Page 1
ORIGINAL PAPER
Geochemistry and tectonic setting of mafic rocksfrom the Othris Ophiolite, Greece
Matthias G. Barth Æ Tatjana M. Gluhak
Received: 26 July 2007 / Accepted: 3 June 2008 / Published online: 17 June 2008
� Springer-Verlag 2008
Abstract We present new geochemical analyses of
minerals and whole rocks for a suite of mafic rocks from
the crustal section of the Othris Ophiolite in central Greece.
The mafic rocks form three chemically distinct groups.
Group 1 is characterized by N-MORB-type basalt and
basaltic andesite with Na- and Ti-rich clinopyroxenes.
These rocks show mild LREE depletion and no HFSE
anomalies, consistent with moderate degrees (*15%) of
anhydrous partial melting of depleted mantle followed by
30–50% crystal fractionation. Group 2 is represented by
E-MORB-type basalt with clinopyroxenes with higher Ti
contents than Group 1 basalts. Group 2 basalts also have
higher concentrations of incompatible trace elements with
slightly lower HREE contents than Group 1 basalts. These
chemical features can be explained by *10% partial
melting of an enriched mantle source. Group 3 includes
high MgO cumulates with Na- and Ti-poor clinopyroxene,
forsteritic olivine, and Cr-rich spinel. The cumulates show
strong depletion of HFSE, low HREE contents, and LREE
enrichments. These rocks may have formed by olivine
accumulation from boninitic magmas. The petrogenesis of
the N-MORB-type basalts and basaltic andesites is in
excellent agreement with the melting conditions inferred
from the MOR-type peridotites in Othris. The occurrence
of both N- and E-MORB-type lavas suggests that the
mantle generating the lavas of the Othris Ophiolite must
have been heterogeneous on a comparatively fine scale.
Furthermore, the inferred parental magmas of the SSZ-type
cumulates are broadly complementary to the SSZ-type
peridotites found in Othris. These results suggest that the
crustal section may be genetically related to the mantle
section. In the Othris Ophiolite mafic rocks recording
magmatic processes characteristic both of mid-ocean rid-
ges and subduction zones occur within close spatial
association. These observations are consistent with the
formation of the Othris Ophiolite in the upper plate of a
newly created intra-oceanic subduction zone.
Keywords Ophiolite � Subduction � Basalt �Magma genesis � Hellenides � Jurassic
Introduction
Ophiolites are products of complex tectonic and magmatic
processes that operated during the initial rifting through
seafloor spreading to subduction-facilitated emplacement
stages of ancient oceanic lithosphere in various tectonic
settings (e.g., Coleman 1977; Nicolas 1989; Dilek et al.
2005). Tethyan-type ophiolites (cf. Moores 1982) experi-
enced active margin tectonics during their incorporation
into continental margins through collisional processes and
in many cases also during their magmatic evolution (Pearce
et al. 1984). Ophiolites representing different stages of
evolution of an ocean basin may display different structural
and petrological features. The crustal section of Tethyan-
type ophiolites may be genetically related to the mantle
section if both formed in an oceanic environment, either by
seafloor spreading or in the upper plate of an intra-oceanic
subduction zone (the Mediterranean-type ophiolites of
Communicated by T.L. Grove.
Electronic supplementary material The online version of thisarticle (doi:10.1007/s00410-008-0318-9) contains supplementarymaterial, which is available to authorized users.
M. G. Barth (&) � T. M. Gluhak
Institut fur Geowissenschaften, Universitat Mainz,
Becherweg 21, 55099 Mainz, Germany
e-mail: [email protected] ; [email protected]
123
Contrib Mineral Petrol (2009) 157:23–40
DOI 10.1007/s00410-008-0318-9
Page 2
Dilek 2003). However, in some ophiolites spatially asso-
ciated depleted peridotites and basaltic rocks are not linked
by a genetic melt and residua relationship (the Ligurian-
type of Dilek 2003). These ophiolites may have formed
during the early stages of opening of an ocean basin, fol-
lowing continental rifting and breakup (Rampone and
Piccardo 2000). Therefore, the evidence (or lack) of a co-
genetic relationship between mantle peridotites and
associated magmatic rocks can help to constrain the tec-
tonic setting and specific mode of generation of an
ophiolite.
In this paper, we present new geochemical analyses of
minerals and whole rocks for a suite of mafic rocks from
the crustal section of the Othris Ophiolite, Greece, and
discuss the melting conditions and mantle sources of these
mafic rocks. The main goals of this contribution are to test
if the crustal section is genetically linked to the mantle
peridotites and to further constrain the evolution and tec-
tonic setting of the Othris Ophiolite. We also integrate
microprobe data we previously collected on ultramafic
rocks of the mantle section that were never fully published
before. Results of our petrographic and microprobe survey
of the Othris mantle rocks are presented in the electronic
supplementary material (eAppendix 1 and eTables 1, 2, 3,
4, 5).
Geological setting
The Othris Ophiolite in central Greece is located in the
Mirdita-Subpelagonian zone and is part of a NNW-trending
belt, which includes the western Hellenic ophiolites in
Greece, the Mirdita ophiolites in Albania, and the Dinaric
ophiolites in Serbia and Croatia. These Jurassic ophiolites
are bounded on the east and the west by the Korabi-Pela-
gonian and the Apulian microcontinents, respectively (see
review by Robertson 2002). The Axios–Vardar Zone east
of the Korabi-Pelagonian microcontinent forms a second
ophiolite belt that is subparallel to the Mirdita-Subpela-
gonian zone. The ophiolites within both belts are
interpreted to be remnants of the Neotethys Ocean, a net-
work of small ocean basins and microcontinents, which
existed between Eurasia and Gondwana during the Meso-
zoic–Early Tertiary (e.g., Robertson et al. 1991; Stampfli
and Borel 2002). The paleogeographic origin of the Othris
Ophiolite within the Neotethys is debated (e.g., Smith
1993; Robertson and Shallo 2000). One view expressed by
many authors is that the Othris Ophiolite formed in the
Mirdita-Pindos ocean, a relatively narrow ocean basin
lying west of the Korabi-Pelagonian microcontinent (e.g.,
Robertson and Karamata 1994). An alternative view is that
the Othris Ophiolite originated in the Meliata-Vardar
ocean, lying east of the Korabi-Pelagonian microcontinent
(e.g., Smith and Spray 1984). In this paper we assume an
origin for the Othris Ophiolite within the Mirdita-Pindos
ocean. According to Robertson and Shallo (2000), the
Mirdita-Pindos oceanic basin, located between the Apulian
continent in the west and the Korabi-Pelagonian micro-
continent in the east, opened during the Late Triassic–Early
Jurassic. During the Middle Jurassic (160–170 Ma) a
south-westward-dipping intra-oceanic subduction zone
became active within the Mirdita-Pindos oceanic basin.
Subsequently, the ophiolites were emplaced onto the con-
tinental margin during the Late Jurassic–Early Cretaceous.
In the Albanian sector of the Mirdita-Subpelagonian
zone, individual ophiolite massifs in the west near the
Apulian microcontinent are mainly composed of mid-
ocean ridge (MOR)-type basalts and lherzolites, whereas
those in the east near the Korabi-Pelagonian microconti-
nent contain dominantly supra-subduction zone (SSZ)-type
basalts and harzburgites (Robertson and Shallo 2000;
Shallo and Dilek 2003; Saccani et al. 2004). In the Greek
sector, a sharp geographical distinction cannot be made.
The Vourinos ophiolite in the east of the Mirdita-Subpel-
agonian zone, which is characterized by mafic and
ultramafic sequences with island arc and boninitic affinities
(Bizimis et al. 2000; Saccani et al. 2004), is a SSZ ophio-
lite, comparable to the eastern belt in Albania. By con-
trast, pure MOR-type ophiolites, similar to the western
belt in Albania, are subordinate. The crustal sections of
the Pindos and Othris ophiolite complexes are thought to
have formed in more than one tectonic setting as their
basalts exhibit both mid-ocean ridge (MOR) and island
arc affinities (Pearce et al. 1984; Jones and Robertson
1991; Photiades et al. 2003; Saccani and Photiades 2004).
In addition, the mantle section of the Othris Ophiolite
shows evidence for both comparatively fertile MOR-type
lherzolites and depleted SSZ-type harzburgites (Barth
et al. 2008).
The Othris Mountains of central Greece extend from
east of Almiros to the Pindos Mountains west of Lamia
(Fig. 1). The Subpelagonian zone in the Othris Moun-
tains consists of a sequence of thrust sheets that had
been emplaced onto the Triassic-Jurassic carbonate
platform overlying the late Carboniferous Hercynian
basement (Hynes et al. 1972; Smith et al. 1975; Ferriere
1985). These thrust sheets formed an ordered progression
from a submarine fan into structurally higher and pene-
contemporaneous pelagic basin sections and finally
ophiolites at the highest stratigraphic level (Hynes 1974;
Menzies and Allen 1974; Smith et al. 1975; Smith
1993). The ophiolitic units of Othris are stacked in
reverse stratigraphic order: cherts and pillow lavas at the
base have been overthrust by sheeted dikes and gabbro
cumulates and then by ultramafic cumulates, near-Moho
mantle rocks, a harzburgite thrust sheet, and finally a
lherzolite-plagioclase lherzolite thrust sheet on top
24 Contrib Mineral Petrol (2009) 157:23–40
123
Page 3
(Rassios and Konstantopoulou 1993; Rassios and Smith
2000).
The interpretation of the origin and emplacement of the
Othris Ophiolite has changed over time and remains con-
troversial (Smith and Rassios 2003). The first models that
placed the Othris Ophiolite in a plate tectonic setting
interpreted it as being formed near a continental margin at
the inception of rifting (Hynes 1974; Menzies and Allen
1974; Menzies 1976) and subsequently being emplaced by
convergent margin tectonics at a subduction zone (Hynes
et al. 1972). Later studies argued for a relatively slow-
spreading mid-ocean ridge environment (Rassios and
Konstantopoulou 1993; Dijkstra et al. 2001, 2003) or an
island-arc environment (Bizimis et al. 2000; Rassios and
Smith 2000). Recently, Barth et al. (2008) proposed that
the Othris Ophiolite originated during the initial stages of
subduction at or near a mid-ocean ridge, when oceanic
extension rapidly changed to convergence.
Sampling
For the present study sampling was focused on mafic
volcanic rocks in order to investigate the geochemical
variability of the crustal section. In the Othris complex
there are two separate magmatic series (Smith et al.
1975): an earlier suite formed in a continental margin, the
Agrilia Formation, which was probably related to the
opening of the ocean basin, and a later suite, the Mirna
Group, in which gabbroic cumulates and ultramafics were
identified representing spreading in that basin. In general,
outcrop is poor in the crustal section of the Othris
Ophiolite. Therefore, the majority of samples were col-
lected from roadcuts on the road from Lamia to Domokos.
A total of eleven samples were collected (Fig. 1;
Table 1): five from the Agrilia Formation (samples A1,
A9, A10, A11, and A32), two from the Mirna group
(samples M3 and M30), and four from the Sipetorrema
Pillow Lava unit (S5, S6, S7, and S8). The Triassic
Agrilia Formation is part of the Othris group (Smith et al.
1975). The Sipetorrema Pillow Lava unit is part of the
Jurassic Mirna Group, which contains the ophiolite com-
plex and which has been thrust over the Othris Group
Katáchloron
Eretria
FournosKaïtsa
10 km
Almiros
Domokos
Lamia
Stillis
Gulf of MaliakosN
Legendultramafic rocks
mafic rockspillow lavas
Vourinos
Othris
22.5°E
39°N
The Othris Ophiolite22°E
Pindos
Kédros
shearzone FK
M30
A1A9A10A11
M3
S5
S6
S7S8
A32
Fig.1 Simplified geological
map of the Othris Ophiolite
showing the location of the
study area. Modified from
Rassios and Konstantopoulou
(1993). Inset: location map
showing the Othris, Pindos, and
Vourinos ophiolites
Table 1 Locations of the sample collection sites of the mafic rocks
from the Othris Ophiolite, Greece
Sample Type Latitude Longitude
Agrilia Formation
A1 Pillow basalt 38�56039.200N 22�24024.800E
A9 Pillow basalt 38�56025.500N 22�23034.400E
A10 Pillow basalt 38�56018.300N 22�23040.100E
A11 Pillow basalt 38�56014.800N 22�23040.400E
A32 Pillow basalt 38�5606.500N 22�24042.800E
Mirna Group
M3 Dolerite 39�0040.800N 22�22037.900E
M30 Dolerite 39�0057.300N 22�1609.900E
Sipetorrema Pillow Lava Unit
S5 Pillow basalt 38�59021.200N 22�22025.500E
S6 Pillow basalt 38�58043.600N 22�22046.900E
S7 Boninitic cumulate 38�5808.600N 22�23012.800E
S8 Boninitic cumulate 38�5808.600N 22�23012.800E
Contrib Mineral Petrol (2009) 157:23–40 25
123
Page 4
(Smith et al. 1975). Samples labeled as Mirna Group are
mafic rocks collected from the Mirna Group excluding the
Sipetorrema Pillow Lava unit.
Analytical methods
Major elements
Whole rock major elements and the trace elements Sc, V,
Cr, Ni, Co, Cu, and Zn were determined by X-ray fluo-
rescence (XRF) on fused glass beads and pressed powder
pellets at the University of Mainz.
Mineral major element compositions of the mafic rocks
were determined using the Jeol JXA-8900 RL wavelength-
dispersive electron microprobe (EMP) at the University of
Mainz. Olivine, orthopyroxene, clinopyroxene, and spinel
were analyzed using an accelerating potential of 20 kV, a
beam current of 12 nA, and a spot size of 2 lm. An
accelerating potential of 15 kV, a beam current of 8 nA,
and a spot size of 5 lm were used to analyze plagioclase.
Trace elements
To get high-precision whole rock data for rare earth
elements and other trace elements, we have applied a
recently developed laser ablation-inductively coupled
plasma-mass spectrometry (LA–ICP-MS) technique using
an automated iridium strip heater (Nehring et al. 2008).
About 40 mg of the rock powder was placed on an irid-
ium strip without any flux agent. The melting of the
samples to glass beads took place in a closed box under
an argon atmosphere to suppress oxidation and to limit
volatilization of elements with low boiling points (e.g.,
Cs, Pb). Melting conditions for basaltic samples were
1,200�C and 10 s. The fused glass beads were analyzed
by LA–ICP-MS at the University of Mainz. Ablation was
achieved with a NewWave Research UP-213 Nd:YAG
laser ablation system, using a pulse repetition rate of
10 Hz, pulse energies of *0.3 mJ, and 100 lm crater
diameters. Analyses were performed on an Agilent
7500ce inductively coupled plasma-mass spectrometer in
pulse counting mode (one point per peak and 10 ms dwell
time). Data reduction was carried out using the software
‘‘Glitter’’. The amount of material ablated in laser sam-
pling is different for each spot analysis. Consequently, the
detection limits are different for each spot and are cal-
culated for each individual acquisition. Detection limits
generally range between 0.001 and 0.5 ppm (lg/g). 44Ca
was used as internal standard. Analyses were calibrated
against the silicate glass reference material NIST 612
using the values of Pearce et al. (1997), and the US
Geological Survey (USGS) glass standard BCR-2G was
measured to monitor accuracy.
Results
Petrography
All samples except S7 and S8 have aphanitic textures
typical for mafic volcanic rocks.
The five samples from the Agrilia formation display
fine-grained microporphyritic textures with euhedral to
subhedral elongate plagioclase microphenocrysts and
granular clinopyroxene microphenocrysts embedded in a
groundmass consisting of lath-shaped plagioclase micro-
lites and granular pyroxenes.
The two samples from the Mirna Group have very fine-
grained microporphyritic to intergranular textures. In mi-
croporphyritic varieties, phenocrysts are represented by
euhedral to subhedral plagioclase, clinopyroxene, and rare
spinel. Clinopyroxene phenocrysts are often zoned.
Samples S5 and S6 from the Sipetorrema Pillow Lava
unit have the finest-grained groundmass among the samples
studies. The microphenocrysts are euhedral to subhedral
elongated plagioclase and granular clinopyroxene.
Samples S7 and S8 from the Sipetorrema Pillow Lava
unit are highly phyric basalts containing abundant euhedral
to subhedral olivine and pyroxene phenocrysts 2–3 mm in
size, suggestive of olivine and pyroxene accumulation.
Additional phenocrysts are represented by orthopyroxene
and minor spinel.
All samples show extensive to high degrees of hydro-
thermal ocean-floor alteration. This alteration is manifested
by the presence of carbonate + iron oxide ± chlorite veins
and crusts replacing the primary lithology. Plagioclase is
partly to completely albitized.
Whole rock chemistry
The major element contents of whole rocks from the Othris
Ophiolite are presented in Table 2 and plotted in element
oxide versus Mg# [100 9 Mg/(Mg + Fe)] abundance plots
(Fig. 2).
As stated in the previous section, all samples are
extensively to highly altered. This observation is reflected
by the high measured LOI values (3.05–11.4%, Table 2).
The chemical effects of hydrothermal alteration, that is,
variable mobilization of CaO, MgO, alkali elements (Na,
K) and large ion lithophile elements (LILE) such as Rb, Ba,
and Sr (Staudigel 2003), can be observed in the samples.
For example, CaO contents in the samples from the Agrilia
Formation are highly variable, ranging from 5.08 to
14.9 wt%, reflecting both loss of CaO in sample A10 and
carbonate addition in sample A9 (Fig. 2c).
The covariation of some selected major and trace ele-
ments against Zr (used as an indicator of differentiation) is
shown in Fig. 3. No systematic elemental variation with
26 Contrib Mineral Petrol (2009) 157:23–40
123
Page 5
Table 2 Whole rock major and trace element composition of mafic rocks from Othris
Agrilia Formation Mirna Group Sipetorrema Pillow Lava
Sample A1 A9 A10 A11 A32 M3 M30 S5 S6 S7 S8Type E-M E-M N-M E-M E-M E-M N-M N-M N-M SSZ SSZ
SiO2 (wt%) 46.0 44.2 49.2 47.9 43.6 47.7 48.8 50.0 45.8 40.5 41.7
TiO2 1.09 0.69 1.60 0.78 1.02 1.05 2.22 1.20 1.25 0.26 0.33
Al2O3 14.3 11.1 14.7 13.0 13.5 15.3 11.8 14.6 15.1 4.43 4.97
FeO 8.60 8.30 11.6 7.41 8.83 8.82 10.9 11.3 10.7 9.59 9.95
MnO 0.11 0.15 0.17 0.14 0.12 0.18 0.18 0.19 0.17 0.14 0.15
MgO 6.51 6.73 6.95 5.35 6.27 7.47 3.75 7.33 7.71 30.8 26.7
CaO 10.3 14.9 5.08 13.4 12.5 7.46 9.39 11.6 11.5 3.43 3.84
Na2O 4.24 3.94 4.14 5.07 3.24 4.01 3.94 3.37 2.84 0.05 0.21
K2O 1.23 0.53 2.04 0.10 1.43 1.60 0.05 0.19 0.11 0.02 0.03
P2O5 0.22 0.08 0.14 0.10 0.20 0.12 0.19 0.08 0.10 0.05 0.08
Cr2O3 0.08 0.16 0.01 0.06 0.10 0.02 0.01 0.05 0.05 0.34 0.36
NiO 0.03 0.03 b.d. 0.01 0.02 0.01 0.01 0.01 0.01 0.18 0.18
LOI 7.18 9.06 4.25 6.71 9.02 6.17 8.61 0.05 4.66 10.05 11.40
Total 99.91 99.92 99.88 99.95 99.90 99.90 99.92 99.88 99.90 99.91 99.89
Mg# 57.4 59.1 51.6 56.3 55.9 60.1 38.0 53.5 56.3 85.1 82.7
Li (ppm) 30.6 5.98 18.5 4.39 28.5 30.0 2.5 30.9 24.6 29.0 73.6
Sc 39.7 35.8 36.5 36.9 39.3 44.6 55.0 44.1 44.6 22.6 24.1
V 288 251 328 262 320 321 555 271 257 120 173
Cr 522 472 36.2 299 698 146 113 332 326 1825 2039
Co 47 47 36 31 43 33 48 46 45 97 103
Ni 191 227 15.6 73.0 193 72.7 60.7 57.6 59.3 1413 1662
Cu 54 51 140 35 51 27 42 103 96 22 47
Zn 78 69 88 63 75 49 97 70 69 56 62
Ga 14 11 18 11 12 15 17 14 19 6 7
Rb 14.6 5.99 15.8 0.93 19.7 28.5 1.02 2.02 0.62 0.75 0.65
Sr 172 151 121 102 203 190 115 221 80.6 43.0 101
Y 19.8 15.6 34.4 17.2 18.5 20.6 51.5 24.6 24.4 5.95 8.18
Zr 67.7 41.2 101 45.8 63.6 75.2 172 58.6 78.0 15.2 22.7
Nb 22.0 3.98 2.48 4.81 20.6 9.28 3.74 1.28 2.43 0.33 0.50
Cs 0.48 0.36 0.10 0.03 0.97 0.43 0.07 0.03 0.02 0.15 0.16
Ba 62.8 74.1 65.2 8.92 69.6 475 7.20 7.48 6.87 1.43 6.40
La 13.9 4.15 4.67 3.95 12.9 7.77 5.80 2.13 3.24 2.78 4.08
Ce 27.7 9.01 13.8 9.67 25.9 17.6 19.2 7.03 9.45 7.25 11.2
Pr 3.25 1.29 2.31 1.39 3.04 2.35 3.24 1.25 1.50 0.95 1.42
Nd 13.5 6.38 12.4 7.00 12.7 10.3 17.7 7.20 8.21 4.02 6.00
Sm 2.95 1.84 3.88 2.07 2.69 2.82 5.99 2.47 2.83 0.78 1.12
Eu 1.00 0.66 1.43 0.77 0.91 1.04 2.18 1.02 1.08 0.26 0.38
Gd 3.28 2.26 5.31 2.64 2.86 3.21 7.57 3.46 3.55 0.94 1.22
Dy 3.49 2.79 6.28 3.03 3.18 3.79 9.33 4.36 4.39 1.10 1.50
Er 2.06 1.68 3.99 1.89 1.94 2.17 5.67 2.63 2.69 0.66 0.84
Yb 2.09 1.74 4.02 1.91 1.97 2.17 5.61 2.62 2.75 0.70 0.94
Lu 0.33 0.26 0.61 0.27 0.30 0.34 0.82 0.38 0.41 0.10 0.13
Hf 1.80 1.17 2.81 1.30 1.69 2.08 4.56 1.66 1.96 0.49 0.70
Ta 1.39 0.25 0.17 0.28 1.28 0.64 0.27 0.10 0.16 0.03 0.03
Pb 0.90 0.56 0.20 0.40 0.64 0.33 1.32 0.23 0.32 0.24 0.37
Th 1.53 0.40 0.32 0.40 1.41 1.15 0.25 0.09 0.15 0.13 0.18
U 0.29 0.19 0.21 0.31 0.35 0.22 0.12 0.04 0.05 0.05 0.13
Major elements, Ga, Co, Cu, and Zn determined by XRF. All other trace elements determined by ICP-MS
b.d. below detection limit, Sample types: E-M E-MORB, N-M N-MORB, SSZ supra-subduction zone
Contrib Mineral Petrol (2009) 157:23–40 27
123
Page 6
magmatic series or geographic location can be observed.
By contrast, the good correlation of most of the elements
with Zr suggests that common processes controlled the
compositions of the different magmatic series. For this
reason, the geochemical features of the analyzed samples
are presented according to the three distinct groups of
mafic rocks recognized on chemical bases: (a) normal mid-
ocean ridge basalt (N-MORB)-type; (b) enriched mid-
ocean ridge (E-MORB)-type; and (c) supra-subduction
zone (SSZ)-type. The major geochemical differences
between these groups lie in the different concentrations of
immobile trace elements such as rare earth elements (REE)
and high field strength elements (HFSE).
Normal mid-ocean ridge-type basalts and basaltic
andesites
Four of the eleven samples (A10, M30, S5, and S6) have
compositions comparable to normal mid-ocean ridge
basalts (N-MORB) and basaltic andesites. The MgO con-
tents (3.75–7.71 wt%) and Mg# [100 9 Mg/(Mg + Fe)]
(38.0–56.3) of these samples indicate a slightly differenti-
ated to differentiated nature. The TiO2 contents range from
1.20 to 2.22 wt% (Fig. 2b), comparable to the high-Ti
basalts and basaltic andesites found in the Agoriani Mel-
ange in the northwestern part of the Othris Ophiolite
(Photiades et al. 2003). The Ti/V ratios range from 26 to 31
(Fig. 4) and plot in the field for MORB (Shervais 1982).
The basalts display low to moderate Cr (36–332 ppm) and
Ni (16–61 ppm) contents (Fig. 2e, f). REE patterns are
consistent with N-MORB compositions (Fig. 5a), as they
have moderate depletions in light REE (LREE; (La/
Sm)N = 0.54–0.76) and flat heavy REE (HREE) patterns
(YbN = 12.5–26.8). The samples show an overall deple-
tion in incompatible trace elements, do not show any HFSE
anomalies, and show both positive and negative Sr ano-
malies (Fig. 6a), indicative of plagioclase accumulation
and fractionation, respectively.
SiO [wt%]2
Mg#
3040506070809044
46
48
50
52
54
56
N-MORB-typeE-MORB-typeSSZ-typeMid-Atlantic Ridge
TiO [wt%]2
Mg#
304050607080900
1
2
3
CaO [wt%]
Mg#
30405060708090
4
8
12
16 FeO* [wt%]
Mg#
304050607080906
8
10
12
14
16
18
Cr [ppm]
Mg#
304050607080900
500
1000
1500
2000
2500Ni [ppm]
Mg#
3040506070809010
100
1000
(a) (b)
(c) (d)
(f)(e)
Fig. 2 Variation diagram of
selected major and trace
elements versus Mg#
[100 9 Mg/(Mg + Fe)] for
whole rock compositions of
mafic rocks from Othris,
Greece, compared to mid-ocean
ridge basalts (MORB) from the
Mid-Atlantic ridge (55�S–52�N;
data source: PetDB database).
Major element concentrations
are recalculated to 100% on
LOI-free bases
28 Contrib Mineral Petrol (2009) 157:23–40
123
Page 7
Enriched mid-ocean ridge-type basalts
Five samples (A1, A9, A11, A32, and M3) have composi-
tions comparable to enriched mid-ocean ridge basalts
(E-MORB). The MgO contents (5.35–7.47 wt%) and Mg#
(55.9–60.1) of these samples indicate a slightly differenti-
ated nature. The TiO2 contents range from 0.69 to 1.09 wt%
(Fig. 2b), comparable to the high-Ti basalts and basaltic
andesites of Photiades et al. (2003). The Ti/V ratios range
from 18 to 24 (Fig. 4) and plot at the border between the
fields for convergent margin basalts and MORB (Shervais
1982). The basalts display moderate to high Cr (146–
698 ppm) and Ni (73–227 ppm) contents (Fig. 2e, f). REE
patterns are consistent with E-MORB compositions
(Fig. 5b), as they have enrichments in LREE [(La/
Sm)N = 1.2–3.0] and flat HREE patterns (YbN = 8.3–
10.4). The samples show an overall enrichment in incom-
patible trace elements, do not show any negative HFSE
anomalies, and show no or only small Sr anomalies (Fig. 6b).
MgO [wt%]
Zr [ppm]
0 50 100 150 200 2500
5
10
15
FeO* [wt%]
Zr [ppm]
0 50 100 150 200 2506
7
8
9
10
11
12
13
14
Y [ppm]
Zr [ppm]
0 50 100 150 200 2500
10
20
30
40
50
60
TiO [wt%]2
Zr [ppm]
0 50 100 150 200 2500
1
2
3
Cr [ppm]
Zr [ppm]
0 50 100 150 200 2500
500
1000
1500
2000
2500
La [ppm]
Zr [ppm]
0 50 100 150 200 2500
5
10
15
20
25
SSZ-typeMid-Atlantic Ridge
E-MORB-typeN-MORB-type
enrichedsource
depletedsource
(a) (b)
(c) (d)
(e) (f)
Fig. 3 Variation diagram of
selected major and trace
elements versus Zr (in ppm) for
whole rock compositions of
mafic rocks from Othris,
Greece, compared to mid-ocean
ridge basalts (MORB) from the
Mid-Atlantic ridge (55�S–52�N;
data source: PetDB database).
Major element concentrations
are recalculated to 100% on
LOI-free bases
bulk rock
Ti/1000 [ppm]
V [p
pm]
00 2 4 6 8 10 12 14 16
100
200
300
400
500
600
N-MORB-typeE-MORB-typeSSZ-typeMid-Atlantic Ridge
Ti/V = 10
20
50
Fig. 4 Vanadium versus titanium discrimination diagram for mafic
rocks from Othris compared to MORB from the Mid-Atlantic
ridge(55�S–52�N; data source: PetDB database). Modified after
Shervais (1982). Ti/V \ 20: SSZ basalts; 20 \ Ti/V \ 50: MORB;
Ti/V [ 50: within-plate basalts
Contrib Mineral Petrol (2009) 157:23–40 29
123
Page 8
Supra-subduction zone-type cumulates
The two highly phyric samples (S7 and S8) have much
higher MgO (26.7–30.8 wt%) and lower SiO2 (40.5–
41.7 wt%) contents than the MORB-type samples
(Fig. 2a). The very high MgO, Cr (1,825–2,039 ppm),
and Ni (1,413–1,662 ppm) contents (Fig. 2e, f) and high
Mg# (82.7–85.1) of these samples are consistent with an
accumulation of olivine and pyroxene phenocrysts. The
TiO2 concentrations are very low (0.26–0.33 wt%,
Fig. 2b), comparable to the very low-Ti basaltic andesites
and andesites of Photiades et al. (2003). The Ti/V ratios
range from 13 to 15 (Fig. 4) and plot in the field for
convergent margin basalts (Shervais 1982). The cumu-
lates display flat HREE patterns with low HREE
concentrations (YbN = 3.4–4.5) and enrichments in
LREE [(La/Sm)N = 2.2–2.3] and other incompatible
trace elements (Figs. 5c, 6c). The samples show strong
negative Nb and Ta anomalies (Nb/La = 0.12). These
characteristics are typical for magmas generated in SSZ
settings.
These rocks have compositions similar to the komatiitic
rocks from the Agrilia Formation described by Cameron
and Nisbet (1982), Paraskevopoulos and Economou
(1986), Economou-Eliopoulos and Paraskevopoulos
(1989), and Paraskevopoulos and Economou-Eliopoulos
(1997).
N-MORB-typeanhydrous near-fractional melting
sam
ple
/N-M
OR
B
1
10
A10M30S5S6
E-MORB-typeanhydrous near-fractional melting
sam
ple
/N-M
OR
B
1
10
A1A9A11A32M3
SSZ-typefluid-induced hydrous melting
La Ce Pr Sr Nd Sm Zr Eu Ti Gd Dy Y YbEr
sam
ple
/N-M
OR
B
0.1
1
10
S7S8
16% accumulated melt(4% garnet, 12% spinel)
7% accumulated melt(4% garnet, 3% spinel)no crystal fractionation
30% olivine addition
(a)
(b)
(c)
0% crystal fractionation
50%
20% accumulated melt(8% garnet, 12% spinel)no crystal fractionation
30%
10% accumulated melt(spinel stability field)
Fig. 5 N-MORB-normalized whole rock trace element diagram for
mafic rocks from Othris. a Normal mid-ocean ridge-type basalts and
basaltic andesites, b enriched mid-ocean ridge-type basalts, c supra-
subduction zone-type cumulates. Also shown are the results obtained
for different melting models (see text for details). Element abun-
dances are normalized to the N-MORB values of Sun and
McDonough (1989)
N-MORB-type
sam
ple
/prim
itive
man
tle
10
100
A10M30S5S6N-MORB
E-MORB-type
sam
ple
/prim
itive
man
tle1
10
100
A1A9A11A32M3E-MORB
SSZ-type
Th U Nb Ta La Ce Pr Sr NdSm Zr Hf Eu Ti Gd Tb Dy Ho Er TmY Yb Lu
sam
ple
/prim
itive
man
tle
1
10
S7S8
(a)
(b)
(c)
Sr Ti
Ti
Ti
Fig. 6 Mantle-normalized whole rock trace element diagrams for
mafic rocks from Othris. a Normal mid-ocean ridge-type basalts and
basaltic andesites, b enriched mid-ocean ridge-type basalts, c supra-
subduction zone-type cumulates. Also shown are the N-MORB and
E-MORB values of Sun and McDonough (1989). Element abun-
dances are normalized to the primitive mantle values of McDonough
and Sun (1995)
30 Contrib Mineral Petrol (2009) 157:23–40
123
Page 9
Mineral chemistry
The major element compositions of plagioclase, clinopy-
roxene, orthopyroxene, spinel, and olivine are presented in
Tables 3, 4, 5, 6, 7.
Plagioclase
Plagioclase has anorthite contents of 20–52% in the
N-MORB-type basalt S5 and of up to 68% in the SSZ-type
cumulate S8. Plagioclase in the E-MORB-type basalts was
completely albitized. Considering that in almost all sam-
ples plagioclase is heavily altered, the measured anorthite
contents may not reflect the original igneous compositions.
Pyroxene
The clinopyroxenes in the mafic rocks have compositions
ranging from aluminian diopsides to Mg-rich augites.
Clinopyroxenes are heterogeneous within samples and are
variable between samples, varying between Wo43En51Fs6
and Wo42En33Fs25. Cr2O3 concentrations are positively and
Al2O3 and TiO2 concentrations are negatively correlated
with Mg#. Na2O does not show a clear correlation with
Mg#. Clinopyroxenes in the three groups of mafic rocks
mirror the whole rock compositions, i.e., the MORB-type
and SSZ-type samples plot in the MORB field and con-
vergent margin field of Beccaluva et al. (1989), respectively
(Fig. 7). At a given Mg# clinopyroxenes in N-MORB-type
rocks have comparatively high Na2O and Cr2O3 concen-
trations coupled with moderate Al2O3 and TiO2
concentrations. Clinopyroxenes in E-MORB-type rocks
have higher Al2O3 and TiO2 concentrations coupled with
moderate Na2O and Cr2O3 concentrations. Clinopyroxenes
in the SSZ-type rocks have cores with very high Mg#, high
Cr2O3, and very low TiO2 and Al2O3 concentrations. Rims
have more evolved compositions similar to clinopyroxenes
in N-MORB-type rocks.
Orthopyroxene was found only in the SSZ-type sample
S8. The orthopyroxenes analyzed are enstatites (Mg# = 79)
with low Cr2O3 concentrations (0.03–0.05 wt%).
Spinel
Spinel was observed in three E-MORB-type samples (A1,
A9, and A11) and in the two SSZ-type samples (S7 and
S8). Spinel composition is slightly heterogeneous within
samples and is variable between samples (Fig. 8). Spinels
in E-MORB-type rocks have Cr# [100*Cr/(Cr + Al)]
ranging from 54 to 77 and Mg# from 59 to 69. These
compositions plot at the Cr-rich end of the MORB field and
in the island arc tholeiite field of Barnes and Roeder
(2001). Spinels from the SSZ-type samples have higher
Cr# (78–89) and lower Mg# (38–60) than spinels from the
MORB-type rocks. These very Cr-rich spinels are compa-
rable to spinels in boninites (Barnes and Roeder 2001). All
samples have low TiO2 contents (0.17–0.66 wt%).
Olivine
Primary olivine has been preserved only in the SSZ-type
samples S7 and S8 (Fig. 9). Olivine in the studied rocks
have high forsterite contents ranging from 88.9 to 94.0,
relatively high NiO contents from 0.24 to 0.47 wt%, and
Cr2O3 from 0.03 to 0.18 wt%.
Discussion
The new mineral major element and whole rock major and
trace element data obtained for the mafic rocks from Othris
suggest that these rocks were generated from different
mantle sources and formed in distinct tectonic settings. The
main objectives of this discussion are to quantify the
variations in the degree of partial melting, to evaluate the
effects of crystal fractionation, to determine the different
mantle sources of the crustal section of the Othris Ophio-
lite, and to constrain the tectonic setting of the Othris
ophiolite.
Trace element modeling
Trace element abundances were used to evaluate if the
compositional spectrum of the MORB-type rocks can be
explained by anhydrous partial melting both in the spinel
and garnet stability field and subsequent crystal
Table 3 Major element composition of plagioclase determined by
electron microprobe
Sample S5 S8
Type N-MORB SSZ
Spots n = 10 1r n = 13 1r
SiO2 62.43 3.65 52.49 1.96
Al2O3 23.16 2.06 29.70 1.48
Fe2O3 0.69 0.25 1.15 0.22
MgO 0.07 0.07 0.43 0.54
CaO 3.92 2.71 12.48 1.81
BaO 0.03 0.02 0.04 0.03
Na2O 9.37 1.64 4.30 0.88
K2O 0.20 0.20 0.07 0.04
Total 99.86 100.65
An# 18.9 13.3 61.5 8.5
Sample locations: S Sipetorrema Pillow Lava unit
Contrib Mineral Petrol (2009) 157:23–40 31
123
Page 10
fractionation events. In addition, a fluid-induced referti-
lization-hydrous melting model was assessed for the SSZ-
type rocks. An incremental, non-modal batch melting
model was used, in which melt was extracted at 0.1%
increments. Initial composition, source mineralogy, melt-
ing phase proportions, and partition coefficients are given
in Table 8. Details of these models have been described in
Barth et al. (2003). Rayleigh fractionation of a solid
assemblage of 60% olivine + 38% clinopyroxene + 2%
spinel was used to model crystal fractionation. This sim-
plified fractionation model does not include plagioclase,
which is observed in the Othris samples as phenocrysts.
However, since plagioclase has low partition coefficients
for Cr and Y (Table 8) its inclusion in the fractionation
assemblage would not alter the results significantly (see
below). Moreover, this simple melting and fractionation
model correctly reproduces the compositional trends of
MORB from the mid-Atlantic ridge (Fig. 10).
The extent of partial melting and crystal fractionation
was calculated based on the concentration of Cr, Y and
HREE (Figs. 5, 10); the mantle source was evaluated based
on the concentrations of Th, Nb, Ta, and LREE as these
elements are judged to be relatively immobile during
hydrothermal alteration (Staudigel 2003).
Table 4 Major element composition of clinopyroxene determined by electron microprobe
Sample A1 A9 A10 A11 A32
Type E-MORB E-MORB N-MORB E-MORB E-MORB
Spots n = 21 1r n = 28 1r n = 24 1r n = 43 1r n = 15 1r
SiO2 50.09 0.62 49.82 1.07 49.95 0.94 50.87 1.07 49.37 0.66
TiO2 1.06 0.20 0.73 0.23 1.37 0.28 0.49 0.31 0.99 0.15
Al2O3 5.92 0.44 4.84 1.10 4.01 0.84 3.49 1.18 5.77 0.77
Cr2O3 0.47 0.32 0.55 0.31 0.07 0.03 0.71 0.35 0.51 0.28
FeO 5.45 0.62 5.44 0.48 10.03 0.49 5.05 1.36 5.38 0.61
MnO 0.14 0.03 0.14 0.03 0.25 0.04 0.14 0.04 0.15 0.03
MgO 16.23 0.43 16.00 0.89 13.21 0.46 16.44 1.05 16.25 0.76
CaO 19.95 0.58 20.88 0.61 20.64 0.46 21.22 0.74 19.97 0.67
NiO 0.03 0.02 0.03 0.02 b.d. 0.02 0.02 0.03 0.02
V2O3 0.09 0.03 0.06 0.02 0.10 0.03 0.05 0.03 0.10 0.03
Na2O 0.22 0.04 0.19 0.03 0.40 0.04 0.20 0.06 0.20 0.03
Total 99.69 98.72 100.07 98.72 98.75
Mg# 84.2 1.7 84.0 1.5 70.1 1.2 85.3 4.2 84.3 1.6
Sample M3 S5 S6 S7 S8
Type E-MORB N-MORB N-MORB SSZ SSZ
Spots n = 32 1r n = 24 1r n = 26 1r n = 30 1r n = 21 1r
SiO2 50.46 0.91 50.19 1.23 50.86 1.13 50.61 3.16 52.49 0.91
TiO2 0.84 0.23 0.74 0.29 0.82 0.38 0.63 0.48 0.23 0.11
Al2O3 4.72 0.88 4.17 1.08 3.36 1.06 4.29 2.74 1.92 0.65
Cr2O3 0.29 0.20 0.56 0.46 0.33 0.16 0.42 0.41 0.72 0.26
FeO 5.67 0.74 6.78 1.68 7.18 2.15 6.55 2.53 4.63 0.87
MnO 0.15 0.03 0.18 0.05 0.18 0.06 0.16 0.05 0.13 0.03
MgO 16.09 0.58 16.52 1.18 15.34 1.32 15.77 2.21 17.62 0.70
CaO 21.02 0.69 19.03 1.10 20.88 0.49 21.03 0.68 21.03 0.36
NiO 0.03 0.02 0.02 0.01 0.01 0.01 0.03 0.02 0.04 0.02
V2O3 0.07 0.03 0.07 0.03 0.06 0.03 0.03 0.03 0.02 0.02
Na2O 0.20 0.04 0.28 0.06 0.33 0.09 0.26 0.05 0.23 0.03
Total 99.57 98.58 99.39 99.82 99.09
Mg# 83.5 2.0 81.3 4.3 79.2 6.3 81.1 8.1 87.1 2.5
CoO, ZnO, and K2O were measured but below the detection limit in all samples
b.d. below detection limit. Sample locations: A Agrilia Formation, M Mirna Group, S Sipetorrema Pillow Lava Unit
32 Contrib Mineral Petrol (2009) 157:23–40
123
Page 11
Low-pressure fractional crystallization of the MORB-type
rocks
The different mantle sources and different degrees of
melting of the N-MORB-type and E-MORB-type rocks are
reflected by the lack of correlation between Zr and highly
incompatible trace elements such as Nb and La (Fig. 3f). In
contrast, the good correlation between Zr and many major
elements, compatible and moderately incompatible trace
elements (Fig. 3) points to fractional crystallization as the
main process that controlled the evolution of the MORB-
type rocks. Petrographical observations demonstrate that
these rocks have experienced fractional crystallization of
Cr-spinel, clinopyroxene, and plagioclase, most likely
preceded by olivine fractionation. This crystallization
sequence is also confirmed by the decrease of MgO, Ni,
and Cr coupled with the increase of FeO with increasing Zr
(Fig. 3; Table 2). As typical for tholeiitic basalts, the
MORB-type rocks do not show an increase of SiO2 with
increasing fractionation.
The co-variation of Ni and Cr is plotted in Fig. 11, since
these elements are mainly distributed within the early mafic
minerals (i.e., olivine, Cr-spinel and clinopyroxene). The
general trend observed in this figure argues for the crys-
tallization of olivine, clinopyroxene, and possibly small
amounts of Cr-spinel, as deduced from the sharp decrease
of both Ni and Cr in the more fractionated rocks.
Clinopyroxene and plagioclase probably started crys-
tallizing almost simultaneously, given that the mafic rocks
display no or only small Eu anomalies (Fig. 5). Negative Sr
anomalies coupled with small negative Eu anomalies point
to plagioclase fractionation in several samples (e.g., sample
A10), whereas positive Sr anomalies in samples A9 and S5
Table 5 Major element composition of orthopyroxene determined by
electron microprobe
Sample S8
Type SSZ
Spots n = 2 1r
SiO2 54.35 0.66
TiO2 0.43 0.06
Al2O3 1.64 0.05
Cr2O3 0.04 0.02
FeO 13.28 0.11
MnO 0.42 0.02
MgO 28.51 0.12
CaO 1.64 0.20
NiO 0.05 0.02
V2O3 0.02 0.01
Na2O 0.05 0.01
Total 100.45
Mg# 79.3 0.2
CoO, ZnO, and K2O were measured but below the detection limit in
all samples
Sample locations: S Sipetorrema Pillow Lava Unit
Table 6 Major element composition of spinel determined by electron microprobe
Sample A1 A9 A11 S7 S8
Type E-MORB E-MORB E-MORB SSZ SSZ
Spots n = 2 1r n = 7 1r n = 3 1r n = 6 1r n = 6 1r
SiO2 0.07 0.04 0.07 0.03 0.15 0.07 0.09 0.07 0.05 0.03
TiO2 0.60 0.07 0.52 0.11 0.49 0.09 0.35 0.14 8.35 16.10
Al2O3 22.85 0.64 14.78 2.11 15.52 1.78 6.84 1.16 6.15 3.54
Cr2O3 42.26 1.17 49.41 2.99 48.53 1.59 55.76 3.53 37.23 25.74
Fe2O3 7.07 0.44 9.00 0.98 9.22 0.38 10.52 1.98 20.16 14.86
FeO 12.37 0.10 13.80 1.02 13.69 1.66 17.82 1.20 20.70 5.71
MnO 0.17 0.01 0.18 0.03 0.22 0.02 0.27 0.02 0.28 0.04
MgO 14.99 0.01 13.22 0.84 13.40 1.29 9.77 0.84 8.02 3.77
ZnO 0.07 0.04 0.06 0.03 0.08 0.04 0.11 0.03 0.08 0.05
NiO 0.14 0.01 0.12 0.04 0.11 0.06 0.10 0.03 0.15 0.09
V2O3 0.23 0.02 0.15 0.03 0.13 0.02 0.08 0.04 0.34 0.44
CoO 0.04 0.01 0.04 0.02 0.03 0.02 0.05 0.02 0.06 0.02
CaO 0.08 0.01 0.10 0.05 0.04 0.03 b.d. b.d.
Total 100.97 101.46 101.62 101.78 101.60
Mg# 68.4 0.2 63.0 3.2 63.5 5.0 49.4 3.8 40.4 18.2
Cr# 55.4 1.4 69.2 4.3 67.8 3.2 84.5 3.0 70.6 18.1
Fe3+ was calculated from the cation sums. Na2O and K2O were measured but below the detection limit in all samples
b.d. below detection limit. Sample locations: A Agrilia Formation, S Sipetorrema Pillow Lava Unit
Contrib Mineral Petrol (2009) 157:23–40 33
123
Page 12
suggest plagioclase accumulation (Fig. 6). The crystalli-
zation of Fe–Ti oxides was most likely not significant,
because the Ti content increases continuously in with
increasing Zr (Fig. 3b).
N-MORB-type basalts and basaltic andesites
Barth et al. (2003) showed that the clinopyroxene com-
position of MOR-type peridotites from the Fournos Kaıtsa
sub-massif of the Othris peridotite massif can be explained
by moderate degrees (*15%) of anhydrous partial melting
of depleted mantle. These authors proposed a multistage
melting model of *4% melting in the garnet stability field
followed by *12% melting in the spinel stability field.
The superchondritic MREE/HREE ratios of the N-
MORB-type basalts point to an episode of partial melting
in the garnet stability field. MREE to HREE patterns are
steeper than the REE patterns produced by single-stage
melting in the spinel stability field. The multistage melting
model of Barth et al. (2003) followed by 30–50% crystal
fractionation reproduces the concentrations of the moder-
ately incompatible trace elements in the basaltic samples
(Fig. 5a). For samples S5 and S6 the calculated and the
measured Cr contents agree reasonably well. However,
sample A10 has much lower Cr contents than the model.
The basaltic andesite M30 requires more than 50% frac-
tionation, which is beyond the scope of the model. The
calculated LREE abundances are lower than the measured
abundances, suggesting a mantle source that is slightly less
depleted than the MORB source of Barth et al. (2003).
E-MORB-type basalts
The E-MORB-type basalts have, on average, higher Mg#,
Cr, Ni, and LREE and lower HREE than the N-MORB-type
basalts, implying a lower degree of crystal fractionation,
lower degrees of partial melting, a more enriched source,
and/or a higher proportion of melting in the garnet stability
field. Samples A1, A32, and M3 can be modeled by *4%
melting in the garnet stability field followed by *3%
Table 7 Major element composition of olivine determined by elec-
tron microprobe
Sample S7 S8
Type SSZ SSZ
Spots n = 14 1r n = 16 1r
SiO2 41.10 0.26 41.02 0.37
Al2O3 0.03 0.01 0.03 0.02
Cr2O3 0.08 0.02 0.08 0.03
FeO 9.29 1.48 8.64 1.63
MnO 0.16 0.03 0.15 0.04
NiO 0.35 0.06 0.38 0.06
MgO 49.78 1.10 50.32 1.25
CaO 0.18 0.06 0.15 0.06
CoO 0.03 0.01 0.03 0.02
Total 101.05 100.84
Fo 90.5 1.5 91.2 1.7
TiO2, V2O3, ZnO, Na2O, and K2O were measured but below the
detection limit in all samples
Sample locations: S Sipetorrema Pillow Lava Unit
MO
RB
IAT
BON
MO
RB
IAT
BON
MORB
IATBON
clinopyroxeneMg# > 80N-MORB-type
S5S6
clinopyroxeneMg# > 80E-MORB-type
Na O20 25 50 75 100
TiO2
0
25
50
75
SiO /1002
25
50
75
100
A1A9A11A32M3
clinopyroxeneMg# > 80SSZ-type
S7S8
Na O20 25 50 75 100
TiO2
0
25
50
75
SiO /1002
25
50
75
100
Na O20 25 50 75 100
TiO2
0
25
50
75
SiO /1002
25
50
75
100
(a)
(b)
(c)
Fig. 7 TiO2-Na2O-SiO2/100 (wt%) discrimination diagram for clin-
opyroxenes from mafic rocks from Othris. a Normal mid-ocean ridge-
type basalts and basaltic andesites, b enriched mid-ocean ridge-type
basalts, c supra-subduction zone-type cumulates. For clarity, only
clinopyroxenes with Mg#[80 are shown. Abbreviations: MORB mid-
ocean ridge basalt, IAT island-arc tholeiites, BON boninites. Modified
after Beccaluva et al. (1989)
34 Contrib Mineral Petrol (2009) 157:23–40
123
Page 13
melting in the spinel stability field (Fig. 5b). Samples A9
and A11 have lower REE contents and lower LREE/HREE
ratios than samples A1, A32, and M3, indicating higher
degrees of melting. The trace element contents of samples
A9 and A11 can be reproduced by *8% melting in the
garnet stability field followed by *12% melting in the
spinel stability field. The measured Cr contents suggest
10–50% crystal fractionation. However, the low HREE con-
tents suggest lower degrees of fractionation (\20%). Since
the calculated LREE abundances are considerably lower than
the measured abundances, an enriched mantle source is
required to explain the observed trace element abundances.
SSZ-type cumulates
The very high MgO, Cr, and Ni contents and high Mg# of
samples S7 and S8 suggest a cumulate origin. Bickle (1982)
discussed an approach to the estimate the original melt
composition in the case of lavas that accumulated pheno-
crysts. Based on experimental data of Fe–Mg partitioning
between olivine and melt, Bickle (1982) proposed a distri-
bution coefficient KD = 0.314 for Mg-rich lavas and
forsteritic olivine. The most magnesian olivines (Fo94) in
samples S7 and S8 would be in equilibrium with melts
containing *30 wt% MgO (calculated on a volatile-free
basis). This is a much higher MgO content than inferred for
the melt composition of komatiitic rocks from the Agrilia
Formation with broadly similar bulk composition (12–
17 wt% MgO; Cameron and Nisbet 1982; Economou-Elio-
poulos and Parakevopoulos 1989). Considering the low SiO2
(46–47 wt%) and Al2O3 (6 wt%) contents of the calculated
melt, we do not judge the calculated melt composition to be
trustworthy. A likely source of error is the forsterite content
of olivine, because the maximum forsterite content of olivine
samples S7 and S8 is higher than that in olivines in komatiitic
rocks from the Agrilia Formation (Fo90–92; Cameron and
Nisbet 1982; Economou-Eliopoulos and Parakevopoulos
1989) and in olivines in peridotites from the Othris Ophiolite
(Fo90–92; eTable 5 and Barth et al. 2003). Python et al. (2007)
argued that the forsterite content of olivine can increase
during high temperature hydrothermal alteration. Using the
average olivine composition (Fo91) of samples S7 and S8, we
calculate a melt composition of 20–22 wt% MgO, 6–7 wt%
CaO, 12 wt% FeO, 49 wt% SiO2, and 8–9 wt% Al2O3. The
calculated melt has MgO, CaO, SiO2, and Al2O3 concen-
trations broadly similar to but considerably higher FeO
concentrations than the experimental boninitic liquids of
Falloon and Danyushevsky (2000).
The fluid-induced refertilization-hydrous melting model
for SSZ-type rocks assumes a mantle source that has been
depleted by 9% anhydrous near-fractional melting in the
spinel stability field. The concentrations of moderately
incompatible trace elements of samples S7 and S8 can be
modeled by *10% hydrous melting induced by addition of
a fluid with the maximum trace element content of Bizimis
et al. (2000), followed by 30–50% olivine addition
(Fig. 5c). However, the modeled concentrations of highly
incompatible trace elements are considerably lower than
the measured concentrations. The most likely reasons for
this discrepancy are (a) the mantle source was less depleted
in highly incompatible trace elements than the model
source; (b) the fluid component had higher concentrations
of highly incompatible trace elements than the model
composition of Bizimis et al. (2000); and/or (c) mixing of a
depleted SSZ-type magma and a more enriched melt.
Geological implications
In agreement with the earlier studies of Pearce et al. (1984)
and Photiades et al. (2003) our melt modeling suggests that
spinel
Mg#304050607080
Cr#
0
20
40
60
80
100
A1A9A11S7S8Othris peridotites
E-MORB-type
SSZ-type
}}
Fig. 8 Diagram of Cr# [100 9 Cr/(Cr + Al)] versus Mg# in spinel
from mafic rocks from Othris compared to spinel from Othris
peridotites (eTable 4 and Barth et al. 2003)
olivine
Fo88 89 90 91 92 93 94 95
NiO
[wt%
]
0.20
0.25
0.30
0.35
0.40
0.45
0.50
S7S8Othris peridotites
SSZ-type}
Fig. 9 Plot of forsterite content versus NiO [in wt%] in olivine from
mafic rocks from Othris compared to olivine from Othris peridotites
(eTable 5 and Barth et al. 2003)
Contrib Mineral Petrol (2009) 157:23–40 35
123
Page 14
the crustal section of Othris Ophiolite records two distinct
melting regimes, i.e., anhydrous MOR-type and hydrous
SSZ-type melting. The occurrence of both N- and
E-MORB-type lavas suggests that the mantle generating
the lavas of the Othris Ophiolite must have been hetero-
geneous on a comparatively fine scale. The E-MORB-type
magmas were erupted without significant mixing with N-
MORB magmas. This observation is consistent with an
origin in a slow-spreading system, since these heteroge-
neities may not have survived in a fast-spreading system
that is underlain by a steady-state melt lens (e.g., Sinton
and Detrick 1992). Thus, the N-MORB and E-MORB
magmatism of the Othris Ophiolite might have been
comparable to the N- to E-MORB magmatism observed on
Macquarie Island, which is the result of a waning spreading
system (Varne et al. 2000).
The N-MORB-type basalts from the Agrilia Formation
and the Sipetorrema Pillow Lava unit are consistent with
about *15% anhydrous partial melting of a depleted
MORB source, in excellent agreement with the degrees of
partial melting inferred for the Othris MOR-type perido-
tites (Barth et al. 2003, 2008). The MOR-type peridotites
from the Fournos Kaıtsa sub-massif and the western
Katachloron sub-massif are moderately depleted and simi-
lar to abyssal peridotites.
The E-MORB-type basalts from the Agrilia Formation
and the Mirna Group record between 7 and 20% anhydrous
partial melting of an enriched mantle source. The minimum
degree of melting inferred for the MOR-type peridotites
(13%; Barth et al. 2008) is not as low as the minimum
degree of melting inferred for the E-MORB-type basalts.
Furthermore, Barth et al. (2008) did not deduce an enriched
mantle source for the MOR-type peridotites. This implies
that the complementary residue of the E-MORB-type
basalts is not exposed or has not been sampled in the Othris
peridotite massif, possibly because such comparatively
undepleted peridotites would probably occur at deeper
stratigraphic levels that are not exposed or have not been
emplaced.
The SSZ-type cumulates from the Sipetorrema Pillow
Lava unit are products of hydrous melting of a previously
depleted mantle source. The Othris peridotites from the
Metalleio, Eretria, and eastern Katachloron sub-massif,
which record evidence of SSZ-type mantle, i.e., they are
Table 8 Partition coefficients, initial mineralogy, melting proportions, MORB source and fluid composition in ppm (lg/g)
Olivine opx cpx Spinel Garnet Plag MORB Fluid
Source Maximum
Ti 0.007 0.12 0.3 0.07 0.29 0.06 927 436
Cr 0.7 4 4 70 2 0.1 2630
Ni 10 1.1 2.6 5 5 0.1
Y 0.006 0.07 0.46 0.002 2.8 0.03 3.44
Zr 0.003 0.05 0.12 0.04 0.27 0.003 6.195 365
Sr 0.01 0.04 0.06 0.0006 0.0025 1.6 12.93 1930
La 0.000007 0.0005 0.0536 0.0006 0.0016 0.16 0.161 18.6
Ce 0.00001 0.0009 0.09 0.0006 0.005 0.1 0.5376 48
Pr 0.00003 0.004 0.13 0.0007 0.018 0.09 0.108 5.88
Nd 0.00007 0.009 0.17 0.0008 0.052 0.07 0.7375 22.2
Sm 0.0007 0.02 0.29 0.0008 0.25 0.06 0.304 4.9
Eu 0.001 0.03 0.35 0.0009 0.4 0.06 0.118 2.85
Gd 0.0012 0.04 0.4 0.0009 0.8 0.05 0.417 3.46
Dy 0.004 0.07 0.45 0.0015 2.2 0.03 0.559 0.5
Er 0.01 0.07 0.47 0.003 3.6 0.02 0.381 0.3
Yb 0.005 0.08 0.49 0.0015 6.6 0.01 0.392 0.3
Xa 0.55 0.25 0.18 0.02
Pa dry -0.20 0.43 0.72 0.05
Pa hydrous -0.10 0.52 0.56 0.02
Xa 0.57 0.21 0.16 0.06
Pa dry 0.08 -0.19 0.81 0.30
Fractionation 0.60 0.38 0.02
Partition coefficients from Kelemen et al. (1993), Johnson (1998), and from the GERM database (http://www.earthref.org/GERM/). Xa min-
eralogy of the MORB source (Johnson et al. 1990). Pa dry melting mode for dry melting. Spinel melting modes from Baker and Stolper (1994);
garnet peridotite melt modes from Walter (1998). Pa hydrous melting mode for hydrous melting in an island arc environment from Bizimis et al.
(2000). Composition of the fluid component from Bizimis et al. (2000)
36 Contrib Mineral Petrol (2009) 157:23–40
123
Page 15
highly depleted and are similar to peridotites from the Izu-
Bonin-Mariana forearc (Barth et al. 2008), may be the
complementary residua to the SSZ-type cumulates.
In order to test if the crustal section of the Othris
Ophiolite is genetically linked to the mantle section we
need to compare the melting conditions of the mafic rocks
of the Mirna Group, including, the Sipetorrema Pillow
Lava unit, to the melting conditions of the Othris peridotite
massif as the Othris peridotite massif is part of the Mirna
Group (Smith et al. 1975). The excellent agreement
between the melting conditions inferred from the
N-MORB-type basalts and the MOR-type peridotites as
well as the broadly complementary nature of the boninitic
cumulates and the SSZ-type peridotites suggest that the
crustal section may be genetically related to the mantle
section. Hence, our modeling results support the interpre-
tation that both the crustal section and the mantle section of
the Othris Ophiolite formed in an oceanic environment,
probably in an infant arc or in a forearc (see below).
Accordingly, the Othris Ophiolite can be classified as a
Mediterranean-type ophiolite (Dilek 2003).
Furthermore, our melt modeling implies that the mantle
sources and melting regimes of the crustal section of the
Othris complex evolved from anhydrous MOR-type melting
during the Middle Triassic–Middle Jurassic opening of the
Mirdita-Pindos oceanic basin, as recorded by the Triassic
Agrilia Formation, to both MOR- and fluid-induced SSZ-
type melting during the closure of the Mirdita-Pindos oce-
anic basin by intra-oceanic subduction during the Middle–
Late Jurassic, as recorded by the Jurassic Mirna Group
(including the Sipetorrema Pillow Lava unit). An important
trait of the Sipetorrema Pillow Lava unit is the close spatial
association of MORB-type and SSZ-type rocks. The close
association of different lava suites has also been reported
for other ophiolites in the Mirdita-Subpelagonian Zone (see
review by Robertson 2002). At several localities including
the Pindos ophiolite in the NW of Othris and in the Alba-
nian ophiolites further north lavas ranging from MORB to
island-arc tholeiites (IAT) and boninites are spatially jux-
taposed or interlayered (Capedri et al. 1980; Pearce et al.
1984; Jones and Robertson 1991; Bebien et al. 2000; Bor-
tolotti et al. 2002; Hoeck et al. 2002; Saccani and Photiades
2004). Taken together, these observations indicate that two
or more distinct melting regimes coexisted spatially and/or
temporally in a relatively restricted sector across an intra-
oceanic subduction setting.
The geological and petrological characteristics shown by
the Othris Ophiolite are compatible with the intra-oceanic
thrusting model for the Hellenic-Dinaric ophiolites pro-
posed by Spray et al. (1984), Bebien et al. (2000),
Insergueix-Filippi et al. (2000), and Barth et al. (2008).
According to this model, the Albanide-Hellenide ophiolites
formed by ridge collapse (Fig. 12), i.e., forced initiation of
subduction at or near a mid-ocean ridge (e.g., Boudier and
Coleman 1981; Boudier et al. 1988). The kinematic
numerical models of Insergueix-Filippi et al. (2000)
Y [ppm]
Cr
[ppm
]
101 10 100
100
1000
10000
MORBsource
residue
fractional
crystallization
accumulatedpartial melt
Othris peridotitesMid-Atlantic Ridge
SSZ-type
N-MORB-typeE-MORB-type
bulk rock
4%8%12%
4%8%
20%20%
40%30%
M30
A10
anhydrous near-fractional melting
Fig 10 Cr versus Y diagram for mafic rocks from Othris compared to
Othris peridotites and MORB from the Mid-Atlantic ridge. Also
shown are the results obtained for a melting model including 4% of
near-fractional melting (incremental batch melting at 0.1% incre-
ments) in the garnet stability field followed by near-fractional melting
in the spinel stability field (solid lines). After 4% of partial melting in
the garnet stability field, a new source modal mineralogy was
calculated to account for the subsolidus garnet to spinel phase
transition. MORB source composition, melting and residue paths for
near-fractional melting, and peridotite compositions are from Barth
et al. (2003, 2008). Numbers along the solid lines are percent melting.
Dashed line shows the Rayleigh fractionation trend of a solid
assemblage of 60% olivine + 38% clinopyroxene + 2% spinel (see
text for details). Numbers along the dashed line are percent crystal
fractionation
bulk rock
Cr [ppm]0 500 1000 1500 2000 2500
Ni [
ppm
]
0
500
1000
1500
2000
SSZ-typeMid-Atlantic Ridge
N-MORB-typeE-MORB-type
primarymelts
olivineaddition
olivine
cpx
opxplagfractionation
spinel
Fig. 11 Ni versus Cr diagram for mafic rocks from Othris compared
to MORB from the Mid-Atlantic ridge. Fractional crystallization
trends for olivine, clinopyroxene (cpx), orthopyroxene (opx), spinel,
and plagioclase (plag) based on the partition coefficients in Table 8
are shown. The arrow indicates the effect of olivine addition to a
primary melt using the average olivine composition of samples S7
and S8. Primary melt composition is taken from Liang and Elthon
(1990)
Contrib Mineral Petrol (2009) 157:23–40 37
123
Page 16
demonstrate that preservation of high temperatures in the
mantle wedge explains the occurrence of boninitic mag-
matism at the earliest stages of subduction initiation, which
is partly contemporaneous with the decline of MORB
magmatism and the initiation of arc magmatism. In the case
of localized fluid flux from the slab, hydrous partial melting
of previously depleted peridotites may be contained to
limited areas, explaining the close proximity of SSZ-type
and MOR-type rocks in the Sipetorrema Pillow Lava unit.
As an alternative, Dilek and Flower (2003), Flower and
Dilek (2003), Saccani et al. (2004), and Beccaluva et al.
(2005) recently proposed a tectonic model of western
Pacific-style arc-trench rollback. The main difference
between these two models is an episode of slab rollback
and associated inter-arc or back-arc opening in the arc-
trench rollback model, causing extension in the upper plate,
whereas in the intra-oceanic thrusting model the upper
plate is under compression. As discussed by Barth et al.
(2008), a western Pacific-style tectonic model is unlikely to
be applicable to the Mirdita-Pindos oceanic basin, because
the geodynamic boundary conditions such as the density
differences between subducting and overriding plates are
very different for subduction initiation in the western
Pacific in the Eocene and in the Mirdita-Pindos oceanic
basin, a comparatively short-lived and narrow marginal
basin.
On the other hand, Benoit et al. (1999), Python and
Ceuleneer (2003), and Nonnotte et al. (2005) suggested
that N-MORB-type and depleted hydrous magmas may
coexist in a single tectonic setting. The depleted hydrous
magmas may be generated at a mid-ocean ridge by shallow
re-melting of partly hydrated peridotites residual after
MORB extraction.
Irrespective of the exact tectonic model, the well-pre-
served ophiolite exposures found from Serbia to Greece
along approximately 1,000 km correspond to infant arc
settings, suggesting that large-scale obduction processes
require the creation of new intra-oceanic subduction zones,
whether spontaneous or induced (Stern 2004; Gurnis et al.
2004). Recently, Agard et al. (2007) proposed a mechanism
in which large-scale obduction events are triggered by
intraplate instabilities resulting from sharp plate accelera-
tions, possibly in response to superplume events. These
authors demonstrated that the inception of the Jurassic
obduction, as testified by ages obtained for metamorphic
soles, also developed during a period of high convergence
velocities.
Conclusions
The crustal section of the Othris Ophiolite contains three
geochemically distinct groups of mafic rocks: N-MORB-
type basalts and basaltic andesites, E-MORB-type basalts,
and boninitic cumulates, indicative of both anhydrous
MOR-type and fluid-induced SSZ-type melting regimes.
Starting in the Triassic, a MORB-type oceanic lithosphere
was generated between the Korabi-Pelagonian and the
Apulian microcontinents. The occurrence of both N-
MORB- and E-MORB-type basalts in the Agrilia For-
mation suggests that the MORB source mantle was
heterogeneous. When oceanic extension rapidly changed
to convergence, intra-oceanic subduction started in the
Middle Jurassic and resulted in the almost contempora-
neous production of MORB-type and boninitic magmas,
as observed in the mafic rocks of the Mirna Group. The
excellent agreement between the melting conditions
inferred from the N-MORB-type basalts and the MOR-
type peridotites in Othris as well as the broadly com-
plementary nature of the boninitic cumulates and the
SSZ-type peridotites found in Othris suggest that the
crustal section may be genetically related to the mantle
section.
Acknowledgments We would like to express our gratitude to Arjan
Dijkstra and Gareth Davies for help with field work. This work was
funded in part by ISES grant 6.2.3 Dinaric-Hellenic Ophiolites.
MORB magmatism
Island Arc magmatism
Boninitic magmatism
ascendingasthenospheric
flowMelting zones:
lithosphere formedat a mid-ocean ridge
high-Ti intermediate to low-TiMORB boniniteIAT
lithosphere formedat a mid-ocean ridge
Jurassic: subduction initiation by intra-oceanic thrusting
slivers ofE-MORBsource
depleted mantle(N-MORB source)
Fig. 12 Cartoon (not to scale) showing the genesis of the Othris
Ophiolite by intra-oceanic subduction during the Middle–Late
Jurassic (modified from Insergueix-Filippi et al. 2000, and Barth
et al. 2008). The Mirdita-Pindos oceanic basin opened during the
Middle Triassic–Middle Jurassic. The subduction zone probably
originated at (or close to) the mid-ocean ridge axis of the Mirdita-
Pindos ocean (ridge collapse). The preservation of high temperatures
in the mantle wedge favors the setting of short-lived boninitic
magmatism in the earliest stages of subduction initiation, which are
partly contemporaneous with a progressive extinction of MORB
magmatism and initiation of arc magmatism. Enriched E-MORB
source material may occur as blobs or slivers in the depleted
N-MORB source mantle. The intermediate-Ti magmatism can result
from tapping of less depleted mantle but also from mixing of high-Ti
and low-Ti magmatism due to the close proximity of the respective
melting zones in the mantle wedge
38 Contrib Mineral Petrol (2009) 157:23–40
123
Page 17
Discussions with Dejan Prelevic and comments by two anonymous
reviewers helped to improve the manuscript. Tim Grove is thanked
for his constructive editorial handling.
References
Agard P, Jolivet L, Vrielynck B, Burov E, Monie P (2007) Plate
acceleration: the obduction trigger? Earth Planet Sci Lett
258:428–441
Baker MB, Stolper EM (1994) Determining the composition of high-
pressure mantle melts using diamond aggregates. Geochim
Cosmochim Acta 58:2811–2827
Barnes SJ, Roeder PL (2001) The range of spinel compositions in
terrestrial mafic and ultramafic rocks. J Petrol 42:2279–2302
Barth MG, Mason PRD, Davies GR, Dijkstra AH, Drury MR (2003)
Geochemistry of the Othris Ophiolite, Greece: evidence for
refertilization? J Petrol 44:1759–1785
Barth MG, Mason PRD, Davies GR, Drury MR (2008) The Othris
Ophiolite, Greece: a snapshot of subduction initiation at a mid-
ocean ridge. Lithos 100:234–254
Bebien J, Dimo-Lahitte A, Vergely P, Insergueix-Filippi D, Dupeyrat
L (2000) Albanian ophiolites I—magmatic and metamorphic
processes associated with the initiation of a subduction. Ofioliti
25:39–45
Beccaluva L, Macciotta G, Piccardo GB, Zeda O (1989) Clinopy-
roxene composition of ophiolite basalts as petrogenetic indicator.
Chem Geol 77:165–182
Beccaluva L, Coltorti M, Saccani E, Siena F (2005) Magma
generation and crustal accretion as evidenced by supra-subduc-
tion ophiolites of the Albanide–Hellenide Subpelagonian zone.
Isl Arc 14:551–563
Benoit M, Ceuleneer G, Polve M (1999) The remelting of
hydrothermally altered peridotite at mid-ocean ridges by intrud-
ing mantle diapirs. Nature 402:514–518
Bickle MJ (1982) The magnesium content of komatiitc liquids. In:
Arndt NT, Nisbet EG (eds) Komatiites. Allen & Unwin, London,
pp 479–494
Bizimis M, Salters VJM, Bonatti E (2000) Trace and REE content of
clinopyroxenes from supra-subduction zone peridotites. Impli-
cations for melting and enrichment processes in island arcs.
Chem Geol 165:67–85
Bortolotti V, Marroni M, Pandolfi L, Principi G, Saccani E (2002)
Interaction between mid-ocean ridge and subduction magmatism
in Albanian ophiolites. J Geol 110:561–576
Boudier F, Coleman RG (1981) Cross section through the peridotite
in the Samail ophiolite, southeastern Oman mountains.
J Geophys Res 86:2573–2592
Boudier F, Ceuleneer G, Nicolas A (1988) Shear zones, thrusts and
related magmatism in the Oman ophiolite: Initiation of thrusting
on an ocean ridge. Tectonophysics 151:275–296
Cameron WE, Nisbet EG (1982) Phanerozoic analogues of komatiitic
basalts. In: Arndt NT, Nisbet EG (eds) Komatiites. Allen &
Unwin, London, pp 29–50
Capedri S, Venturelli G, Bocchi G, Dostal J, Garuti G, Rossi A (1980)
The geochemistry and petrogenesis of an ophiolitic sequence
from Pindos, Greece. Contrib Mineral Petrol 74:189–200
Coleman RG (1977) Ophiolites. Springer, Heidelberg, p 229
Dijkstra A, Drury MR, Vissers RLM (2001) Structural petrology of
plagioclase peridotites in the West Othris Mountains (Greece):
melt impregnation in mantle lithosphere. J Petrol 42:5–24
Dijkstra AH, Barth MG, Drury MR, Mason PRD, Vissers RLM
(2003) Diffuse porous melt flow and melt-rock reaction in the
mantle lithosphere at a slow-spreading ridge: A structural
petrology and LA-ICP-MS study of the Othris Peridotite Massif
(Greece). Geochem Geophys Geosys 4:8613. doi:
10.1029/2001GC000278
Dilek Y (2003) Ophiolite concept and its evolution. In: Dilek Y,
Newcomb S (eds) Ophiolite concept and the evolution of
geological thought. Geol Soc Am Special Paper, vol 373.
Boulder, Colorado, pp 1–16
Dilek Y, Flower MFJ (2003) Arc-trench rollback and forearc
accretion: 2. A model template for ophiolites in Albania,
Cyprus, and Oman. In: Dilek Y, Robinson PT (eds) Ophiolites
in Earth history. Geol Soc London Special Publication, vol 208,
pp 43–68
Dilek Y, Shallo M, Furnes H (2005) Rift-drift, seafloor spreading, and
subduction tectonics of Albanian ophiolites. Int Geol Rev
47:147–176
Economou-Eliopoulos M, Parakevopoulos GM (1989) Platinum-
group elements and gold in komatiitic rocks from the Agrilia
formation, Othrys ophiolite complex, Greece. Chem Geol
77:149–158
Falloon TJ, Danyushevsky LV (2000) Melting of refractory mantle at
1.5, 2 and 2.5 GPa under anhydrous and H2O-undersaturated
conditions: implications for the petrogenesis of high-Ca boni-
nites and the influence of subduction components on mantle
melting. J Petrol 41:257–283
Ferriere J (1985) Nature et developpement des ophiolites Helleniques
du secteur Othrys-Pelion. Ofioliti 10:255–278
Flower MFJ, Dilek Y (2003) Arc-trench rollback and forearc
accretion: 1. A collision-induced mantle flow model for Tethyan
ophiolites. In: Dilek Y, Robinson PT (eds) Ophiolites in Earth
history. Geol Soc London Special Publication, vol 218, pp 21–41
Gurnis M, Hall C, Lavier L (2004) Evolving force balance during
incipient subduction. Geochem Geophys Geosys 5, Q07001. doi:
10.1029/2003GC000681
Hoeck V, Koller F, Meisel T, Onuzi K, Kneringer E (2002) The
jurassic south Albanian ophiolites: MOR-vs SSZ-type ophiolites.
Lithos 65:143–164
Hynes A (1974) Notes on the petrology of some ophiolites, Othris
mountains, Greece. Contrib Mineral Petrol 46:233–239
Hynes AJ, Nisbet EG, Smith GA, Welland MJP, Rex DC (1972)
Spreading and emplacement ages of some ophiolites in the
Othris region (eastern central Greece). Z Deutsch Geol Ges
123:455–468
Insergueix-Filippi D, Dupeyrat L, Dimo-Lahitte A, Vergely P, Bebien
J (2000) Albanian ophiolites II—model of subduction zone
infancy at a mid-ocean ridge. Ofioliti 25:47–53
Johnson KTM (1998) Experimental determination of partition
coefficients for rare earth and high-field-strength elements
between clinopyroxene, garnet, and basaltic melt at high
pressures. Contrib Mineral Petrol 133:60–68
Johnson KT, Dick HJB, Shimizu N (1990) Melting in the oceanic
upper mantle: an ion microprobe study of diopsides in abyssal
peridotites. J Geophys Res 95:2661–2678
Jones G, Robertson AHF (1991) Tectono-stratigraphy and evolution
of the Mesozoic Pindos ophiolite and related units, Northwestern
Greece. J Geol Soc London 148:267–288
Kelemen PB, Shimizu N, Dunn T (1993) Relative depletion of
niobium in some arc magmas and the continental crust:
partitioning of K, Nb, La and Ce during melt/rock reaction in
the upper mantle. Earth Planet Sci Lett 120:111–134
Liang Y, Elthon D (1990) Evidence from chromium abundances in
mantle rocks for extraction of picrite and komatiite melts. Nature
343:551–553
McDonough WF, Sun S-S (1995) Composition of the earth. Chem
Geol 120:223–253
Menzies M (1976) Rifting of a Tethyan continent—rare earth
evidence of an accreting plate margin. Earth Planet Sci Lett
28:427–438
Contrib Mineral Petrol (2009) 157:23–40 39
123
Page 18
Menzies M, Allen C (1974) Plagioclase lherzolite-residual mantle
relationships within two eastern Mediterranean ophiolites. Con-
trib Mineral Petrol 45:197–213
Moores EM (1982) Origin and emplacement of ophiolites. Rev
Geophys Space Phys 20:735–760
Nehring F, Jacob DE, Barth MG, Foley SF (2008) Laser-ablation ICP-
MS analysis of siliceous rock glasses fused on an iridium strip
heater using MgO dilution. Microchim Acta 160:153–163
Nicolas A (1989) Structures of ophiolites and dynamics of oceanic
lithosphere. Petrology and structural geology.. Kluwer, Dordr-
echt, p 367
Nonnotte P, Ceuleneer G, Benoit M (2005) Genesis of andesitic-
boninitic magmas at mid-ocean ridges by melting of hydrated
peridotites: geochemical evidence from DSDP Site 334 gab-
bronorites. Earth Planet Sci Lett 236:632–653
Paraskevopoulos GM, Economou MI (1986) Komatiite-type ultra-
mafic lavas from the Agrilia formation, Othrys Ophiolite
complex, Greece. Ofioliti 11:293–304
Paraskevopoulos GM, Economou-Eliopoulos M (1997) Geochemistry
and geotectonic setting of basaltic rocks from the Othrys
ophiolite complex. Ann Geol Pays Hell 37:601–612
Pearce JA, Lippard SJ, Roberts S (1984) Characteristics and tectonic
significance of supra-subduction zone ophiolites. In: Kokelaar
PB, Howells MF (eds) Marginal Basin Geology. Geol Soc
London Special Publication, vol 16, pp 77–94
Pearce NJG, Perkins WT, Westgate JA, Gorton MP, Jackson SE, Neal
CR, Chenery SP (1997) A compilation of new and published
major and trace element data for NIST SRM 610 and NIST SRM
612 glass reference materials. Geostand Newsl 21:115–144
Photiades A, Saccani E, Tassinari R (2003) Petrogenesis and tectonic
setting of volcanic rocks from the subpelagonian ophiolitic
melange in the Agoriani area (Othrys, Greece). Ofioliti 28:121–
135
Python M, Ceuleneer G (2003) Nature and distribution of dykes and
related melt migration structures in the mantle section of the
Oman ophiolite. Geochem Geophys Geosys 4:8612. doi:
10.1029/2002GC000354
Python M, Ceuleneer G, Ishida Y, Barrat J-A, Arai S (2007) Oman
diopsidites: a new lithology diagnostic of very high temperature
hydrothermal circulation in mantle peridotite below oceanic
spreading centres. Earth Planet Sci Lett 255:289–305
Rampone E, Piccardo GB (2000) The ophiolite-oceanic lithosphere
analogue: New insights from the Northern Apennines (Italy). In:
Dilek Y, Moores EM, Elthon D, Nicolas A (eds) Ophiolites and
oceanic crust: New insights from field studies and the Ocean
Drilling Program. Geol Soc Am Special Paper, vol 349. Boulder,
Colorado, pp 21–34
Rassios A, Konstantopoulou G (1993) Emplacement tectonism and
the position of chrome ores in the Mega Isoma peridotites, SW
Othris, Greece. Bull Geol Soc Greece 28:463–474
Rassios A, Smith AG (2000) Constraints on the formation and
emplacement age of western Greek ophiolites (Vourinos, Pindos,
and Othris) inferred from deformation structures in peridotites.
In: Dilek Y, Moores E, Elthon D, Nicolas A (eds) Ophiolites and
oceanic crust: new insights from field studies and the ocean
drilling program. Geol Soc Am Special Paper, vol 349. Boulder,
Colorado, pp 473–484
Robertson AHF (2002) Overview of the genesis and emplacement of
Mesozoic ophiolites in the eastern Mediterranean Tethyan
region. Lithos 65:1–67
Robertson AHF, Karamata S (1994) The role of subduction-accretion
processes in the tectonic evolution of the Mesozoic Tethys in
Serbia. Tectonophysics 234:73–94
Robertson A, Shallo M (2000) Mesozoic-tertiary tectonic evolution of
Albania in its regional Eastern Mediterranean context. Tectono-
physics 316:197–254
Robertson AHF, Clift PD, Degnan PJ, Jones G (1991) Palaeogeo-
graphic and palaeotectonic evolution of the Eastern
Mediterranean Neotethys. Palaeogeogr Palaeoclimatol Palaeo-
ecol 87:289–343
Saccani E, Photiades A (2004) Mid-ocean ridge and supra-subduction
affinities in the Pindos ophiolites (Greece): implications for
magma genesis in a forearc setting. Lithos 73:229–253
Saccani E, Beccaluva L, Coltorti M, Siena F (2004) Petrogenesis and
tectono-magmatic significance of the Albanide-Hellenide Sub-
pelagonian ophiolites. Ofioliti 29:75–93
Shallo M, Dilek Y (2003) Development of the ideas on the origin and
of Albanian ophiolites. In: Dilek Y, Newcomb S (eds) Ophiolite
concept and the evolution of geochemical thought. Geol Soc Am
Special Paper, vol 373. Boulder, Colorado, pp 351–363
Shervais JW (1982) Ti-V plots and the petrogenesis of modern and
ophiolitic lavas. Earth Planet Sci Lett 59:101–118
Sinton JM, Detrick RS (1992) Mid-ocean ridge magma chambers.
J Geophys Res 97:197–216
Smith AG (1993) Tectonic significance of the Hellenic-Dinaric
ophiolites. In: Prichard HM, Alabaster T, Harris NBW, Neary
CR (eds) Magmatic processes and plate tectonics. Geol Soc
London Special Publication, vol 76. Blackwell, Oxford,
pp 213–243
Smith AG, Rassios A (2003) The evolution of ideas for the origin and
emplacement of the western Hellenic ophiolites. In: Dilek Y,
Newcomb S (eds) Ophiolite concept and the evolution of
geological thought. Geol Soc Am Special Publication, vol 373.
Boulder, Colorado, pp 337–350
Smith AG, Spray JG (1984) A half-ridge transform model for the
Hellenic-Dinaric ophiolites. In: Dixon JE, Robertson AHF (eds)
The geological evolution of the Eastern Mediterranean. Geol Soc
London Special Publication, vol 17. Blackwell, Oxford, pp 629–
644
Smith AG, Hynes AJ, Menzies M, Nisbet EG, Price I, Welland MJ,
Ferriere J (1975) The stratigraphy of the Othris Mountains,
eastern central Greece: a deformed Mesozoic continental margin
sequence. Eclogae Geol Helv 68:463–481
Spray JG, Bebien J, Rex DC, Roddick JC (1984) Age constraints on
the igneous and metamorphic evolution of the Hellenic-Dinaric
ophiolites. In: Dixon JE, Robertson AHF (eds) The geological
evolution of the Eastern Mediterranean. Geol Soc London
Special Publication, vol 17. pp 619–627
Stampfli GM, Borel GD (2002) A plate tectonic model for the
Paleozoic and Mesozoic constrained by dynamic plate bound-
aries and restored synthetic oceanic isochrons. Earth Planet Sci
Lett 196:17–33
Staudigel H (2003) Hydrothermal alteration processes in the oceanic
crust. In: Holland HD, Turekian KK (eds) Treatise on geochem-
istry, vol 3, pp 511–535
Stern RJ (2004) Subduction initiation: spontaneous and induced.
Earth Planet Sci Lett 226:275–292
Sun S-S, McDonough WF (1989) Chemical and isotopic systematics
of ocean basalts: implications for mantle composition and
processes. In: Saunders AD, Norry MJ (eds) Magmatism in the
ocean basins. Geol Soc London Special Publication, vol 42,
pp 313–345
Varne R, Brown AV, Falloon T (2000) Macquarie Island: its geology
and structural history, and the timing and tectonic setting of its
N-MORB to E-MORB magmatism. In: Dilek Y, Moores EM,
Elthon D, Nicolas A (eds) Ophiolites and oceanic crust: new
insights from field studies and the ocean drilling program, vol
349. Geological Society of America, Boulder, pp 301–320
Walter MJ (1998) Melting of garnet peridotite and the origin of
komatiite and depleted lithosphere. J Petrol 39:29–60
40 Contrib Mineral Petrol (2009) 157:23–40
123