is scattered.

In addition, oxygen isotope disequilibrium between coexisting

pyroxenes and amphiboles, both in host-rocks and cognate en-

claves have been observed, as well as additional mineralogical

and petrographical alteration effects (some alteration of prima-

ry mafic minerals, miarolitic cavities and veins filled with

quartz and calcite). These features are characteristics of hydro-

thermal systems with a short lifetime. Therefore, for the

Mãguri intrusions the most probable explanation for the deple-

tion in

O is the interaction between intrusive bodies and

heated meteoric water. However, the isotopic heterogeneities

observed within the Mãguri andesites might be due to an inho-

mogeneous source. This could be an incompletely homoge-

nized MASH zone (mixing, assimilation, storage and homoge-

nization; Hildreth & Moorbath 1988), which operated at the

lower-crustal depths. The presence of MASH zones was previ-

ously suggested for the CãlimaniGurghiuHarghita segment

(Mason et al.1996).

Cornii and Colibiþa andesites, from the high-K series, are

characterized by the highest δ

O (6.7 ) and

Sr/

Sr ratio

(0.709), which imply extensive crustal assimilation. These

rocks are well crystallized, and have high K and Sr contents.

They form large intrusions. All these features can be explained

by stagnation of magma in large chambers, situated in the

mid-crustal depth, where AFC processes took place. Distinc-

tive equilibrium mantle-like δ

O values of the pyroxene and

amphibole from the cognate enclaves within Cornii andesite

indicate an origin from a mantle-derived uncontaminated

source. The higher δ

O values of the host rocks are not con-

sistent with assimilation of a highly altered crust.

Conclusions

The geotectonic evolution and the magmatic processes in

the Rodna-Bârgãu Mountains are an intrinsic part of the gen-

eral evolution of the Carpatho-Pannonian region during Ter-

tiary times. The East Carpathian magmatic arc is closely relat-

ed to the subduction processes located at the southwestern

border of the Eurasian plate.

The results of the mineralogical and geochemical study (ma-

jor and trace elements, oxygen and strontium isotopes) allowed

us to evaluate the petrogenetic processes which influenced mag-

matic evolution. Each intrusive unit encountered specific differ-

entiation processes by fractional crystallization, crustal assimi-

lation and magma mixing. Two different series of rocks have

been separated: one medium-K and another high-K.

Fig. 7. δ

O (mineral separate) vs.

Sr/

Sr (whole-rock) diagram

for Rodna-Bârgãu igneous rocks. Bulk rock fields of Pannonian Ba-

sin lower crust (LC) (Kempton et al. 1997) and for East Carpathian

arc including CãlimaniGurghiuHarghita segment (CGH) (Mason

et al. 1996) and Ukrainian Carpathians (Ukr) (Seghedi et al. 2001)

are also shown for comparison. The Cãlimani low-K group (LKC) is

also highlighted. See text for discussions. Mineral phases from the

same rock are indicated by tie lines.

PETROGENESIS OF CALC-ALKALINE ROCKS: NEOGENE SUBVOLCANIC AREA (EASTERN CARPATHIANS) 89

The rocks of the medium-K series are the oldest, being em-

placed at about 10.6 Ma. Mantle-derived magma, contaminat-

ed with lower-crustal material is considered to be the source of

these rocks. The magmas had a rapid ascent toward the sur-

face, without a long-lasting stagnation period in intermediate

crustal magmatic chambers, as proven by the presence of pri-

mary garnet bearing rocks, or by the sporadic occurrence of

cognate enclaves. This is the case of Parva, Sturzii-Bucnitori,

Pleºii-Mal, Oala and Iliuþa Intrusive Units, which form small

intrusive bodies. An exception is Heniu Intrusive Unit, which

may have paused in the upper/middle crust where it experi-

enced an AEC process. Within this series, the decreasing of

the δ

O values as

Sr/

Sr ratios and SiO

increase is inter-

preted as a progressive contamination of a mantle derived

magma with a contaminant depleted in δ

O and enriched in

Sr/

Sr (i.e. hydrothermally altered lower crustal rocks).

The high-K series has been emplaced later in the evolution

of the magmatic activity (~9 Ma). The evolution of these mag-

mas was more complex; thus, the presence of intermediate

magma chambers needs to be considered. The magmas of the

Cornii and Valea Vinului Intrusive Units stagnated in large

chambers (see the large volume of the intrusions) where AFC

processes took place. These rocks are well crystallized, have

high K and Sr content, as well as higher

Sr/

Sr and

ratios. The higher δ

O values and

Sr/

Sr ratios displayed

by Cornii and Colibiþa Intrusive Units are consistent with as-

similation of a different contaminant from that involved in

petrogenesis of the medium-K series. This could be the local

upper-crustal schists. The Mãguri Intrusive Units are the

youngest rocks in the area. They show characteristics of a hy-

drothermal system: oxygen isotope disequilibrium between

coexisting pyroxenes and amphiboles, both in host-rocks and

cognate enclaves, as well as additional mineralogical and pet-

rographical alteration effects.

Acknowledgments: Special thanks are addressed to Prof. G.

Cavarretta who made it possible that an important part of the

present study was carried out at the CNR Centro di Studio

per il Quaternario e lEvolutione Ambientale, Rome, within

the No. 52 NATO Fellowships Programme of which D.C.

Papp benefited. Laboratory managers and staff are thanked for

their help during analytical work: Prof. M. Barbieri for carry-

ing out strontium isotope analyses, A.M. Conte for the XRF

measurements, M. Serracino for technical assistance during

microprobe determinations and Prof. B. Turi for providing ac-

cess to oxygen isotope facility. M.L. Frezzotti and F. Tecce

are thanked as well for their invaluable help at various stages

of this work. The authors thank Dr. J. Lexa and Dr. M. Munte-

anu for the critical reviewing of the manuscript.

References

Balintoni I. 1996: Alpine structural outline of the Pannonian-Car-

pathian realm. Stud. Univ. Babes-Bolyai, Geol., XL, 1, 5572.

Csontos L. 1995: Tertiary tectonic evolution of the Inter-Carpathian

area: a review. Acta Vulcanol. 7, 114.

Day R.A., Green T.H. & Smith I.E.M. 1992: The origin and signifi-

cance of garnet phenocrysts and garnet-bearing xenoliths in

Miocene calc-alkaline volcanics from Northland, New Zealand.

J. Petrology 33, 125161.

Dobosi G., Kempton P., Downes H., Embey-Isztin A., Thirlwall M.

& Greenwood P. 2003: Lower crustal granulite xenoliths from

the Pannonian Basin, Hungary. Part 2: SrNdPbHf and O

isotope evidence for formation of continental lower crust by

tectonic emplacement of oceanic crust. Contr. Mineral. Petrol-

ogy 144, 67168.

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. Acta Vulcanol. 7, 2941.

Eiler J.M. 2001: Oxygen isotope variation of basaltic lavas and up-

per mantle rocks. In: Valley J.W. & Cole D.R. (Eds.): Stable

isotope geochemistry. Rev. Mineral. Geochem. 43, 319364.

Green T.H. 1992: Experimental phase equilibrium studies of garnet-

bearing I-type volcanics and high-level intrusives from North-

land, New Zeeland. Trans. Roy. Soc. Edinb. Earth Sci. 83,

429438.

Harangi Sz., Downes H., Kosa L., Szabo Cs., Thirlwall M.F., Mason

P. & Mattey D. 2001: Almandine garnet in calc-alkaline volca-

nic rocks of the Northern Pannonian Basin (Eastern-Central

Europe): geochemistry, petrogenesis and geodynamic interpre-

tations. J. Petrology 42, 18131843.

Hildreth W. & Moorbath S. 1988: Crustal contribution to arc mag-

matism in the Andes of Central Chile. Contr. Mineral. Petrolo-

gy 98, 455489.

Hoefs J. 1997: Stable isotope geochemistry. Springer-Verlag, Ber-

lin, 1201.

Hupert H.E. & Sparks R.S.J. 1985: Cooling and contamination of

mafic and ultramafic magmas during ascent through the conti-

nental crust. Earth Planet. Sci. Lett. 74, 371386.

Johnson M.C. & Rutherford M.J. 1989: Experimental calibration of

the aluminium-in-hornblende geobarometer with application to

Long Valley Caldera (California) volcanic rocks. Geology 17,

837841.

Kempton P.D., Downes H. & Embey-Isztin A. 1997: Mafic granu-

lites in Neogene alkali basalts from Western Pannonian Basin:

insight into the lower crust of collapsed orogen. J. Petrology

38, 7, 940969.

Leake B.E., Woolley A.R., Arps C.E.S., Birch W.D., Gilbert M.C.,

Grice J.D., Hawthorne F.C., Kato A., Kisch H.J., Krivovichev

V.G., Linthout K., Laird J., Mandarino J.A., Maresch W.V.,

Nickel E.H., Rock N.M.S., Schumacher J.C., Smith D.C.,

Stephenson N.C.N., Ungaretti L., Whittaker J.W. & Youzhi G.

1997: Nomenclature of amphiboles: report of the Subcommit-

tee on Amphiboles of the International Mineralogical Associa-

tion, Commission on New Minerals and Mineral Names. Amer.

Mineral. 82, 19191037.

LeMaitre R.W. 1989: A classification of igneous rocks and glossary

of terms. Blackwell, Oxford, 1193.

Ludwig K.R. 1994: Analyst version 2.20: a computer program for

control of a thermal-ionization, single collector mass-spectrom-

eter. U.S. Department of Interior Geologcal Survey, Open File,

Report, 92543.

Mason P., Downes H., Thirlwall M., Seghedi I., Szakács A., Lowry

D. & Mattey D. 1996: Crustal assimilation as a major petroge-

netic process in the East Carpathian Neogene and Quaternary

continental margin arc, Romania. J. Petrology 37, 4, 927959.

Mason P.R.D., Seghedi I., Szakács A. & Downes H. 1998: Magmat-

ic constraints on geodynamic models of subduction in the East-

ern Carpathians, Romania. Tectonophysics 29, 157176.

Mattey D. & Mcphearson C. 1993: High-precision oxygen isotope

microanalysis of ferromagnesian minerals by laser fluorination.

Chem. Geol. 105, 305318.

Mattey D., Lowry D. & Mcphearson C. 1994: Oxygen isotope compo-

sition of mantle peridotite. Earth Planet. Sci. Lett. 128, 231241.