Download Trace element composition and degree of partial melting of pelitic

Gondwana Research 9 (2006) 176 – 188
www.elsevier.com/locate/gr
Trace element composition and degree of partial melting of pelitic migmatites
from the Aoyama area, Ryoke metamorphic belt, SW Japan:
Implications for the source region of tourmaline leucogranites
Tetsuo Kawakami a,b,*, Tomoyuki Kobayashi c
a
Institute of Geology and Geoinformation, National Institute of Advanced Industrial Science and Technology (AIST), Central 7, Tsukuba, Ibaraki 305-8567, Japan
b
Department of Earth and Marine Sciences, The Australian National University, Canberra, ACT 0200, Australia
c
Department of Geology and Mineralogy, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan
Received 25 October 2004; accepted 24 June 2005
Available online 9 January 2006
Abstract
Degree of partial melting of pelitic migmatites from the Aoyama area, Ryoke metamorphic belt, SW Japan is determined utilizing whole-rock
trace element compositions. The key samples used in this study were taken from the migmatite front of this area and have interboudin partitions
filled with tourmaline-bearing leucosome. These samples are almost perfectly separated into leucosome (melt) and surrounding matrix (solid).
This textural feature enables an estimate of the melting degree by a simple mass-balance calculation, giving the result of 5 – 11 wt.% of partial
melting. Similar calculations applied to the migmatite samples, which assume average migmatite compositions to be the residue solid fraction,
give degree of melt extraction of 12 – 14 wt.% from the migmatite zone. The similarity of the estimated melting degree of 5 – 11 wt.% with that in
other tourmaline – leucogranites, such as Harney Peak leucogranite and Himalayan leucogranites, in spite of differences in formation process
implies that the production of tourmaline leucogranites is limited to low degrees of partial melting around 10 wt.%, probably controlled by the
breakdown of sink minerals for boron such as muscovite and tourmaline at a relatively early stage of partial melting. Because the amount of boron
originally available in the pelitic source rock is limited (on average ¨100 ppm), ¨10 wt.% of melting locally requires almost complete breakdown
of boron sink mineral(s) in the source rock, in order to provide sufficient boron into the melt to saturate it in tourmaline. This, in turn, means that
boron-depleted metapelite regions are important candidates for the source regions of tourmaline leucogranites.
D 2005 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.
Keywords: Degree of partial melting; Leucogranite; Migmatite; Trace element; REE
1. Introduction
Estimation of the degree of partial melting in the pelitic
migmatite zones is important for understanding how much melt
is produced, segregated and extracted from the region to
produce leucogranite bodies. Leucosomes in interboudin
partitions and shear bands in migmatitic rocks are often
interpreted to represent small volumes of felsic magma
squeezed out from the deformed matrix and transferred into
low-pressure site due to pressure gradients (e.g., Sawyer, 1991;
Brown, 1994; Brown et al., 1995; Oliver and Barr, 1997;
* Corresponding author. Institute of Geology and Geoinformation, National
Institute of Advanced Industrial Science and Technology (AIST), Central 7,
Tsukuba, Ibaraki 305-8567, Japan.
E-mail address: [email protected] (T. Kawakami).
Milord et al., 2001). This type of melt segregation process
separates the melt fraction and solid fraction almost perfectly
into leucosome and melanosome, respectively (Milord et al.,
2001), and the texture preserving this process is useful in
estimating the degree of partial melting by simple mass-balance
calculation if the compositions of melt and solid fractions are
known (Prinzhofer and Allegre, 1985; Sawyer, 1991).
Partial melting in the Ryoke metamorphic belt has been
recently described in detail and there is a wide occurrence of
migmatites (e.g., Hokada, 1996; Morikiyo, 1998; Brown, 1998;
Kawakami, 2001a; Kawakami and Ikeda, 2003). The study of
partial melting in the Ryoke metamorphic belt, however, still
remains qualitative and the degree of partial melting has not
been estimated quantitatively. In this paper, the degree of
partial melting of migmatites from the Aoyama area, Ryoke
metamorphic belt, SW Japan is estimated utilizing samples
1342-937X/$ - see front matter D 2005 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.gr.2005.06.009
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
collected from the migmatite front that preserves a texture
indicative of almost perfect separation of the melt and solid
fractions. The estimated degree of partial melting is then
compared with that in other tourmaline leucogranite suites, and
the similarity of the degree of melting and its importance are
discussed. Mineral abbreviations are after Kretz (1983).
2. General geology of the Ryoke metamorphic belt and the
Aoyama area
The Ryoke metamorphic belt, SW Japan (Fig. 1a) is a highly
elongate high-T/low-P type metamorphic belt of approximately
800 km in length (Miyashiro, 1965; Okudaira et al., 1993;
Ikeda, 1998a,b; Brown, 1998; Nakajima et al., 1990; Nakajima,
1994; Suzuki and Adachi, 1998). It is mainly composed of
pelitic and psammitic metamorphic rocks and metacherts, and
the highest grade zones are considered to have reached
granulite facies conditions at metamorphic peak (e.g., Ikeda,
2002). The metamorphic belt grades into the unmetamorphosed
Jurassic accretionary Mino – Tanba complex to the north
(Wakita, 1987). The low-T/high-P type Sanbagawa belt is
located to the south of the Ryoke metamorphic belt and the two
belts are separated by a major strike-slip fault, the Median
Tectonic Line (MTL). Eastward younging of the K –Ar and
Rb – Sr ages of the Ryoke granites and metamorphic rocks has
been ascribed to an eastward along-arc shift of granitic activity
(Nakajima et al., 1990; Nakajima, 1994). The chemical Th –
U – total Pb isochron method (CHIME) dating of monazite
from gneisses and granitoids, however, gives ages between 102
177
and 98 Ma both in the east and west of the belt, suggesting
more or less simultaneous igneous and metamorphic activity
along the whole length of the Ryoke belt followed by
denudation that was more rapid in the west than the east
(Suzuki and Adachi, 1998).
The samples used in this study are from the Aoyama area
(Fig. 1), one of the well-studied areas of the Ryoke
metamorphic belt, where high-grade metasedimentary rocks
are widely distributed (Yoshizawa et al., 1966; Hayama et al.,
1982; Takahashi and Nishioka, 1994; Kawakami, 2001a).
Subordinate amounts of calcareous metasediments, metachert
and metabasite are also distributed. The rock facies of the
pelitic-psammitic rocks are schists in the northern half of the
area (white part of Fig. 1b), and are anatectic migmatites
(metatexites to inhomogeneous diatexites; nomenclature after
Brown, 1973) in the southern half of the area (gray part of Fig.
1b). There is a tendency that inhomogeneous diatexites are
more common in the southwestern part of the migmatite zone
(Kawakami, 2001a). The schistosity of pelitic rocks in the
northern part of the area generally strikes E –W to WSW– ENE
and dips moderately either N or S, due to later upright folds
with fold axes trending E – W to WSW – ENE. The migmatitic
banding in central and southern parts strikes NW – SE to
WNW – ESE and dips, in most cases, NE to NNE. Intrafolial
folds with axial planes parallel to the penetrative migmatitic
layering are overprinted by upright folds with fold axes
trending NW –SE to WNW – ESE (Takahashi and Nishioka,
1994; Kawakami, 2001a). Massive, post-regional metamorphic
granodiorite and two-mica granite both intrude discordantly to
Fig. 1. (a) Simplified geological map of southwest Japan showing the location of the Aoyama area. (b) Geological map of the Aoyama area, modified from
Yoshizawa et al. (1966), Hayama et al. (1982) and Yoshida et al. (1995). Sample localities and metamorphic isograds are also shown.
178
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
the foliations of metasediments, and granodiorite is accompanied by the contact aureole (Fig. 1b). The evidences of contact
metamorphism in the adjacent country rocks are not found
around the two-mica granite.
The Aoyama area were previously divided into two regional
metamorphic zones and one contact metamorphic zone of postregional metamorphic origin, utilizing mineral assemblages in
pelitic lithology (Kawakami, 2001a). The regional metamorphic zones in the order of ascending metamorphic grade are (i)
sillimanite – K-feldspar zone, where the muscovite + quartz
assemblage is unstable and sillimanite + K-feldspar + biotite is
stable, and (ii) garnet – cordierite zone, where garnet + cordierite + biotite T sillimanite is stable. The contact metamorphic
zone is recognized by the occurrence of randomly oriented
andalusite porphyroblasts partly replaced by sillimanite and is
defined adjacent to the granodiorite pluton in the sillimanite –
K-feldspar zone (Fig. 1b). The granodiorite pluton is discordant
with respect to the schistosity of the regional metamorphic
rocks and the pluton contains xenoliths of surrounding regional
metamorphic rocks (Yoshida et al., 1995). Therefore, the pluton
clearly postdates regional metamorphism (Takahashi and
Nishioka, 1994, Kawakami, 2001a).
The peak P – T conditions estimated using garnet –biotite
geothermometers and GASP geobarometers are 3.0– 4.0 kbar,
615 –670 -C for the sillimanite – K-feldspar zone, and 4.5 –
6.0 kbar, 650 – 800 -C for the Grt –Crd zone (Kawakami,
2001a). However, these estimations are affected by the
retrograde re-equilibrium of Fe –Mg exchange in between
garnet and biotite, and thus are giving low-temperature
estimates. It is also probable that introduction of a spessartine
content into garnet stabilized the garnet + cordierite assemblage
in the low-temperature part of the garnet – cordierite zone,
giving low-temperature estimates for the garnet + cordierite
assemblage (Kawakami, 2001b). Taking these points and result
of pseudosection studies (e.g., White et al., 2001) into account,
Kawakami (2001b) considered that P –T paths of low- and
high-temperature parts of the garnet– cordierite zone went
through the metamorphic peak of ca. 730 -C, 4.5 kbar and ca.
800 -C, 5.5 kbar, respectively.
In the garnet – cordierite zone, sillimanite and biotite are
included in cordierite that constitutes the melanosome. In
addition, euhedral plagioclase that is indicative of the presence
of melts is observed in the leucosome. These textures suggest
that a dehydration melting reaction that consumed biotite and
sillimanite, such as
Bt þ Sil þ Qtz ¼ CrdFKfs þ melt
ð1Þ
and
Bt þ Sil þ Qtz ¼ Grt þ CrdFKfsFIlm þ melt
ð2Þ
took place in the garnet – cordierite zone. At the lowtemperature part of the garnet – cordierite zone, reaction (1) is
likely because garnet is not very common. At the hightemperature part of the garnet – cordierite zone, reaction (2) is
likely. These reactions are probably responsible for the
formation of migmatites in this area (Kawakami, 2001a,b).
Besides the garnet – cordierite isograd, a line marking the
disappearance of prograde tourmaline with increasing metamorphic grade may be mapped, and is termed the Ftourmalineout isograd_ (Kawakami, 2001a). Near the isograd, textures
involving the breakdown of tourmaline to sillimanite + cordierite, and interboudin partitions filled with tourmaline-bearing
leucosome, which is the key sample used in this study, are
found. The whole-rock composition of the leucosome is
suitable for the frozen melt (Kawakami, 2002). Based on these
observations, Kawakami (2001a, 2005) considered that the
breakdown of tourmaline occurred at the tourmaline-out
isograd during regional metamorphism to form sillimanite +
cordierite + boron-bearing melt, and that the melt was transferred into low-pressure interboudin partitions to crystallize
retrograde tourmaline. The reaction responsible for the
tourmaline-out isograd is considered to be:
TurFAbFKfs þ Qtz ¼ Sil þ Crd þ boron-bearing melt:
ð3Þ
Because these pelitic schists with interboudin partitions
filled with leucosome are important in this study, description of
them is summarized below.
3. Analytical methods and sample description
Twenty-six samples including pelitic schists, metatexites
and inhomogeneous diatexites (leucocratic and melanocratic
diatexites) are taken from the studied area (Fig. 1b) and major
and trace element compositions are determined. Nineteen of the
25 samples were previously analyzed for major elements (by
XRF) and boron (by PGNAA), and reported by Kawakami
(2001b). A 1 : 10 ratio of powdered rock sample (0.4 g) and
anhydrous lithium borate flux (4.0 g) was weighed into a Pt
crucible and fused at 1200 -C to prepare a glass bead sample.
Utilizing these glass bead samples, whole-rock major element
compositions of 6 samples are newly determined by Rigaku
Simultix system 3550 X-ray fluorescence spectrometer (see
details for Goto and Tatsumi, 1994). Trace element compositions including rare earth elements (REE) are determined by
LA-ICPMS analyses at the Research School of Earth Sciences,
Australian National University (Eggins, 2003). Glass bead
samples prepared for XRF analyses as mentioned above are
also used in these analyses. Trace element compositions of 2
samples were measured by Rigaku System 3070 (X-ray
fluorescence spectrometer) on pressed powder pellets at the
Institute for Geothermal Sciences, Kyoto University, Beppu,
Japan (Goto and Tatsumi, 1994; 1996).
Samples T14, H9, H20, L5, L10, W28 and 85b are pelitic
schists, samples U6, U4, 19, 66, 83M, 98, Y32A and Y26 are
metatexites, and samples P2, P3 and Y33 are inhomogeneous
diatexites (Fig. 1b; Table 1). Appearance of Y26 and P2 is the
most melanocratic in metatexites and inhomogeneous diatexites, respectively. Major constituting minerals of these samples
are given in Kawakami (2001b). Pelitic schists are considered
to be low-grade equivalents of migmatites.
Besides these samples, pelitic schists with interboudin
partitions filled with leucosome (Fig. 2a) are commonly
Table 1
Major and trace element concentrations of pelitic schists and migmatites from the Aoyama area
Rock
type
Pelitic schists
Sample
no.
T14
H20
L5
L10
W28
85b
Leucosomes in interboudin partitions
SAIT1
SAI991B
SAIT2
SAIT3
SAIT1B
SAIT2B
SAIT3B
59.87
0.89
18.20
5.39
0.04
2.56
1.64
2.25
7.90
0.17
98.93
3
1263.5
36.2
171.4
177.7
15.8
n.a.
256.6
19.2
25.1
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
–
54.24
1.21
18.59
9.81
0.14
2.04
1.84
2.81
8.29
0.45
99.42
31
558.2
17.1
599.1
174.8
65.2
n.a.
154.5
9.3
42.6
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
–
68.55
0.85
15.33
7.03
0.11
2.65
0.28
0.61
4.33
0.10
99.85
79
656.2
18.0
213.5
84.6
16.0
45.7
157.0
17.1
25.1
2.0
1.5
40.6
83.1
33.2
6.2
0.8
8.8
5.3
4.8
0.9
1.1
2.7
0.4
2.7
0.4
10.2
64.49
0.80
18.33
5.82
0.08
2.17
1.47
2.41
4.09
0.21
99.87
77
816.6
22.0
243.8
255.8
13.7
22.7
128.7
15.3
29.4
3.1
1.5
38.7
77.9
35.1
6.7
0.9
8.9
6.0
5.4
1.1
1.2
3.1
0.5
3.2
0.5
8.1
66.25
0.74
17.73
5.45
0.11
1.84
1.51
2.55
3.55
0.14
99.87
19
641.6
21.5
263.0
151.1
12.9
21.3
108.1
14.9
31.3
2.8
1.3
34.1
73.5
31.6
6.1
0.9
8.1
5.7
5.8
1.2
1.1
3.5
0.6
3.5
0.5
6.6
68.49
0.66
15.76
5.45
0.14
2.08
1.70
1.96
3.53
0.18
99.97
64
725.4
16.8
184.2
155.5
10.7
32.4
114.1
13.3
25.6
3.0
1.2
32.8
71.3
29.3
5.5
0.8
7.6
5.2
4.9
1.0
1.3
2.7
0.4
2.6
0.4
8.6
65.04
0.76
17.75
5.72
0.13
2.02
2.06
2.83
3.52
0.18
100.01
2
651.0
21.5
260.0
204.7
13.0
34.2
113.0
12.9
32.6
3.2
1.5
34.1
69.8
31.7
6.3
0.9
8.0
6.3
5.8
1.2
1.4
3.5
0.5
3.3
0.6
7.0
64.28
0.74
19.05
5.59
0.11
1.82
1.90
3.03
3.36
0.16
100.06
n.a.
442.2
18.1
254.5
160.4
12.8
34.7
111.6
12.4
28.8
2.8
1.8
30.7
67.1
28.1
6.0
0.9
7.4
5.7
5.7
1.1
1.1
3.3
0.5
3.5
0.5
6.0
64.41
0.75
18.58
5.79
0.14
1.94
0.99
1.91
5.10
0.14
99.74
n.a.
959.6
26.0
301.8
130.8
13.9
26.4
137.0
17.7
35.1
3.9
1.6
36.5
80.6
33.6
6.7
1.0
8.7
6.2
6.4
1.4
1.0
4.0
0.6
4.3
0.7
5.8
66.12
0.67
18.00
4.98
0.09
1.62
1.72
2.77
4.05
0.12
100.14
n.a.
786.8
22.9
241.5
173.6
9.6
37.5
102.7
11.4
25.1
2.4
1.1
25.2
55.8
24.2
4.9
0.7
6.2
4.7
4.5
0.9
1.0
3.0
0.4
3.0
0.5
5.7
74.98
0.14
14.72
0.93
0.02
0.42
0.66
1.57
5.75
0.24
99.43
745
720.9
37.1
43.4
147.3
3.6
14.6
98.7
3.0
12.0
2.2
1.3
8.1
17.9
7.0
1.6
0.3
1.9
1.9
2.2
0.4
0.7
1.4
0.2
1.6
0.2
3.4
73.96
0.08
15.12
0.47
0.01
0.18
0.64
1.77
7.12
0.21
99.56
283
1080.2
54.6
15.2
154.1
2.7
15.8
139.4
1.3
9.4
6.4
1.5
3.7
8.9
4.0
1.1
0.2
0.9
1.1
1.8
0.4
0.9
1.0
0.2
1.2
0.2
2.1
75.10
0.09
14.29
0.56
0.01
0.25
0.51
1.58
6.89
0.18
99.46
273
1288.4
64.5
28.6
154.8
2.2
17.1
128.9
2.3
10.3
2.0
0.6
5.5
13.1
4.9
1.0
0.2
1.3
1.3
1.7
0.4
1.0
1.1
0.2
1.2
0.2
3.0
72.54
0.16
14.94
1.34
0.03
0.50
0.63
1.35
7.38
0.16
99.03
n.a.
1317.8
50.5
41.6
173.7
4.2
19.1
131.8
2.6
11.1
5.4
2.0
7.0
15.7
6.5
1.6
0.3
1.7
1.6
1.8
0.4
1.0
1.2
0.2
1.6
0.2
2.9
–
XRF,
Kyoto U.
–
XRF,
Kyoto U.
0.56
LA-ICPMS,
ANU
0.57
LA-ICPMS,
ANU
0.56
LA-ICPMS,
ANU
0.75
LA-ICPMS,
ANU
0.66
LA-ICPMS,
ANU
0.56
LA-ICPMS,
ANU
0.49
LA-ICPMS,
ANU
0.61
LA-ICPMS,
ANU
1.24
LA-ICPMS,
ANU
2.31
LA-ICPMS,
ANU
2.47
LA-ICPMS,
ANU
1.90
LA-ICPMS,
ANU
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
SiO2
TiO2
Al2O3
Fe2O3
MnO
MgO
CaO
Na2O
K2O
P2O5
Total
B
Ba
Pb
Zr
Sr
Nb
Ni
Rb
Th
Y
U
Ta
La
Ce
Nd
Sm
Tb
Pr
Gd
Dy
Ho
Eu
Er
Tm
Yb
Lu
La(N) /
Yb(N)
Eu / Eu*
Trace element
analyses
H9
Intervening matrix parts of pelitic schists with
interboudin partitions filled with leucosome
Major element and boron concentrations are from Kawakami (2001b) and this study.
wt.% for major elements and ppm for trace elements (B – Lu).
Total Fe as Fe2O3.
n.a. = not analyzed, < 2 = below 2 ppm.
All of the trace element data except for T14, H9, and boron data are the average of three analyses.
179
180
Table 1 (continued)
Metatexites
Inhomogeneous diatexites
Sample no.
U6
U4
19
66
83M
98
Y32A
Y26
P2
P3
Y33
SiO2
TiO2
Al2O3
Fe2O3
MnO
MgO
CaO
Na2O
K2O
P2O5
Total
B
Ba
Pb
Zr
Sr
Nb
Ni
Rb
Th
Y
U
Ta
La
Ce
Nd
Sm
Tb
Pr
Gd
Dy
Ho
Eu
Er
Tm
Yb
Lu
La(N) / Yb(N)
Eu / Eu*
Trace element
analyses
69.03
0.65
15.83
4.49
0.05
1.73
2.23
3.68
2.31
0.20
100.20
10
193.4
18.0
223.0
265.7
10.6
21.1
94.0
13.1
24.1
2.8
1.4
31.1
64.2
27.5
5.4
0.7
7.0
4.6
4.5
0.9
1.2
2.8
0.4
2.7
0.4
7.7
0.69
LA-ICPMS,
ANU
69.26
0.59
16.09
5.23
0.22
1.59
1.41
3.13
2.50
0.10
100.11
<2
246.8
16.9
233.1
147.2
20.6
28.5
139.3
14.5
27.0
3.6
3.0
31.1
71.1
27.4
5.7
0.7
7.1
4.9
4.8
1.0
0.8
3.3
0.5
3.6
0.6
5.8
0.43
LA-ICPMS,
ANU
65.18
0.89
16.40
6.44
0.08
2.33
1.92
3.29
3.13
0.18
99.83
<2
276.8
14.6
364.8
172.2
16.2
31.1
174.1
22.5
33.1
2.7
1.2
50.8
99.5
43.0
7.9
1.0
11.3
7.0
6.2
1.3
1.0
3.5
0.5
3.2
0.5
10.7
0.41
LA-ICPMS,
ANU
66.04
0.72
17.34
5.22
0.11
2.02
1.77
2.65
4.02
0.17
100.06
2
772.3
23.3
221.6
224.7
14.5
20.7
132.7
15.8
22.7
2.1
1.7
34.8
71.3
30.6
5.7
0.7
7.6
5.0
4.3
0.8
1.2
2.5
0.4
2.8
0.5
8.5
0.67
LA-ICPMS,
ANU
61.73
0.88
17.99
7.09
0.19
2.39
2.19
4.06
3.09
0.12
99.75
<2
180.2
17.6
281.8
159.5
15.9
45.5
170.2
16.3
29.2
3.2
1.4
39.4
86.0
34.8
7.2
0.9
9.0
6.0
5.5
1.1
0.9
3.2
0.5
3.7
0.5
7.3
0.41
LA-ICPMS,
ANU
66.94
0.58
16.56
5.04
0.19
1.84
2.53
4.33
2.03
0.11
100.15
<2
279.6
17.3
225.9
160.1
9.4
36.3
103.8
12.5
30.0
2.7
1.1
31.3
69.3
29.6
6.1
0.9
7.2
5.6
5.6
1.2
1.3
3.4
0.6
3.8
0.6
5.5
0.67
LA-ICPMS,
ANU
63.81
0.96
17.79
7.24
0.09
2.43
1.53
2.54
3.63
0.10
100.12
4
457.4
18.8
289.4
209.5
15.6
36.4
128.8
20.4
29.0
2.0
1.3
47.6
94.9
40.8
7.0
0.9
10.4
6.1
5.8
1.1
1.2
3.3
0.5
3.5
0.5
9.3
0.56
LA-ICPMS,
ANU
60.80
1.19
19.26
8.32
0.09
3.22
0.94
1.54
4.19
0.08
99.64
<2
525.5
15.1
333.1
131.8
23.8
56.9
169.5
19.0
31.6
2.1
2.1
43.2
85.4
37.0
6.9
0.9
9.5
5.9
5.6
1.2
0.8
4.0
0.6
4.7
0.7
6.3
0.37
LA-ICPMS,
ANU
63.34
1.01
17.15
6.76
0.07
2.35
2.49
3.66
2.70
0.11
99.65
<2
203.2
18.6
279.7
228.2
25.2
33.1
124.9
19.0
16.4
2.6
2.3
41.5
81.6
33.8
5.8
0.6
8.9
4.8
3.5
0.7
1.1
1.6
0.2
1.5
0.2
18.9
0.62
LA-ICPMS,
ANU
69.13
0.55
15.96
3.50
0.05
1.12
1.36
2.87
5.30
0.30
100.14
3
883.3
41.8
227.2
197.7
14.8
23.8
155.0
17.2
79.4
4.1
2.0
36.3
77.3
34.1
8.0
1.8
8.5
9.7
13.6
2.8
1.0
8.6
1.1
7.3
1.0
3.4
0.35
LA-ICPMS,
ANU
70.14
0.59
16.43
3.99
0.04
1.62
1.47
2.44
3.23
0.10
100.04
5
787.4
18.0
205.7
220.5
13.7
16.9
92.4
13.9
15.5
2.3
1.6
33.1
67.7
27.6
5.5
0.7
7.5
5.0
3.7
0.6
1.1
1.4
0.2
1.2
0.1
19.2
0.65
LA-ICPMS,
ANU
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
Rock type
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
181
Fig. 2. (a) A picture of pelitic schist with an interboudin partition filled with leucosome. Black crystals in the midst of the leucosome are tourmaline. (b) CIPW
normative Qtz – Ab – Or plot of whole-rock compositions of the leucosomes in interboudin partitions (white circles). Gray shaded area represents the glass
compositions obtained by melting two-mica, plagioclase-poor pelites at 800 and 850 -C, 3 kbar (Spicer et al., 2004). Liquidus phase relations in the system Qtz –
Ab – Or – H2O at 5 kbar and X(H2O) = 1.0 are taken from Holtz et al. (1992).
observed near the tourmaline-out isograd that is defined near
the schist/migmatite boundary of the Aoyama area (Fig. 1b; see
also Kawakami, 2002). The leucosome is composed of quartz,
K-feldspar, plagioclase, tourmaline, biotite, muscovite,
T andalusite and apatite. Plagioclase, apatite, andalusite and
tourmaline are euhedral. The whole-rock composition of
leucosomes is consistent with the leucosome being a frozen
melt (Kawakami, 2002). Fig. 2b shows the CIPW normative
plot of the compositions of leucosome in interboudin partitions.
It is clear from this figure that the leucosomes have similar
compositions to melts produced by the melting of pelitic schists
of low Na2O / K2O ratio (0.21 – 0.41 in wt.% ratio) at 3 kbar,
800 –850 -C (Spicer et al., 2004). This point will be further
discussed below.
Tourmaline showing a partial breakdown texture was found
in the intervening matrix, which lead Kawakami (2001a) to
conclude that the boron in the leucosome was derived from the
breakdown of tourmaline in the surrounding country rocks
(i.e., closed-system behavior of boron and melt on the sample
scale). The interboudin partitions found near the tourmalineout isograd are often connected with each other in the Aoyama
area (Kawakami, 2005). However, it is important that all of the
leucosome samples used in this study have three-dimensionally closed shapes and not connected to veins or shear zones.
Those leucosomes do not contain mafic aggregates and are
pure leucosomes. From these lines of evidence, the leucosome
was interpreted to be a frozen melt collected from the
surroundings (Kawakami, 2001a, 2002) and thus the SAIseries samples are the typical example of partially molten
pelites that are almost perfectly separated into the melt fraction
(leucosome part; samples SAIT1, SAIT2 and SAIT3) and into
the residual solid fraction (intervening matrix parts; SAI991B,
SAIT1B, SAIT2B and SAIT3B). Samples SAITn and SAITnB
(n = 1, 2 and 3) are separated from the same sample (i.e.,
SAIT1B is the leucosome in interboudin partitions and SAIT1
is the intervening matrix part of the same sample) and utilized
in XRF and LA-ICPMS analyses explained above. Samples
L5, L10 and 85b are the pelitic schists collected close to the
tourmaline-out isograd, but boudinage structures are not
developed.
4. Whole-rock trace element compositions
Whole-rock major and trace element compositions of
analyzed samples are summarized in Table 1. All of the trace
element data listed in Table 1 are the average of three analyses,
except for T14, H9, and boron data of all samples. Fig. 3 is the
multi-element variation diagrams normalized to primordial
mantle of Taylor and McLennan (1985). It is clear from this
plot that pelitic schists and migmatites have almost the same
trace element composition except for Ba. Leucosomes are
enriched in Ba, K and Ta, and depleted in Th, Nb, La, Ce, Nd,
Zr, Sm, Ti, Tb and Y than pelitic schists and metatexites.
Compositional difference between pelitic schists and the
intervening matrix parts of boudinaged rocks (samples SAIT1,
SAIT2 and SAIT3) is not observed in this diagram. Inhomogeneous diatexites are different from metatexites in the
abundance of Tb and Y.
Fig. 4 is the multi-element variation diagrams of pelitic
schist compositions normalized to the average trace element
abundances of Faverage pelitic schist_, which is the average of
samples H20, L5, L10, W28 and 85B (Table 2). Apparently,
samples H9 and T14 have extraordinary high concentrations of
LIL or HFS elements, such as K, Zr, Ti and Nb for sample H9,
and K, Rb and Ba for sample T14, indicative of the effect of Kmetasomatism and/or originally untypical whole-rock chemistry. These samples have very low SiO2 contents compared with
other pelitic schist samples. However, these cannot be restites
because they are taken from Fschist zone_ where pelitic and
psammitic schists dominate and significant partial melting is
not observed (Fig. 1). Because samples H9 and T14 are not
Ftypical_ pelitic schists and are not appropriate for the aim of
182
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
matrix part of boudinaged samples, metatexites, and inhomogeneous diatexites. Pelitic schists show negative Eu
anomaly (Eu / Eu* = 0.56 – 0.75) with flat HREE pattern,
where Eu / Eu* = EuN / (SmN / 2 + GdN / 2). Leucosomes filling
the interboudin partitions are depleted in REE relative to
pelitic schists and metatexites, and show positive Eu anomaly
(Eu / Eu* = 1.24 – 2.47) with flat HREE pattern. Metatexites
have almost the same REE patterns with pelitic schists
except for the stronger negative Eu anomaly (Eu / Eu* = 0.37–
0.69) than the pelitic schists. Inhomogeneous diatexites show
variations in terms of HREE abundance, and no significant
difference between metatexites and inhomogeneous diatexites
is found in terms of LREE abundances. High concentrations
of HREE in sample P3 may be not explained by the high
modal abundance of garnet and apatite in P3, because garnet
is absent and apatite is very rare in this sample. High modal
amount of zircon is excluded because Zr content is not
prominently high in P3, relative to other inhomogeneous
diatexites. The fact that P3 is also enriched in P (Table 1)
and Y (Fig. 3) rather suggests that REE pattern of P3 is
strongly reflecting the high modal abundance of xenotime
(e.g. Bea, 1996). Intervening matrix parts of boudinaged
samples have almost the same pattern as pelitic schists, and
the difference between them may be best recognized in the
REE-diagram normalized to Faverage pelitic schist_ (Fig. 7).
There is a tendency that intervening matrix parts are a little
richer in HREE and slightly depleted in LREE, especially
Eu, relative to pelitic schists. This may reflect the restitic
nature of the intervening matrix parts. Weak negative Eu
anomaly observed in the intervening matrix samples may
reflect the effect of disequilibrium melting; REE that are
mainly contained in accessory minerals went through
disequilibrium melting, whereas Eu that is mainly contained
in major mineral as feldspars went through equilibrium
Fig. 3. Trace element compositions of (a) pelitic schists and intervening matrix
parts of boudinaged sample, (b) metatexites and leucosomes in interboudin
partitions, and (c) inhomogeneous diatexites from the Aoyama area, normalized
to primordial mantle of Taylor and McLennan (1985).
this study to estimate the degree of partial melting, they are not
used in this study. Excluding these two, the compositional data
of pelitic schists show good concentration (Fig. 4). Compositional difference between pelitic schists and other rock types
mentioned above may be better recognized by the multielement variation diagrams normalized to Faverage pelitic
schist_ (Fig. 5). The trend that metatexites are depleted
significantly in Ba and slightly in K than the average pelitic
schist is clearly observed in this plot.
Fig. 6 is the chondrite-normalized REE-diagram for pelitic
schists, leucosomes in interboudin partitions, intervening
Fig. 4. Trace element compositions of the pelitic schists from the Aoyama area,
normalized to Faverage pelitic schist_. See Table 2 for compositional data of
Faverage pelitic schist_.
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
Table 2
Average composition of pelitic schists, leucosomes in interboudin partitions,
intervening matrix of boudinaged samples, and metatexites used in calculations
Rock type (number
of averaged samples)
Pelitic
schist
(n = 5)
Intervening
matrix
(n = 3)
Leucosome
(n = 4)
Metatexite
(n = 8)
SiO2
TiO2
Al2O3
Fe2O3
MnO
MgO
CaO
Na2O
K2O
P2O5
Total
B
Ba
Pb
Zr
Sr
Nb
Ni
Rb
Th
Y
U
Ta
La
Ce
Nd
Sm
Tb
Pr
Gd
Dy
Ho
Eu
Er
Tm
Yb
Lu
La(N) / Yb(N)
Eu / Eu*
66.6
0.8
17.0
5.9
0.1
2.2
1.4
2.1
3.8
0.2
99.9
48.2
698.2
20.0
232.9
170.3
13.2
31.3
124.2
14.7
28.8
2.8
1.4
36.0
75.1
32.2
6.2
0.8
8.3
5.7
5.3
1.1
1.2
3.1
0.5
3.1
0.5
8.1
0.62
64.94
0.7
18.5
5.5
0.1
1.8
1.5
2.6
4.2
0.1
100.0
n.a.
729.5
22.4
265.9
154.9
12.1
32.9
117.1
13.8
29.6
3.0
1.5
30.8
67.8
28.6
5.9
0.9
7.4
5.6
5.5
1.1
1.0
3.4
0.5
3.6
0.6
5.8
0.55
74.1
0.1
14.8
0.8
0.0
0.3
0.6
1.6
6.8
0.2
99.4
433.7
1101.8
51.7
32.2
157.5
3.2
16.6
124.7
2.3
10.7
4.0
1.3
6.1
13.9
5.6
1.3
0.3
1.5
1.5
1.9
0.4
0.9
1.2
0.2
1.4
0.2
2.8
1.98
65.3
0.8
17.2
6.1
0.1
2.2
1.8
3.2
3.1
0.1
100.0
2.6
366.5
17.7
271.6
183.8
15.8
34.6
139.0
16.8
28.3
2.6
1.6
38.7
80.2
33.9
6.5
0.8
8.7
5.6
5.3
1.1
1.0
3.3
0.5
3.5
0.5
7.6
0.53
Major element and boron concentrations are from Kawakami (2001b) and this
study.
wt.% for major elements and ppm for trace elements (B – Lu).
Total Fe as Fe2O3.
n.a. = not analyzed.
melting (that is, Eu may be more rapidly taken into melt than
other REE).
5. Discussion
5.1. Leucosomes in interboudin partitions — are they frozen
melts?
The leucosomes in interboudin partitions now consist of
granitic mineral assemblages including euhedral plagioclase.
Generally, plagioclase is only rarely euhedral in metamorphic
rocks (Spry, 1969), and the euhedral shape implies its growth
in free space, melt or fluid. Three-dimensionally closed shape
183
of leucosomes is the supporting evidence that melt or fluid
segregated from intervening matrix and that melt or fluid did
not contain significant amount of crystals including those
formed simultaneously with the melt or fluid, when they were
segregated into the interboudin partitions.
The normative Qtz –Ab – Or plots of the leucosomes are in
good agreement with experimental melt compositions (Fig. 2),
showing that the leucosomes can be interpreted to be frozen
melts. Although the experiments cited in Fig. 2 were performed
at lower pressures (3 kbar) than the estimated pressure of the
Aoyama area (¨5 kbar), this difference may not affect the
conclusion that the leucosomes possess the composition of
frozen melts, because leucosome compositions (SAI series) are
plotted near the quartz– orthoclase cotectic line at 5 kbar and
because location of the quartz – orthoclase cotectic line in the
Qtz –Ab – Or system does not differ very much between 3 to 5
kbar (e.g. Holtz et al., 1992). The only major difference caused
by the pressure difference will be the temperature that is needed
to produce the same melt composition, and at 5 kbar, the
leucosome (melt) compositions of the Aoyama samples may be
produced at 700– 740 -C under X(H2O) = 1.0 and at about
820 -C under X(H2O) = 0.7 (Holtz et al., 1992). This is
concordant with the temperature estimates of the Aoyama area.
The aluminum saturation index [ASI = Al 2O3 / (CaO +
Na2O + K2O)] (Zen, 1986) of the leucosomes range from 1.28
to 1.47. Based on Acosta-Vigil et al. (2003), hydrous granitic
melts with ASI = 1.28 – 1.47 coexist with tourmaline at about
730– 810 -C at 2 kbar. The normative corundum of the SAIseries leucosomes range from 3.59 to 5.29. Such melts coexist
with tourmaline at above 780 -C at 2 kbar. These temperatures
are almost concordant with the estimated temperature of the
migmatite front (730 -C).
Bea (1996) reported that Eu is always essentially contained
within feldspars, and epidote may also contain a significant
proportion. However, epidote is not contained in the pelitic
rocks from the Aoyama area and thus Eu will be mostly
contained in feldspars. Therefore, behavior of Eu will be not
affected by the disequilibrium melting process related to
accessory minerals, and solely controlled by the melting
reactions containing feldspars. Positive Eu anomaly of leucosomes and negative Eu anomaly of intervening matrix parts
relative to pelitic schists (Fig. 7) suggest that feldspars were
consumed in the course of partial melting to produce
leucosome melts, and feldspar components are taken into melt.
This is the supporting evidence for the model proposed by
Kawakami (2001a) that the melt was produced in the
intervening matrix parts of boudinaged pelitic samples and
segregated into interboudin partitions. From these lines of
evidence, leucosomes that are filling the interboudin partitions
of boudinaged pelitic schists (SAI-series samples) are interpreted to be nearly frozen melts.
The zirconium contents of the leucosomes in interboudin
partitions are below the concentration that is required to
saturate the peraluminous melt of SiO2 = 73 wt.% in zircon
(Watson, 1988), and the calculated Zr / Zr* ratio are 0.16– 0.45.
The leucosomes may be classified into a Fdisequilibrium melt_
of Sawyer (1991).
184
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
Fig. 5. Trace element compositions of (a) leucosomes and metatexites, and (b) inhomogeneous diatexites from the Aoyama area, normalized to Faverage pelitic
schist_.
5.2. Estimating degree of partial melting by mass-balance
calculation
Kawakami (2005) estimated the degree of partial melting of
the pelitic rock sample with interboudin partitions filled with
leucosome to be 12 wt.%, using average boron contents of the
leucosomes in interboudin partitions, the intervening matrix
parts, and of the average pelitic schist compositions. He
assumed the closed-system behavior of boron and melts within
the sample scale. His conclusion can be evaluated utilizing
other trace elements, and the new estimates are given below.
Under the assumption that SAI-series samples are almost
perfectly separated into melt and solid fractions, mass balance
can be expressed as
C0 ¼ C1 ð1 F Þ þ C2 F
where C 0 (ppm) is the whole-rock trace element
of the pelitic schists before partial melting, C 1
residues, C 2 is that of the leucosomes, and F is
partial melting in wt.% basis (e.g. Prinzhofer
ð4Þ
concentration
is that of the
the degree of
and Allegre,
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
185
Fig. 6. Chondrite-normalized REE plots of (a) pelitic schists, (b) leucosomes and intervening matrix parts of boudinaged samples, (c) metatexites and (d)
inhomogeneous diatexites from the Aoyama area. Chondrite values are taken from Taylor and McLennan (1985).
Fig. 7. REE plot of leucosomes and intervening matrix parts of boudinaged
samples normalized to Faverage pelitic schist_.
1985). In this study, composition of samples SAITn is used for
C 1, and the composition of SAITnB is used for C 2. Using these
values, C 0 is calculated for given F (= calculated C 0) and
compared with observed C 0 values (= Faverage pelitic schist_ of
Table 2). In order to determine which value of F gives the bestfit composition to Faverage pelitic schist_ (= observed C 0), the
least square method was used. In this calculation, all the major
and trace element compositions available are used and dealt
equally. As a result, SAIT1 – SAIT1B and SAIT2 – SAIT2B
pairs gave F values of 5 and 11 wt.% (Table 3). A SAIT3 –
SAITB3 pair gave F = 0 wt.% to be the best-fit result. This may
be due to the inappropriate assumption of protolith composition
for a SAIT3 –SAITB3 pair. The degree of partial melting at the
migmatite front of the Aoyama area may be estimated to be
5– 11 wt.%.
Alternative way of estimating the degree of partial melting
is to separate the pure restitic part and analyze its composition
to utilize in this calculation. However, it is difficult to do so in
the case of Aoyama area because melanosome patches
contained in the boudinaged samples are small and irregularly
186
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
Table 3
Summary of F values calculated by Eq. (4)
Protolith (C 0)
Average
Average
Average
Average
Average
Average
Average
pelitic
pelitic
pelitic
pelitic
pelitic
pelitic
pelitic
schist
schist
schist
schist
schist
schist
schist
Residue (C 1)
Leucosome (C 2)
SAIT1
SAIT2
SAIT3
Average
Average
Average
Average
SAITB1
SAITB2
SAITB3
SAITB1
SAITB2
SAITB3
Average leucosome
migmatite
migmatite
migmatite
migmatite
shaped (see Fig. 3b of Kawakami, 2001a). Even if the pure
restite parts are successfully separated, one has to be careful
because the F value obtained from the mass-balance calculation may represent the degree of partial melting in the specific
rock part where partial melting reactions selectively took place
(this is why melanosome patches are produced). In order to
obtain Fbulk_ degree of partial melting, modal information of
restitic part should be taken into account, which will result in
lowering the degree of partial melting estimated by this
method.
Metatexite migmatites are presumably a mixture of residual
solid and leucosome, where the leucosome may represent some
mixture of residual felsic minerals, early crystallized felsic
minerals (cumulates) (Brown, 2001), and possibly unextracted
felsic melts. Because SAI-series leucosomes may represent
melt compositions that were produced almost in situ at the
migmatite front, it is interesting to consider the degree of melt
extraction from the metatexite migmatites under some assumptions. The assumptions made here are (i) protolith composition
of metatexites are Faverage pelitic schist_, and (ii) melt
composition that extracted from metatexites is equivalent to
average leucosome compositions of SAI series (Table 2).
Following calculation is based on Eq. (4), and composition of
average metatexite (Table 2) is used for C 1, and the
composition of SAITnB (n = 1, 2 and 3) is used for C 2. In
this calculation, F does not represent the degree of partial
melting in metatexites but represents the degree of melt
extraction (in wt.%) from the metatexites. Then, C 0 is
calculated for a given F and compared with observed C 0
values (= Faverage pelitic schist_ of Table 2). In order to
determine which value of F gives the best-fit composition to
Faverage pelitic schist_, the least square method was used. All
the major and trace element compositions available are used
and dealt equally. As a result, F values of 12 –14 wt.% are
obtained (Table 3). If average leucosome composition (Table 2)
is used for C 2 instead of SAITnB, then the F value is 13 wt.%.
As some melt probably remained unextracted in metatexites, it
may be concluded that the degree of partial melting in most
part of the migmatite zone exceeded the degree of melt
extraction of 12 –14 wt.%.
5.3. Implications for the source region of tourmaline
leucogranites
Because pelitic migmatite zones are potential sources for
leucogranites, and because the presence of tourmaline-bearing
Calculated F (wt.%)
5
11
0
12
13
14
13
leucosome in the migmatite front of the Aoyama area shows
that tourmaline – leucogranite was produced there, it is
interesting to compare the degree of partial melting of the
Aoyama area with that estimated in the tourmaline leucogranite suites. The Harney Peak leucogranite (Nabelek et al.,
1992; Wilke et al., 2002) and Himalayan leucogranites (Harris
and Inger, 1992) are well-studied examples of tourmaline
leucogranites and are considered to have been produced via
muscovite dehydration melting reaction. The degrees of
partial melting to produce the leucogranites from muscovite
schists are estimated to be 10– 14 wt.% for the Harney Peak
leucogranite (Nabelek et al., 1992; Wilke et al., 2002), and
about 12 wt.% for the Himalayan leucogranites (Harris and
Inger, 1992). The result of this study shows that more than
5– 11 wt.% of partial melting in the migmatite front of the
Aoyama area, which is quite similar to the values estimated in
other leucogranite suites, produced the tourmaline leucogranite although it did not segregate into a pluton scale. The
source of boron to produce boron-bearing melt in the Aoyama
area is considered to be the breakdown of tourmaline, and
other melting reactions involving biotite, such as reactions (1)
and (2), are also responsible for the production of melts at
that metamorphic grade (Kawakami, 2001a). In the migmatite
zone where higher degrees of partial melting are inferred from
the calculation given above, no tourmaline-bearing leucogranite is found. This is probably because boron content in the
leucosome melt was not high enough to crystallize tourmaline
(e.g., Wolf and London, 1997) or high Ti content preferred
crystallization of biotite instead of tourmaline (e.g., Nabelek
and Bartlett, 1998). Anyway, the mechanism of producing
tourmaline leucogranite, including the reaction to produce it,
and P – T conditions of its production are different from those
considered for the Harney Peak leucogranite and Himalayan
leucogranites. The similarity of the degree of partial melting,
in spite of these differences, implies that production of the
tourmaline leucogranite is limited to low degrees of partial
melting (¨10 wt.%) regardless of P – T conditions and
reaction by which it is produced. This low degree of melting
of ¨10 wt.% is probably controlled by the breakdown of sink
minerals of boron such as muscovite and tourmaline at a
relatively early stage of the partial melting of the pelitic
source rocks. Low-temperature breakdown of tourmaline may
require the coexistence with quartz (i.e., breakdown of
tourmaline + quartz; von Goerne et al., 1999) and appropriate
tourmaline composition, possibly tourmaline with substantial
amount of X-site vacancy and schorl component (e.g.,
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
Schreyer and Werding, 1997; Kawakami, 2001b; Kawakami
and Ikeda, 2003). Because of the limited amount of boron
originally available in the pelitic source rocks (on average
¨100 ppm; Harder, 1970), ¨10 wt.% of melting requires
almost complete breakdown of sink mineral(s) of boron in the
source rock in some cases (e.g., Kawakami, 2001a, b;
Kawakami and Ikeda, 2003; Wilke et al., 2002), in order to
provide sufficient amounts of boron into the melt to saturate
in tourmaline, although the concentration of boron required
for the tourmaline saturation in leucogranite melts is still
controversial (e.g., Benard et al., 1985; Holtz and Johannes,
1991; Scaillet et al., 1995; Spicer et al., 2004; Wolf and
London, 1997). This, in turn, means that boron-depleted
metapelite regions, except for the boron-depleted contact
aureoles around some plutons (e.g., Wilke et al., 2002), are
important candidates for source regions of tourmaline
leucogranites.
Acknowledgements
We are grateful to C. Allen and D. Rubatto for the LAICPMS analyses at RSES, ANU, and T. Shibata, K. Sato, Y.
Tatsumi and T. Hoshide for the XRF analyses at Kyoto
University. We would like to thank T. Ikeda, K. Miyazaki and
M. Owada for reading the manuscript and giving us valuable
comments. Thanks are also due to I. Buick and M. Brown for
constructive reviews and K. Sajeev for editorial efforts. This
study was mainly done during T. Kawakami’s stay in DEMS,
ANU as a visiting fellow, and was financially supported
by the Grant-in-Aid for JSPS Fellows (No. 05864) to T.
Kawakami.
References
Acosta-Vigil, A., London, D., Morgan VI, G.B., Dewers, T.A., 2003. Solubility
of excess alumina in hydrous granitic melts in equilibrium with peraluminous minerals at 700 – 800 -C and 200 MPa, and applications of the
aluminum saturation index. Contrib. Mineral. Petrol. 146, 100 – 119.
Bea, F., 1996. Residence of REE, Y, Th and U in granites and crustal protoliths;
implications for the chemistry of crustal melts. J. Petrol. 37, 521 – 552.
Benard, F., Moutou, P., Pichavant, M., 1985. Phase relations of tourmaline
leucogranites and the significance of tourmaline in silicic magmas. J. Geol.
93, 271 – 291.
Brown, M., 1973. The definition of metatexis, diatexis and migmatite. Proc.
Geol. Assoc. 84, 371 – 382.
Brown, M., 1994. The generation, segregation, ascent and emplacement of
granite magma: the migmatite-to-crustally-derived granite connection in
thickened orogens. Earth-Sci. Rev. 36, 83 – 130.
Brown, M., 1998. Unpairing metamorphic belts: P – T paths and a tectonic
model for the Ryoke Belt, southwest Japan. J. Metamorph. Geol. 16, 3 – 22.
Brown, M., 2001. Orogeny, migmatites and leucogranites: a review. Proc.
Indian Acad. Sci., A Earth Planet. Sci. 110, 313 – 336.
Brown, M., Averkin, Y.A., McLellan, E.L., 1995. Melt segregation in
migmatites. J. Geophys. Res. 100, 15,655 – 15,679.
Eggins, S.M., 2003. Laser ablation ICP-MS analysis of geological materials
prepared as lithium borate glasses. Geostand. Newsl. 27, 147 – 162.
Goto, A., Tatsumi, Y., 1994. Quantitative analysis of rock samples by an X-ray
fluorescence spectrometer (I). Rigaku J. 11, 40 – 59.
Goto, A., Tatsumi, Y., 1996. Quantitative analysis of rock samples by an X-ray
fluorescence spectrometer (II). Rigaku J. 13, 20 – 38.
187
Harder, H., 1970. Boron content of sediments as a tool in facies analysis.
Sediment. Geol. 4, 153 – 175.
Harris, N., Inger, S., 1992. Trace element modeling of pelite-derived granites.
Contrib. Mineral. Petrol. 110, 46 – 56.
Hayama, Y., Yamada, T., Ito, M., Kutsukake, T., Masaoka, K., Miyakawa, K.,
Mochizuki, Y., Nakai, Y., Tainosho, Y., Yoshida, M., Kawarabayashi, I.,
Tsumura, Y., 1982. Geology of the Ryoke Belt in the eastern Kinki District,
Japan — the phase-divisions and the mutual relations of the granitic rocks.
Jpn. J. Geol. Soc. 88, 451 – 466R (in Japanese with English abst).
Hokada, T., 1996. Orthopyroxene from a metabasite of the Ryoke belt in the Ina
district, Nagano Prefecture, central Japan: textural evidence of partial
melting. J. Jpn. Assoc. Mineral. Petrol. Econ. Geol. 91, 253 – 265.
Holtz, F., Johannes, W., 1991. Effect of tourmaline on melt fraction and
composition of first melts in quartzofeldspathic gneiss. Eur. J. Mineral. 3,
527 – 536.
Holtz, F., Pichavant, M., Barbey, P., Johannes, W., 1992. Effects of H2O on
liquidus phase relation in the haplogranite system at 2 and 5 kbar. Am.
Mineral. 77, 1223 – 1241.
Ikeda, T., 1998a. Progressive sequence of reactions of the Ryoke metamorphism in the Yanai district, southwest Japan: the formation of cordierite.
J. Metamorph. Geol. 16, 39 – 52.
Ikeda, T., 1998b. Phase equilibria and the pressure – temperature path of the
highest-grade Ryoke metamorphic rocks in the Yanai district, SW Japan.
Contrib. Mineral. Petrol. 132, 321 – 335.
Ikeda, T., 2002. Regional occurrence of orthopyroxene-bearing basic rocks in
the Yanai district, SW Japan: evidence for granulite-facies Ryoke
metamorphism. Isl. Arc 11, 185 – 192.
Kawakami, T., 2001a. Tourmaline breakdown in the migmatite zone of the
Ryoke metamorphic belt, SW Japan. J. Metamorph. Geol. 19, 61 – 75.
Kawakami, T., 2001b. Boron depletion controlled by the breakdown of
tourmaline in the migmatite zone of the Aoyama area, Ryoke metamorphic
belt, SW Japan. Can. Mineral. 39, 1529 – 1546.
Kawakami, T., 2002. Magmatic andalusite from the migmatite zone of the
Aoyama area, Ryoke metamorphic belt, SW Japan, and its importance in
constructing the P – T path. J. Mineral. Petrol. Sci. 97, 241 – 253.
Kawakami, T., 2005. Tourmaline and boron as indicators of the presence,
segregation and extraction of melt in pelitic migmatites: examples from the
Ryoke metamorphic belt, SW Japan. Trans. R. Soc. Edinb. Earth Sci. 95,
111 – 124.
Kawakami, T., Ikeda, T., 2003. Depletion of whole-rock boron controlled by
the breakdown of tourmaline and retrograde formation of borosilicates in
the Yanai area, Ryoke metamorphic belt, SW Japan. Contrib. Mineral.
Petrol. 145, 131 – 150.
Kretz, R., 1983. Symbols for rock forming minerals. Am. Mineral. 68,
277 – 279.
Milord, I., Sawyer, E.W., Brown, M., 2001. Formation of diatexite migmatite
and granite magma during Anatexis of semi-pelitic metasedimentary rocks:
an example from St. Malo, France. J. Petrol. 42, 487 – 505.
Miyashiro, A., 1965. Metamorphic Rocks and Metamorphic Belts. Iwanami
Shoten Publishers, Tokyo. 458 pp. (in Japanese).
Morikiyo, T., 1998. Mode of occurrence of migmatites and metabasites in the
Ryoke metamorphic belt. Excursion Guidebook for the 105th Annual
Meeting of the Geological Society of Japan, pp. 205 – 230 (in Japanese).
Nabelek, P.I., Bartlett, C.D., 1998. Petrologic and geochemical links between
the post-collisional Proterozoic Harney Peak leucogranite, South Dakota,
USA, and its source rocks. Lithos 45, 71 – 85.
Nabelek, P.I., Russ-Nabelek, C., Denison, J.R., 1992. The generation and
crystallization conditions of the Proterozoic Harney Peak leucogranites,
Black Hills, South Dakota, USA; petrogenic and geochemical constraints.
Contrib. Mineral. Petrol. 110, 173 – 191.
Nakajima, T., 1994. The Ryoke plutonometamorphic belt: crustal section of the
Cretaceous Eurasian continental margin. Lithos 33, 51 – 66.
Nakajima, T., Shirahase, T., Shibata, K., 1990. Along-arc lateral variation of
Rb – Sr and K – Ar ages of Cretaceous granitic rocks in southwest Japan.
Contrib. Mineral. Petrol. 104, 381 – 389.
Okudaira, T., Hara, I., Sakurai, Y., Hayasaka, Y., 1993. Tectono-metamorphic
processes of the Ryoke belt in the Iwakuni-Yanai district, southwest Japan.
Mem. Geol. Soc. Jpn. 42, 91 – 120.
188
T. Kawakami, T. Kobayashi / Gondwana Research 9 (2006) 176 – 188
Oliver, N.H.S., Barr, T.D., 1997. The geometry and evolution of magma
pathways through migmatites of the Halls Creek orogen, western Australia.
Mineral. Mag. 61, 3 – 14.
Prinzhofer, A., Allegre, C.J., 1985. Residual peridotites and the mechanisms of
partial melting. Earth Planet. Sci. Lett. 74, 251 – 265.
Sawyer, E.W., 1991. Disequilibrium melting and the rate of melt-residuum
separation during migmatization of mafic rocks from the Grenville Front.
Quebec. J. Petrol. 32, 701 – 738.
Scaillet, B., Pichavant, M., Roux, J., 1995. Experimental crystallization of
leucogranite magmas. J. Petrol. 36, 663 – 705.
Schreyer, W., Werding, G., 1997. High-pressure behaviour of selected boron
minerals and the question of boron distribution between fluids and rocks.
Lithos 41, 251 – 266.
Spicer, E.M., Stevens, G., Buick, I.S., 2004. The low-pressure partial-melting
behaviour of natural boron-bearing metapelites from the Mt. Stafford area,
central Australia. Contrib. Mineral. Petrol. 148, 160 – 179.
Spry, A., 1969. Metamorphic Textures. Pergamon, London. 350 pp.
Suzuki, K., Adachi, M., 1998. Denudation history of the high T/P Ryoke
metamorphic belt, southwest Japan: constraints from CHIME monazite ages
of gneisses and granitoids. J. Metamorph. Geol. 16, 23 – 37.
Takahashi, Y., Nishioka, Y., 1994. Mode of plagioclase twinnings in Ryoke
metamorphic rocks in the western area of Tsu City, Mie Prefecture. J. Jpn.
Assoc. Mineral. Petrol. Econ. Geol. 89, 261 – 268 (in Japanese with English
abst).
Taylor, S.R., McLennan, S.M., 1985. The Continental Crust; its Composition
and Evolution; an Examination of the Geochemical Record Preserved in
Sedimentary Rocks. Blackwell, Oxford. 312 pp.
von Goerne, G., Franz, G., Robert, J.L., 1999. Upper thermal stability of
tourmaline + quartz in the system MgO – Al2O3 – SiO2 – B2O3 – H2O and
Na2O – MgO – Al2O3 – SiO2 – B2O3 – H2O – HCl in hydrothermal solutions
and siliceous melts. Can. Mineral. 37, 1025 – 1039.
Wakita, K., 1987. The occurrence of Latest Jurassic – Earliest Cretaceous
radiolarians at the Hida – Kanayama area in the Mino terrane, central Japan.
Jpn. J. Geol. Soc. 93, 441 – 443.
Watson, E.B., 1988. The role of accessory minerals in granitoid geochemistry.
The Origin of Granites. Royal Society of Edinburgh, Royal Society of
London Hutton Meeting, Edinburgh, pp. 19 – 20R abstract.
White, R.W., Powell, R., Holland, T.J.B., 2001. Calculation of partial melting
equilibria in the system Na2O – CaO – K2O – FeO – MgO – Al2O3 – SiO2 –
H2O (NCKFMASH). J. Metamorph. Geol. 19, 139 – 153.
Wilke, M., Nabelek, P.I., Glascock, M.D., 2002. B and Li in Proterozoic
metapelites from the Black Hills, U.S.A.: implications for the origin of
leucogranitic magmas. Am. Mineral. 87, 491 – 500.
Wolf, M.B., London, D., 1997. Boron in granitic magmas: stability of
tourmaline in equilibrium with biotite and cordierite. Contrib. Mineral.
Petrol. 130, 12 – 30.
Yoshida, F., Takahashi, Y., Nishioka, Y., 1995. Geology of the Tsu-Seibu
district. With geological sheet map at 1:50,000. Geological Survey of Japan
(in Japanese with English abst.).
Yoshizawa, H., Nakajima, W., Ishizaka, K., 1966. The Ryoke metamorphic
zone of the Kinki district, southwest Japan: accomplishment of a regional
geological map. Mem. Coll. Sci., Univ. Kyoto, Ser. B 32, 437 – 453.
Zen, E., 1986. Aluminum enrichment in silicate melt by fractional
crystallization: some mineralogic and petrographic constraints. J. Petrol.
27, 1095 – 1117.