1980 Qasim Jan

The Mineralogy and Geochemistry of the
Metamorphosed Basic and Ultrabasic Rocks of the

Jijal Complex, Kohistan, NW Pakistan
by M. QASIM JAN 1 and R. A. HOWIE 2
Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

' Department of Geology, University of Peshawar, Pakistan
2
Department of Geology, King's College, Strand, London WC2R 2LS
(Received 8 November 1979; in revisedform 2 September 1980)
ABSTRACT
The Jijal complex, covering more than 150 sq. km in the extreme north of Pakistan, is a tectonic
wedge of garnet granulites intruded in the south by a 10 x 4 km slab of ultramafic rocks. The
granulites are divisible into plagioclase-bearing (basic to intermediate) and plagioclase-free
(ultrabasic to basic) types, the two types reflecting differences in bulk chemistry. Garnet +
plagioclase + clinopyroxene + quartz + rutile ± hornblende ± epidote is the most common
assemblage. The plagioclase-free rocks are composed mainly of two or three of the minerals garnet,
amphibole, clinopyroxene and epidote. Orthopyroxene occurs in websteritic rocks devoid of epidote.
Much of the amphibole and some epidote appear to be prograde products. Although variation
diagrams do not reveal a genetic link between the two types of granulite, it is considered that they
are comagmatic rather than the products of two or more unrelated magmas.
The compositions of garnet (Py28î6Alm27-43Groi6-28), clinopyroxene (Mg44-34Fe5-nCa5i-49,
A12O3 3-0-9-9 per cent), orthopyroxene (with up to 5-5 per cent A12O3), amphibole (with up to
16-3 per cent A12O3 and high Alvi/Aliv), and the abundance of garnet suggest a high-pressure origin
for the granulites. The rocks appear to have differentiated from a tholeiitic magma of oceanic
affinity or they may be genetically related to the pyroxene granulites of Swat considered to have
originally crystallized from a calc-alkaline magma of island arc or continental margin affinity. They
probably crystallized in the ancient Tethyan crust/upper mantle (or less likely in a continental
margin), later to be metamorphosed to granulites (670-790 °C, 12-14 kb) during the collision of
the Indian-Asian landmasses, and carried upwards during later Himalayan orogenic episodes.
The ultramafic rocks are alpine-type in nature and devoid of garnet. They are dominated by
diopsidites; dunites, peridotites, and harzburgites together form <50 per cent of the area of outcrop.
The chemistry of the rocks, and their olivines (Fo92_89) a n d clinopyroxenes (Mg49.5_48Fe2.g_5.2
Ca47.4_46.8) are similar to those of alpine complexes of the harzburgite subtype. It is not clear
whether they represent a faulted slab of suboceanic crust/upper mantle, mantle diapirs in deep
orogenic roots, or dismembered ultramafic rocks differentiated from a basaltic magma. They seem
to have a complex history; their present mineralogy is suggestive of high grade metamorphism
(800-850 °C, 8-12 kb). They are magmatically unrelated to the garnet granulites and were
probably intruded into the latter as plastic crystalline material after both had been independently
metamorphosed, but before the entire complex was carried tectonically into its present
surroundings. The abundances of the diopsidites is in marked contrast to other alpine-type
complexes and the possibility of Ca and Si metasomatism during or before their metamorphism
should not be totally ruled out.
INTRODUCTION
MUCH of Kohistan and adjacent areas, the northernmost part of the North-West
Frontier Province of Pakistan, is occupied by the rocks of the Kohistan complex.
The latter extends E-W for more than 200 km between Nanga Parbat and eastern
Afghanistan and N-S, along the Swat river, for about 70 km (Jan & Kempe,
1973). The complex consists of three NE-trending belts, consisting of amphibolites,
[Journal of Petrology, Vol. 22, Part 1, pp. 8 5 - 1 2 6 , 19811
86 M. Q. JAN AND R. A. HOWIE
pyroxene granulites, and quartz diorites. In the south the amphibolites have a
faulted contact with Palaeozoic schists. Occurrence along or near this contact of
glaucophane schists, serpentinites and other ultramafic rocks is suggestive of a
major tectonic activity (Coleman, 1971, 1972; Moores, 1973). The fault is
probably an extension of the Indus suture line which, according to Gansser (1974),
marks 'the subduction of the Indian plate and its thrust below the Tibetan mass'.
Along the Indus near Jijal (35° 05' N, 72° 55' E), however, the amphibolites
and Palaeozoic schists are separated from each other by a fault-bounded block or
wedge of garnet granulites and ultramafic rocks, which constitute the Jijal complex.

It covers an area of at least 150 sq. km (Fig. 1); its eastern and western extensions
are not known because of the inaccessibility of the area; however, to the west it
does not extend beyond the drainage divide between the Indus and Swat valleys.
(To avoid confusion the term ultramafic, in this paper, refers to the alpine-type,
garnet-free, rocks. The term is not used for the plagioclase-free garnet granulites,
e.g. garnetites, hornblende-rocks, garnet pyroxenites, epidote-rich rocks.)
The northern part of the Jijal complex is occupied by the garnet granulites and
the southern part by the ultramafic rocks. The latter form a NW-trending and
northerly dipping body of rocks more than 45 sq. km in area (Jan & Tahirkheli,
1969). However, small bodies of serpentinized ultramafic rocks (some containing a
green spinel) also occur in the granulites, especially along the Patan fault, and
xenolithic lenses and patches of the granulites occur in the ultramafic rocks and
along the southern margin of the complex. The small bodies generally have sheared
contacts; the contact between the two main masses, however, is interfingered or
'gradational' (Qureshi & Jan, 1977), possibly because the ultramafic rocks were
intruded as plastic material. The contact relationship is further exacerbated here by
the occurrence of pyroxenites which represent both rock-types and which are
difficult to assign to one or the other type in the field.
This paper presents a detailed account of the geochemistry and mineralogy of
the complex with the help of chemical analyses for 28 rocks and 40 minerals.
Physical constants for most of the minerals are given and the distribution of iron
and magnesium between the co-existing minerals is discussed, along with estimates
of temperature and pressure of formation for the two types of rocks.
Analytical techniques
Disc and Frantz magnetic separators and heavy liquids were used for mineral
separation. Only those rocks were chosen for separation which showed equilibrium
texture and in which the minerals, especially plagioclase if present, showed no (or
only marginal) zoning. Small quantities of rocks, weighing less than 500 g, were
crushed for separation in the hope that uniform conditions of chemical equilibrium
would have been reached in the smaller volume of the chemical system. The
minerals were ground to between 200 and 300 B.S.M. but the garnets were
powdered to <300 mesh; the analysed concentrates, especially garnets, were purer
than 99-5 per cent. Densities were determined by taking about 3-0 g of material in
density bottles; the reported values are at 0 °C and should be accurate in most
THE JIJAL COMPLEX, PAKISTAN 8?
O I
73 00 E
2 Km.
35° OO' N
FIG. 1. Geological sketch map of the Jijal complex. Numbers indicate the locations of the analysed samples.
F = fault.
88 M. Q. JAN AND R. A. HOWIE -
cases to ±0-03. Refractive indices, measured at room temperature, are accurate to

±0-002, and optic axial angles (universal-stage) to ±2°.
Chemical analyses were carried out by combining instrumental and classical
methods and using suitable laboratory standards for reference. SiO2 was
determined gravimetrically; FeO by titration (Wilson, 1966); TiO2 and P2O5
colorimetrically, following Riley (1958) and using a Unicam SP 600 series 2
spectrophotometer; H2O by the modified Penfield method; and F by the rapid
method of Hall & Walsh (1969). The remainder of the elements were determined
by atomic absorption, using emission flame photometric mode for the alkalis.

Trace elements in all the rocks, and major elements (except H2O and FeO) in 14
rocks were, however, analysed by XRF, broadly following the suggestions of
Leake et al. (1969). Pressed tablets, consisting of rock powders and cellulose
binder in a 1:1 ratio, were used and the results extrapolated graphically. Each
quoted figure is the average of four readings, two on each side of the tablet, the
accuracy being ±2 per cent of the actual amounts but MgO and A12O3 are within 3
per cent of the amounts present. The microprobe analyses were made with a
Cambridge Instruments Geoscan at an accelerating voltage of 15 kV and a current
of 0-5 x 10~7 amps; analysed silicates and oxides were used as standards. Results
were corrected after the method outlined by Sweatman & Long (1969) using a
BM/IC/NPL computer programme.
PETROGRAPHY
The garnet granulites

The granulites are characterized in most places by an abundance of garnet and
can be classified broadly into two types: (1) those in which plagioclase is an
important constituent, and (2) those in which plagioclase is either accessory or is
absent. Rocks in the first type make up most of the granulites and are generally
gneissose and medium-grained, whilst those in the second type are non-gneissose
and may locally be coarse-grained (e.g. some hornblende rocks with crystals up to
6j cm long). Porphyroblasts of garnet, up to 9 cm across and in some cases
surrounded by a few centimetres of thick mafic and/or felsic envelopes, are
common. In some outcrops the granulites are cut by feldspathic veins locally
studded with garnet porphyroblasts. Also, the type 1 granulites are locally banded,
and in a few places these bands are pegmatitic. The banding observed along the
Patan stream may be due to shearing along the northern fault, but elsewhere it may
be due to metamorphic segregation or relic layering.
The type 2 granulites are represented by garnet-, hornblende-, and/or
pyroxene-rich rocks and rare (clino)zoisite-garnet—hornblende rocks. They are
scattered throughout the granulites, but are more common along the southern part
of the granulite mass where they form a belt 250 to 600 metres broad (Qureshi &
Jan, 1977). They may either form distinct bodies of their own, up to a few tens of
metres across and considerably longer, or they may occur in confused mixtures:
patches and bands of 'garnetites', pyroxenites, and type 1 granulites occurring in
the hornblende rocks and vice versa.
THE JIJAL COMPLEX, PAKISTAN 89
TABLE 1
Mineral assemblages in the Jijal garnet granulites
Assemblages Mineral phases (+ =: present, 0 = absent)
Typel No Plag Gar Cpx Hbl Epi (etc)
(+,0) 0
+ + + + +
Granulites 1
o + o + +
with 2 0 (+,0)
essential 3 + (+,0)

plagioclase 4
(all + rutile &/or 5 + + + + (Paragonite)
opaques)
Type 2 Garnet-rich types + rutile and other opaque minerals
Granulites 6 (+,0) 0 + 0
with 7 0 0 (+,0) 0 (+,0) (Spinel)
little 8 0 + 0 0 + 0
or no 9 (+,0) + (+,0) (+,0)
plagioclase
Hornblende-rich types, all with rutile &/or opaque minerals
10 0 + 0 0 + 0
11 0 0 + 0 (+, 0) (Spinel)
12 (+.0) + 0 0
Pyroxene-rich assemblages; most have opaque minerals (no rutile)
13 0 0 (+,0) 0
14 0 (+,0) + 0 (+.0) 0 + (OPx)
15 0 0 0 +
(Clino)zoisite-rich types, with minor rutile (sphene) and/or opaque minerals

16 0 (+,0) 0 (+,0) + clz
17 0 (+,0) clz
The 17 mineral assemblages identified in the granulites are listed in Table 1,

assemblage 1, with rutile being the most common. The textures are indicative of
equilibrium (see Jan, 1977, 1979).
Comparison with other areas

The Jijal granulites, according to the subdivision of the granulite facies by Green
& Ringwood (1967), belong to the high-pressure type. This type is much rarer than
the low- and intermediate-pressure types and has been reported from a limited
number of localities.
Comparison of other high-pressure granulites with those of Jijal illustrates two
points: (a) that there is a greater number of mineral assemblages in the Jijal rocks
than in those from other areas, and (b) that the Jijal rocks may constitute the
largest mass of garnet granulites so far described. Although the granulite complex
at Cabo Ortegal, NW Spain (Den Tex et al., 1962, 1972) is equally large, the
garnet-clinopyroxene-plagioclase assemblage in that area is less extensive. The

two areas have remarkable similarities in some respects; many of the assemblages
are common to the two areas, although in Spain the epidote- and horn-
blende-bearing assemblages are more common, epidote is more abundant, and
garnetites have not been reported. Both areas have ultramafic rocks associated with
the granulites; in Spain pelitic assemblages and eclogites also occur.
The ultramafic rocks

The ultramafic rocks are mostly represented by diopsidites; dunites, harzburgites
and websterites make up less than half the outcrops, whilst orthopyroxenites and
lherzolites have not been found. There is no clear distinction of top, bottom, or
chilled margins and the contacts between the various units may be sharp or
gradational (Qureshi & Jan, 1977); the pyroxenites locally contain lenses of dunites
with sharp contacts. In places the harzburgites contain trails of coarse crystals of
pyroxene, giving the impression of thin, crude banding. Streaks and bands, up to 2
cm x 5 m, of chromite also occur locally in the harzburgites and dunites. They are
generally folded, some of the bands variously squeezed and, locally, tapering
suddenly at the ends. The general trends of the chromite as well as pyroxenite
streaks are at high angles to the NW trend of the ultramafic mass (and its main
contact with the granulites). Thus, if the spinel concentrations (Thayer, 1970) and
coarse pyroxene streaks represent some kind of original structure (? stratification),
then the ultramafic rocks are not in structural harmony with the granulites. Plots of
500 joint poles for the two rocks also show different patterns (Qureshi & Jan,
1977).
The rocks are medium- to coarse-grained and allotriomorphic. Deformation is
illustrated by the frequent fractures, non-uniform strain extinction, and hour-glass
texture and kink bands in pyroxene and olivine grains. Some of the rocks are
intensely granulated and have a mortar texture. The pyroxenites may or may not
contain olivine. Chromitic spinel of a brownish black colour is a typical accessory
in the olivine-rich rocks. Most of the rocks contain magnetite formed during
serpentinization; it is not clear whether there is any primary magnetite.
Serpentinization is selective, probably due to differential shearing, even on the scale
of a thin section, and the rocks are only mildly altered except where strongly
sheared.
The abundance of the diopsidites is in marked contrast to other alpine-type
ultramafic complexes, where peridotites and dunites are the common rock-types
(Jackson & Thayer, 1972). Detailed work is needed before a final assessment can
be made of the relative abundance of the diopsidites; however, it is possible that:
(1) the Jijal rocks are either a dismembered part of a much larger 'normal' alpine
ultramafic mass; (2) olivine-rich rocks are present at depth or in those parts of the
complex which have not yet been studied in detail; or (3) the diopsidites are a
product of large scale Ca and Si metasomatism. Coarse-grained clinopyroxenite
veins up to a few centimetres thick, found along Duber stream in a harzburgite
mass, are an example of the latter phenomenon.
The characteristics of alpine-type ultramafic rocks have been reviewed by many
authors. Tectonic fabrics, Mg-rich olivine and pyroxenes (Mg/Fe ~9/l), common
concentrations of Cr-spinel (Thayer, 1960, 1967; Jackson & Thayer, 1972), and
little compositional variation in the rocks (Moores, 1969) are among their special
features. The Jijal rocks share all these characteristics; they are devoid of
plagioclase, micas, and primary amphibole. Den Tex (1969) subdivided the alpine
peridotites into two genetic classes, the ophiolitic or truly alpine-type, and the
root-zone peridotites found in very high-grade metamorphic terrains in deeply
eroded mountain roots. The Jijal rocks probably should be placed in the second
group.

GEOCHEMISTRY
The garnet granulites

Analyses of 18 granulites presented in Table 2 show that the rocks are ultrabasic
to intermediate in composition. The presence or absence of plagioclase is essentially
a function of bulk chemistry; plagioclase seems to have disappeared more readily in
the more basic rocks (those with SiO2 <45 per cent and Na2O <1 per cent except
in one (?)metasomatized rock, SI 289) than in the less basic rocks with higher silica
and soda (cf. Kushiro & Yoder, 1966; Green & Ringwood, 1967). Because the
plagioclase-free rocks are more abundant in the south and the three most silicic
rocks come from Patan and to the north, it appears that the granulites as a whole
become more silicic towards the north.
A characteristic feature of the rocks is their poverty in K, Rb, Cr and Ni
compared with most igneous rocks of equivalent silica content. K2O is less that 0-1
per cent in the plagioclase-free rocks (except 289) and increases to 0.-3 per cent
only in the plagioclase-bearing rocks. Although the possibility of some K, Rb (?
and Na) having been lost to the upper levels during high-grade metamorphism
(Sighinolfi, 1971) cannot be excluded, the rocks may well have been poor in these
elements to begin with.
The plagioclase-bearing basic granulites are chemically closer to continental
basalts than to other types but their Na/K ratios are like the oceanic basalts (Engel
et ai, 1965; Bryan et ai, 1976). On Kuno's (1968) diagrams of (Na2O + K2O) vs
A12O3 and SiO2, they fall mostly in the field of tholeiites rather than in that of
high-alumina basalts. They also have similarities to some rocks from Skaergard
and the Bushveld, and to certain norite and basalt analyses. The plagioclase-free
rocks do not have chemical analogues in the common volcanic rocks; however, the
ultrabasic and basic granulites of Jijal resemble closely the rocks of the Huntly
region (Weedon, 1970), except for the significantly higher Cr 2 O 3 in the latter. The
three most silicic rocks of Jijal (SiO2 >54 per cent) depart from the common
volcanic rocks of similar SiO2 content by their relatively high CaO, low alkalis, and
very high Ca/alkalis ratio. However, scattered examples of plutonic,rocks with
somewhat similar chemistry can be found in the literature. It cannot be ascertained
whether the rocks represent a mixed plutonic-volcanic or an entirel;,' plutonic
association. However, the lack of common volcanic equivalents for the rocks
represented by the ultrabasic and intermediate analyses and the apparent genetic
^<ôoQ' i < irroo©<NO «o r*i © r*i »*v io O ^ *O (N
•O — ' T ' T ' — © — f - 00 II
II
' © ^ r£
to
r- — oo—•
Ooooov. NOVJOI— o o o o

V V M V -"
CT\ *o —
O - • 0O \
8 j _ O — ' O
«o ' t •** «o I
i — — r-1 — I
v v— v
# i~- • £ - ^ a> - v <N — • i — «N I —
O r^
•I O — O— t w - i O O O — O
t r - T ^ m ^ - — — — — o»"i
I I 1 ^ 1 I
s V "» V
r*j —
a^ o o «O 00
o — , - — — 6
43
18
V — V V
—
ÔfOOfNI^MNOtD-O I N N jOOOOsTf^r- —
OrÔOOOOOOO
r-—,,*! — ^ - — (N — C * ~
1
I* , IO <N O !
OQ a
i I
"? I 9
i — I © O — rMr- — *
rîrÔf~~fN
-66^-
V *N
sc f\
O\ 8: > O L.
k. *N
8 © CN «O 6 — O0 <
5)
i-co©< sr > — r - © f - . ô>o i —•
iOooo<-nr-mrnO©Oi_ i *e
s:
a
"a
**» f-vÔ'o — O
S OO - — O O — OO
© r*> r~- •
i (N O * N X
— V "" V V
^ » * i t o o — ^rr— ov»o©'«
O ^ r ^ ' r - j — — r— •*!•(— rN
- OOOOO* - — so i n
v vv v
00 »O O 00 w-i ©
I 00
I — - I I in
O.
* S ^ g J2 £ t
om http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015
mt 8-63 3-77 117 4-44 415 3-31 0-68 5-31 0-71 512 001 2-87 2-90 4-63 6-26 2-77 1-62 1-45
il 2-35 0-42 0-61 0-38 0-42 0-42 0-30 118 0-32 017 213 1-18 1-31 101 103 203 110 1-20
ap — — 014 007 — — 009 — 0-28 007 109 0-38 0-33 0-31 005 0-62 019 0-36
Niggli values
si 76-7 791 80-8 790 79-2 79-8 86-2 85-1 900 935 99-6 113-9 120-3 115-8 121-0 146-5 136-8 1801
al 18-3 19-6 18-4 20-9 18-7 190 281 11-7 18 5 3-7 25-9 30-9 26-9 26-7 27-9 281 27-4 31-3
fm 540 57-5 60-3 54-9 581 50-7 42-7 49-5 47-4 76-6 41-8 381 41-9 41-2 39-3 40-5 37-7 33-6
c 22-6 220 20-3 23-4 22-7 29-8 28-1 37-8 33-1 19-2 26-6 261 26-9 26-6 27-3 260 26-8 261
alk 51 0-9 1-0 0-9 0-6 0-6 1-2 10 10 05 5-7 4-9 4-4 5-5 5-5 5-4 80 90
ti 1-77 0-31 0-46 0-28 0-31 0-31 0-24 0-86 0 25 012 1-82 1-09 1 26 0-94 0-99 218 110 1-44
p — — 005 002 — — 003 — 010 002 0-42 0-16 0-14 013 002 0-30 009 019
k 007 006 009 004 006 008 000 002 008 009 007 003 007 002 003 006 006 005
mg 0-62 0-55 0-65 0 71 0-65 0-73 0-74 0-73 0-85 0 81 0-38 0-47 0-49 0-47 0-49 0-38 0-50 0-46
-i
. a
Locations of the analysed rocks in Tables 2 and 3 are given on the map (Fig. 1). The XRF analyses are distinguished from wet chemical analyses by the absence of an OH" value for the rn
former. ,_
Columns SI 289 to SI 297a represent granulites with little or no plagioclase whilst SI 326 to SI 330 represent granulites with essential plagioclase. "
SI 289. Garnetiferous hornblende rock' containing 90-8 per cent ferroan pargasite, 4-8 per cent garnet (Gar). 1-6 per cent epidote (Epi), 1-2 per cent opaque minerals (Ore), 0-8 per >
cent rutile (Rt). and 0-8 per cent clinopyroxene (Cpx), quartz (Qz), chlorite (Ch), and sphene (Sph). The latter three and Epi are retrograde. The sample is traversed by a thin felsic vein t~
with traces of a white mica. Density (D) = 3-27. ^
SI 292. Clinopyroxene 'garnelite' consisting of 71-3 per cent Gar, 20-7 per cent Cpx, 7-4 per cent amphibole (Amph, brownish green and bluish green varieties), and 0-7 per cent Ore, O
Epi. Ch. Sph, Rt. The bluish green Amph. Epi. Ch and Sph are retrograde. D = 3-52. 5*
SI 317. Granulite composed of Gar. Hbl. Epi. with small amounts of (?) margarite. clinopyroxene. plagioclase (Pig), Ore, Ch. Marg, Ch, some Epi and (?) Hbl retrograde. D — 3-36. *ta
SI 283. Granulite containing Gar. Amph, Cpx, with small amounts of Pig, Rt, (?) Qz. Ch, Epi and carbonate along fractures, mainly in Gar. D = 3-49. *"*
SI 297b. Granulile containing Gar, Cpx. Hbl (brownish) with small amounts of Ore and Rt. Epi, Ch. Sph and an amphibole are the retrograde products. D = 3-51. Pj
SI 297c. Granulite containing Gar. Cpx. pleonaste and Amph. A lot of secondary Epi, Ch, (?) Serp. Amph. Spinel surrounded by a complex corona structure which has been made com-
plicated by alteration. *g
SI 285. (Clino)zoisite rich granulite also containing Gar, Amph, Cpx and minor Qz, Ore, and Ch. Most (probably all) of the hydrous minerals retrograde. D = 3-34. >
SI 270. Garnet clinopyroxenite consisting of 77 1 per cent Cpx, 16-8 per cent Gar, 5-7 per cent Amph, and 0-4 per cent Ore. Opx also present in some parts of the outcrop sampled. ?^
D - 337. £
SI 299. Granulite (light grey in colour) containing Cpx. Gar (C)zoi. Amph, Ch, (?)Plg: the last four may be entirely retrograde. D = 3-20. H
SI 297a. Garnet free websteritic granulite composed of Opx. Cpx and abundant secondary talc, serpentine (?) tremolite and minor magnetite. The three samples. SI 297a, b, c, are from a >
granulite xenolilh in ultramafic rocks. For further details, see text under spinels. ^
SI 326. Granulite containing 32-5 per cent Amph. 30-6 per cent Gar. 22-7 per cent Pig. 5-8 per cent Epi, 4 5 per cent Cpx, 2-9 per cent Qz, 0-6 per cent Rt and 0-3 per cent apatite.
Pig complexly twinned and zoned, Epi also zoned. D - 3-20.
SI 332. Granulite consisting of 46-6 per cent Pig. 25-6 per cent Gar. 17-1 per cent Cpx, 2-5 per cent Epi, 1-8 per cent Amph, 3-8 per cent Qz, 2-3 per cent Ore and 0-3 per cent Ap.
Gar contains Rt needles. Epi, and some Qz and Amph retrograde. D = 3-2. Contains 0-23 per cent C in the norm.
SID 50. Granulite composed of 46-8 per cent Pig, 2 5 1 per cent Gar, 21-3 per cent Cpx, 4-6 per cent Qz, 1-3 per cent Ore, 0-6 per cent Epi (retrograde), 0-2 per cent Rt, and 0-2 per
cent Ap. Secondary Amph after Cpx and Sph/leucoxenc after Ore. D = 3-29.
SI 290. Granulite containing 45-5 per cent Gar, 24-7 per cent Pig, 13-4 per cent Cpx, 10-4 per cent Qz, 5-2 per cent Epi, 0-6 per cent Rt, and 0-2 per cent Ap. Most Epi primary but
some retrograde, as is minor Ch. D —3-13.
SI 291. Granulite consisting of Gar, Amph, Pig (An 24), Epi, Qz, paragonite, Rt and Ap. Paragonite is retrograde but Amph and Epi look primary. D = 3-27.
SI 328. Granulite composed of zoned Pig (andesine). Gar, Cpx, Qz and minor Rt, Hbl and hypersthene. Hbl may be retrograde but Hy is in unstable relics. D = 3-16.
SI 338. Granulite containing 39-2 per cent Pig (An 40 ), 28-3 per cent Gar, 19-2 per cent Cpx, 12-4 per cent Qz, 0-3 per cent Rt, 0-2 per cent each of Amph, Epi, Ore and traces of Ap.
D = 319. ^
SI 330. Sheared and banded granulite containing Pig, Gar, Cpx, Qz, Amph, Rt, Ore, Epi; the later being retrograde. D = 2-95. <••»
relationship between the plagioclase-bearing basic and intermediate rocks (close

proximity and chemistry, e.g. Fig. 3) suggest that the entire mass of the granulites is
plutonic.
The extent to which CIPW norms represent modal composition is debatable.
However, the more basic rocks of Jijal (SiO2 <47 per cent) seem to have been
essentially composed of highly calcic plagioclase and magnesian pyroxene(s) and
olivine (troctolites and olivine gabbros), with minor pyroxenites (with or without
plagioclase and olivine) and feldspathic allivalites (SI 285). Four of these rocks are
nepheline normative (ne 0-2-2-1); the notably higher ne (8-84) in the hornblende

rock (SI 289) might be a consequence of metasomatism. The plagioclase-bearing
granulites (SiO2 >48 per cent) were, probably, quartz-hypersthene gabbros and
diorites. Analysis 332 has 0-2 per cent corundum in the norm due perhaps to a
relative overestimation of A12O3 or to alteration.
Variation diagrams
Oxide percentages have been plotted against the Thornton & Tuttle (1960)
differentiation index, DI = sum of weight per cent normative Q + or + ab + ne + Ic
+ ks, in Fig. 2. With increase in DI, the plagioclase-bearing rocks show an increase
in SiO2 and total alkalis (essentially sodium), and a decrease in A12O3, Fe oxides,
CaO, and MgO. Such trends cannot be seen in the plagioclase-free rocks because
of their restricted DI range due to low alkalis and silica. Interestingly, however,
there is a gap in the DI values between 5-41 and 18-95 (ignoring sample 289, with
DI = 16-26, which is clearly behaving independently due to (?)metasomatism).
This lack of intermediate DI values could arise from a lack of analyses of truly
representative rocks; nevertheless, the possibility of two magmas cannot be ruled
out.
Fig. 3 shows the plot of several cation percentages against (j Si + K — Ca —
Mg). Here also Na and Si increase and Ca, Mg, and Al decrease with increase in
the index for the plagioclase-bearing granulites. The plagioclase-free rocks behave
similarly in the case of Ca and Mg but have an almost perpendicular trend for Si
and (?) Al. The two types of rock also seem to have opposed trends for the other
elements despite the fact that this type of variation diagram produces smoother
trends than most other variation diagrams.
Some of the trace elements have also been plotted against (j Si + K — Ca — Mg)
in Fig. 4. Whether any conclusions can be drawn from the distribution of trace
elements is debatable; significant differences in the pre- and post-metamorphic
modal composition between the two types of rocks, lack of suitable laboratory
standards, and high-grade metamorphism are some of the main factors which could
render such conclusions invalid.
The low Cr and Ni contents and the different variation trends for the two types
of rocks raise a serious question: are the plagioclase-free granulites really
meta-igneous?
The Cr and Ni values for the granulites are lower than those reported for
basic-ultrabasic rocks of somewhat similar composition. Cr, apart from making its
• FeO
12 *
•
o o 2 . Ti O2
8 • o o 1 O O
CD O O O O
o O
4
v •
- •• 3 . Na 2 O+K 2 O #
<9 O
O O
18 • • (Fe 2 O 3 HhFeO) 2 - °O O
14 - • • •
•
1
H
10
•
CO <d o o
O
X
m
•
o
IUAL
6
18 • Ca O
I t o •
22 AI2O3 16 O
O
20 o o 14 2
o o
PLEX.
18 O° o 12
10 CO 8 O
O
o O 13
AKIS
55 oo 14 " < % Mg 0
H
12
50 o 10 *•
• (P o Si O 2
8 •
0
o
45 6
o°
•
0 O O
4 O O
1 • 1 1 I 1 1 1 1 1
0 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
DI DI
FIG. 2. Plots of oxide percentages against differentiation index (DI = normative wt. per cent Q + or + ab + ne + Ic + ks) for the plagioclase-bearing (open circles) and
plagioclase-free (filled circles) garnet granulites.
28
26
o o
24
Si
22
20

13
12
11
Al
10 o o
O O
9
8 t i
6
Mg
4
14
12
10
Ca
8
Na
in '**'
-14 -10 -6 -2 0 +2
( j S i + K - Ca - M g )
FIG. 3. Plots of Na, Ca. Mg. Al. and Si against (jSi + K - Ca — Mg) for the plagioclase-bearing (open circles)
and plagioclase free (filled circles) garnet granulites.
own compounds, is preferentially accommodated in pyroxenes, whilst Ni enters

olivine; both of these elements are strongly withdrawn in the earlier phases of
magmatic crystallization. Thus, one would expect higher values of these elements in
the Jijal granulites, especially Ni since the rocks contain up to 38 per cent
normative olivine. It seems that the source material which produced the Jijal rocks
was, for some reason, depleted in Ni and Cr.
The possibilities of metasedimentary, or metasomatic origin (the latter being
popular with some Soviet petrologists when discussing the origin of certain

Gros +Andr
Spess
FIG. 4. Plots of the analysed garnets from Jijal on a (almandine + spessartine) — pyrope — (grossular +
andradite) = 100 diagram showing the fields of garnets from eclogites (1), granulites (2), amphibolites (3), and
epidote amphiboles (4) (after Dobretsov et ai, 1972).
eclogite-like rocks, e.g. Pinus et ai, 1970) were explored using Niggli's approach as
recommended by Leake (1964). It is commonly agreed that in basic magmas a
systematic decrease with increasing differentiation occurs in the Niggli numbers mg
(mg = Mg/(Mg + Fe tota , + Mn) and fm, and a systematic increase in si, al, alk, and
ti. By plotting these values in various ways, the trends generated by igneous rocks
are significantly different from those given by the mixtures of sediments (in this
case the appropriate combinations of pelitic rocks and limestones/dolomites)
which can duplicate the chemistry of basic and ultrabasic rocks. Metasomatic
rocks, on the other hand, are expected to show scattered plots on these variation
diagrams. A few of the plagioclase-free rocks may have been metasomatized
(epidote, sericitic mica, amphibole and chlorite are common as retrograde
products), but the effects of metasomatism are not pronounced except in the
garnetiferous hornblende rock (? and other hornblende-rich rocks). The few
departures from igneous trends, except in the hornblende rock, seem to be due
essentially to bulk mineralogy.
The following values were used for plotting: mg against c, ti, Ni, Cr, and si; c
against si, al, and al-alk; and a 100 mg + (al —alk) + c = 100 triangular plot.
Although in some cases the rocks fall in the overlapping fields of igneous and
metasedimentary rocks, it can be concluded on the following grounds that the Jijal

granulites, including the plagioclase-free type, are meta-igneous: (1) they generally
follow the igneous trends on various diagrams; (2) they lack associated calcareous
or pelitic rocks; (3) their normative compositions are not abnormal for igneous
rocks and they have chemical similarities with some magmatic rocks; (4) the
plagioclase-free rocks occur as local masses throughout the complex; and (5)
sporadic occurrences of xenoliths have been noted in the plagioclase-bearing type.
Taken as a unit, the Jijal granulites show an increase in Sr, Zn, and Zr, and a
decrease in Ni and, less obviously, in Cr with increase in the index. The variations
in Ba, Co, Cu, and Ga, however, seem to be independent for the two types of
granulites. For Co, the plagioclase-free rocks show a sharp rise and the plagioclase-
bearing rocks a sharp decline as the index increases. The variation in Ba, Cu, and
Ga can be interpreted in more than one way. The behaviour of trace elements,
therefore, cannot discriminate between the rival hypotheses of comagmatic or
independent origin for the two types of granulites.
The variations in Ba and Sr merit comment. Ba is generally assumed to be
captured in K-bearing minerals, but in the Jijal rocks K2O is typically low and there
is no potassium mineral. Yet the distribution of Ba is different for the two types of
granulites and it is not clear which major element it follows. Such breaks in the
distribution of trace elements in the Huntly region were considered by Weedon
(1970) to be of petrogenetic significance.
Sr is normally considered to substitute for Ca or even K; it is usually more
concentrated in apatite, epidote, and plagioclase than in other common calcic
minerals. In Jijal, the Sr content is generally very low in the plagioclase-free
granulites. Out of the three rocks with appreciable Sr content, one is the possibly
metasomatized rock (SI 289) while the other two are considerably altered,
containing retrograde epidote. Whether or not Sr has been added to the three rocks
from an unknown source is difficult to tell; ignoring these three, the values of Sr for
the two types of granulites are markedly different.
On the basis of chemical variations alone, it is hard to suggest that the
plagioclase-bearing and the plagioclase-free granulites are comagmatic. However,
chemical breaks on variation diagrams between basic rocks and their comagmatic
ultrabasic rocks are not totally unexpected. The intimate association of the two
types of granulite can be better explained if they are regarded as comagmatic rather
than being the products of two or more unrelated magmas. There is a likelihood
that the granulites represent a metamorphosed sequence of hypersthene gabbros
with some more and some less basic rocks which differentiated from an
alumina-rich, (?) alkali-poor tholeiitic magma.
The ultramafic rocks

Analyses of the representative ultramafic rocks (Table 3) show that the principal
variations are in the values of SiO2, A12O3, FeO, CaO, and MgO due to different
TABLE 3
Chemical and XRF analyses of the ultramafic rocks from the Jijal complex
SI 318 SI 278 SI 275 SIJ 76 SI 272 SID 52 SI 294 SIJ 81 SI 275a SI 295

SiO 2 40-63 42-20 44-50 5012 50-21 50-32 50-40 51-05 52-92 53-75
TiO 2 006 ' 009 - 012 0-23 011 015 012 009 010 0-23
A12O3 0-45 0-49 0-41 3-81 0-83 2-30 1-65 1-45 0-78 1-49
Fe 2 O 3 8-95 4-35 2-78 0-29 1-36 0-27 1-33 0-94 113 2-17
FeO 9-14 5-54 5-62 5-70 3-52 5-76 605 4-52 1-87 4-40
MnO 0-30 0-15 0-14 015 011 012 016 015 010 018
MgO 38-29 44-69 44-86 21-48 26-29 25-23 25-50 2608 19-00 21-63
CaO 0-82 1-39 0-72 17-56 1610 15-25 14-88 15-42 21-79 14-55
Na 2 O 010 003 009 015 0-20 015 019 0-28 0-23 016
K2O 0-04 002 0-01 0-04 0-01 0-02 004 002 002 001
H 2 O+ 1-07 0-11 0-32 1-79 1-06 2-29 1-57 1-82 1-01 0-42
HjO— 019 008 006 — — 012 019
P 2 O, 000 000 000 007 002 010 009 008 000 000
TOTAL 99-85 99-25 99-65 101-4 99-88 1020 1020 101-9 99-07 99-18
PPM
Co 355 137 145 23 28 64 48 27 <10 28
Cr 191 5900 3445 2873 2990 1746 2206 2819 3080 2835
Cu <30 137 55 <30 77 <30 <30 39 <30 59
Ga <10 <10 <10 <10 <10 <10 <10 17 <10 <10
Ni 545 1736 2483 206 433 282 273 235 240 210
Rb 10 11 <10 <10 12 <10 10 10 13 <10
Sr <15 <15 <15 24 <15 20 22 30 20 20
Y < 10 <10 < 10 < 10 < 10 < 10 < 10 18 < 10 < 10
Zn 60 47 40 <30 150 <30 <30 <30 40 60
Zr
CIPW norms
Q — — — — — — — — — 1-42
or 0-24 012 0-06 0-24 006 0-12 0-24 0-12 012 006
ab 0-85 0-25 0-76 1-27 1-69 1-27 1-61 2-37 1-95 1-35
an 0-66 114 0-69 9-60 1-34 5-54 3-53 2-64 1-04 3-32
wo-v 1-42 2-40 1-20 32-17 32-73 2900 2910 30-61 44-69 28-75
en >di 1-14 1-98 0-99 24-30 26-63 22-27 22-43 24-28 37-12 22-83
fs J 012 012 007 4-60 2-18 3-66 3-57 2-86 1-98 2-65
en \ . 24-01 15-28 23-62 4-95 7-05 9-93 1119 1010 5-35 31-02
fs ; h y 2-48 0-92 1-71 0-94 0-58 1-63 1-78 119 0-29 3-59
fo \ , 49-18 65-88 61-02 16-98 22-27 21-45 20-93 21-41 3-39
fa) 01 5-60 4-38 4-37 3-54 201 3-89 3-67 2-78 0-20 —
mt 12-98 6-31 403 0-42 1-97 0-39 1-93 1-36 1-64 3-15
il 0-11 0-17 0-23 0-44 0-21 0-28 0-23 0-17 019 0-44
ap — — — 017 005 0-24 0-21 0-19 — —
SI 318. Moderately altered dunite containing olivine, magnetite, serpentine and chromitic spinel.
SI 278. Mildly altered harzburgite consisting of olivine (Fo 90 ), orthopyroxene, chromite, magnetite, serpentine and
(?) talc.
SI 275. Harzburgite composed of 86-5 per cent olivine (Fo,j.,), 10-1 per cent enstatite, 1 0 per cent diopside, 1 • 1 per cent
serpentine, 0-45 per cent talc, and 0-87 per cent of chromite with minor magnetite.
SIJ 76. Pyroxenite containing diopside, (?) tremolite (talc), serpentine, orthopyroxene, chromite, magnetite. The
amphibole and/or talc seem to replace the serpentine.
SI 272. Olivine clinopyroxenite containing also serpentine, talc, (?) orthopyroxene, chromite and magnetite. Olivine
(Fo,,).
SID 52. Similar to above but also containing a carbonate, no orthopyroxene.
SI 294. Similar to 272 but without orthopyroxene.
SIJ 81. Similar to 294 but also containing a little tremolite which may be later than serpentine.
SI 275a. Coarse-grained clinopyroxenite vein in harzburgite 275, composed of 86-4 per cent diopside, 12-5 per cent
serpentine, and 1-2 per cent chromite and a little magnetite.
SI 295. Websterite containing diopside. magnesian bronzite, chromite, serpentine, talc. (?) amphibole. and magnetite.
proportions and, to a lesser extent, the composition of the three principal

minerals—clinopyroxene, olivine, and orthopyroxene. Ti, Al, and P are higher, and
Ni and Co lower in the pyroxene-rich rocks than in the olivine-rich rocks,
suggesting a mineralogical control over their distribution.
The variations in Fe2O3 and H 2 O + are due to the degree of serpentinization so
that if the amount of Fe 2 O 3 is recalculated to FeO, then the mole ratio MgO/(MgO
+ FeO total + MnO) can be compared for the various rocks (Coleman & Keith,
1971; Loney et al., 1971; Himmelberg & Loney, 1973). This ratio for some alpine
peridotites and dunites from Oregon and California varies generally from 0-90 to

0-92. The values of this ratio for the Jijal rocks are: SI 318 = 0-79; 278 = 0-90;
275 = 0-91; 76 = 0-86; 272 = 0-91; 52 = 0-88; 294 = 0-86; 81=0-90; 275a =
0-93; and SI 295 = 0-86. Sample 318 has a considerably lower value for this ratio;
it also has higher Fe2O3, lower Cr and, compared to the other two olivine-rich
rocks (278 and 275), lower Ni contents. It contains a higher proportion of
magnetite along serpentinized fractures than would be likely to form by isochemical
alteration of Mg-rich minerals. There is a likelihood that some iron has been added
to such a rock, either due to redistribution within the rock body or introduced from
an external source by the incoming water.
The composition of the clinopyroxenite vein (275a) is very similar to that of its
host harzburgite (275) for minor and trace elements (ignoring Ni), and for the
Mg/(Mg + Fe + Mn) ratio; the essential difference being in CaO, SiO2 and, to a
lesser extent, in H 2 O + and A12O3. Thus it is probable, as is also suggested by
petrographic evidence, that the pyroxenite veins have been produced by the action
of Ca- and Si-bearing aqueous solutions on the host rock at temperatures above the
stability field of serpentine {i.e. above about 500 °C; Moody, 1976). A somewhat
similar origin (Si-metasomatism) has been suggested by Barth (1962) for the
formation of enstatite veins (and larger masses) in some peridotites.
The two peridotite analyses (278 and 275) compare well with that of average
alpine peridotite (Velinskiy et al., 1970). The Jijal rocks are dominated by
diopsidites, a feature not commonly found in the harzburgite-subtype alpine
complexes (Jackson & Thayer, 1972) to which the Jijal rocks seem to belong. The
lherzolite subtype complexes (the high-temperature peridotites of Green, 1964)
contain higher A12O3 in the whole rock, as well as in the spinel and pyroxene
analyses, than in the Jijal rocks. According to Zoloyev's (1970) classification, only
one rock (295) should be called pyroxenite; the others would be olivine pyroxenites
and peridotites. The analyses suggest also that the Jijal rocks are 'barren
peridotites' not capable of producing basaltic magmas on partial melting.
A comparison of the analyses of the Jijal ultramafic rocks and granulites
suggests that they are probably not genetically related; the granulites have
distinctly lower Cr and Ni contents even when ultrabasic in composition. The
major element content of the garnet-free pyroxenite member (297a) of the
granulites is similar to those of the pyroxene-rich members of the ultramafic rocks.
However, the former has distinctly lower molar MgO/(MgO + FeO total = MnO)
ratio (0-81), Cr and Ni, and higher A12O3 and Co. A lack of comagmatic
relationship is further suggested by the distinct differences in their clinopyroxenes.
MINERALOGY
Plagioclase. Plagioclase is present only in the granulites. In many rocks it is
cloudy or saussuritized and full of tiny prisms of (clino)zoisite, but in a few rocks it
may also contain patches of an epidote with intergrown quartz. White mica, quartz,
and albite are among its alteration products. In some rocks where the feldspar is
fresh, its composition, based on maximum symmetrical extinction angle on albite
twins, is from An24 to An45. Twinning is usually poorly developed and is on albite
and/or pericline/acline laws. In some rocks the plagioclase appears to lack

twinning almost completely whilst in a few it is complexly twinned on Carlsbad +
albite + pericline/acline laws (sample SI 326).
Olivine. The occurrence of olivine is restricted to the ultramafic rocks. The
compositions of six olivines were determined optically; one of these was also
chemically analysed (Table 4, anal. 1). The composition (Fo92.5) of the olivine from
harzburgite 275, determined by plotting cron the curve given in Deer et al. (1962a),
TABLE 4
Analyses of an olivine and spinels from the Jijal complex
/ 2 3 4 PPM 1 2 3 4
SiO2 40-90 003 007 Co 140 441 705 630

TiO2 0-22 — 0-02 0-03 Cr 56 — 547 684
A1A 0-01 13-40 6303 63-19 Cu 5 105 — —
Cr 2 O, — 47-28 — — Li 1 1 — —
Fe 2 O 3 1-67 8-50* 3-50 3-50 Pb 5 110 — —
FeO 7-51 2118 1402 13-93 Ni — 663 786 786
MnO 013 0-25 006 009 V 5 884 450 450
MgO 48-83 7 91 17-64 18-26 Zn 35 1222 — —
CaO 0-17 0-32 000 0-00
99-91 98-84 98-30 99-07
Number of ions on the basis of four (anal. I) or thirty-two oxygens

1 2 3 4
Si 0-999 0006^ 0-014"!
Al — 4-235 I 15-485 15-397
Cr 10-000l,sq4q — H6-042 — M5-960
Fe 3+ 0031 1-714 f 0-548 0-544
Ti 0 004 — J 0-003J 0005J
Mg 1-778 •1 982t 3 160 5 478 5 625
Fe
' ] ' ] ' ]
0153 2443 2 408
Mn 0003 4-751 I I 7-931 - 1 8-049
Ca 0004 0056 f 8 0 5 4 0010 f ' y J 1 0-016 f a U 4 y
0-087 J —J — J
1. Forsterite from harzburgite (SI 275), Duber stream, Jijal complex (Total includes 0 0 1 per cent each of
Na 2 O, K2O and 0-45 per cent NiO). a 1-648, D = 3-331; Mg = 92-1. Analyst: M.Q.J.
2. Chromite from SI 275. Analyst: M.Q.J.
3. Green spinel from garnet-spinel-clinopyroxene-amphibole lens (SI 297c) in Jijal ultramafic rocks.
(Microprobe analysis by R. F. Symes).
4. Dark-coloured spinel associated with the green spinel in SI 297c. (Microprobe analysis by R. F. Symes).
* Total iron was determined as Fe 2 O 3 in the spinel analyses. The amounts of Fe 2 O 3 and FeO was re-
calculated to give approximately the spinel R 2 + : R3+ ratio of 1:2, following Deer et al. (19626).
t Includes 0 0 0 9 Ni.
matches closely that determined from the analysis (Fo92.j); however, the value
(Fo86) .derived from the density is markedly low in Mg. Most of the Mg-rich
olivines plotted by Deer et al. (1962a) also give lower Fo contents on the density
curve.
The variation in the composition of the Jijal olivines is small (Fo92.5 to Fo89) and
within the range (Fo 94 -Fo 87 ) displayed by the olivines of alpine peridotites (Green,
1964; Challis, 1965; Jackson & Thayer, 1972). Olivines from peridotites in layered
complexes may be as magnesian as those in alpine peridotites but the range in

composition of the former is greater, extending to Fo g0 (Loney et al, 1971) and
below (Jackson & Thayer, 1972). Il'vitskiy and Kolbantsev (1968), however,
report comparatively closer ranges in the composition of the olivines (Fo92.6_90.5
and Fo93.7_84.5)from alpine and layered complexes, respectively.
In some alpine complexes, olivine is less magnesian in pyroxenites than in
olivine-rich rocks but such a tendency is not apparent in Jijal. Similarly,
Himmelberg & Loney (1973) found that in the Vulcan Peak peridotite, Oregon,
chromitites (and a peridotite with a higher than usual amount of accessory
Cr-spinel) contain the most magnesian olivine. Olivine compositions have not been
determined from the 'chromitites' of Jijal; however, no sympathetic relationship is
observable between the Fo content of the olivine and the Cr content in the whole
rock analysis.
Spinel. Analyses of three spinels are given in Table 4. One of these (column 2) is
from harzburgite 275 and the other two from a granulite. In all three, total iron was
determined as Fe 2 O 3 and FeO was recalculated to give an approximate R2+:R3+
ratio of 1:2. According to Deer et al. (19626) and Irvine (1967), natural spinels are
usually fairly close to the ideal R1+R\+ formula and the recalculation seems
justified.
The analysed chromite has a dark brown colour when it occurs as very thin
grains; most of the grains, however, are almost black in thin-section. The Fe
content of the chromite is higher than in most chromites from alpine complexes
(Thayer, 1970), including those of Hindubagh (Bilgrami, 1968) and Malakand
(Ahmed, 1978) which also have much higher MgO. However, occasional analyses
of alpine chromites with high Fe can be found in the literature. The chromite in
dunite (4-VP-68) from Vulcan Peak (Himmelberg & Loney, 1973), for example, is
very similar to the Jijal chromite, whilst Ehrenberg (1975, Table 1, anal. JJ) has
reported a chromite from Sierra Nevada with 43-1 per cent FeO. The relatively
high CaO content of the Jijal analysis may be due to contamination resulting from
the repeated addition and evaporation of acids to dissolve the chromite.
The Fe 3+ /(Cr + Al + Fe 3+ ) ratio of the analysis is 0-106; such low values seem
to be characteristic of the chromites in alpine complexes (Himmelberg & Loney,
1973); however, Rodgers (1973) has shown that such low values can also be found
in the chromites of stratiform complexes. On Mg/(Mg + Fe 2+ ) vs A1/(A1 +
Cr + Fe 3+ ) and total Fe plots such as the ones used by Rodgers (1973), the Jijal
chromite, like the New Caledonian alpine chromites, falls in the marginal part of
the field for stratiform complexes but well within the field of alpine complexes.
The two spinels (columns 3 and 4) are intimately associated in granulite 297c; in
THE JIJAL C O M P L E X , PAKISTAN 103
some, the dark coloured, almost black, spinel is sprinkled in the form of dense dust
in the green variety. The chemical composition of the two is remarkably similar and
the cause for the difference in their colour is not clear. In both, considerable Fe 2+ is
replacing Mg (Mg/Fe 2+ = 2-24 in the green and 2-34 in the dark variety),
indicating pleonaste.
Spinel is a rare accessory in the Jijal granulites except in the case of rock 297c,
where it becomes an important phase. The spinel-bearing rock occurs in the form of
a band a few centimetres thick that separates a garnet granulite part (297b) from a
websterite part (297a) in a large granulite xenolith in the ultramafic rocks. The

spinel-bearing rock is chemically very similar to the garnet granulite (297b) except
for a higher CaO and lower FeO in the former. It is not clear whether the spinel
was produced during the regional or later retrograde metamorphism or by contact
metamorphism/metasomatism imposed by 297a (? hot websterite intrusion)
surviving the regional metamorphism which affected the websterite as well as the
granulite.
The two spinels are very low in Cr in contrast to those of the pyroxenite
inclusions in basalts (Wilkinson, 1973) and of alpine peridotites (Medaris, 1972).
Although their host rock is not a part of the alpine ultramafic rocks of Jijal, it may
be mentioned that Cr-depleted spinels have been reported from ultramafic rocks
affected by contact metamorphism (Springer, 1974; Frost, 1975) and by
metasomatism accompanied by low oxygen activity (Hamlyn, 1975).
Garnet. Analyses of 10 garnets from the granulites, together with physical data
for eight of them, are presented in Table 5. The structural formulae, recalculated on
the basis of 24 oxygens, approach closely the ideal formula Xf+ F 2 3+ Z 3 O I2 .
However, in 297b, 283, 270, and 338 the amount of cations in Z i s slightly less and
in Y slightly more than that required by the ideal formula, probably due to slight
underestimation of their FeO contents (see Deer et al., 1962a). The garnets are
composed essentially of the components almandine (27-43 per cent), pyrope
(28-46 per cent), and grossular (16-28 per cent).
The Mg/(Fe 2+ + Ca) ratio of the garnets is influenced by this ratio in their host
rocks. They are richer in Fe 2+ , Mn, Fe 2+ /Mg and (Fe + Mg)/Ca than their host
rocks. Similar relationships have already been found by Buddington (1952), Howie
& Subramaniam (1957), and Subramaniam (1962). Like the garnets of eclogites
(Coleman et al., 1965) and other high-pressure metamorphic rocks, they are
generally lower in Cr, Cu, Ni, and higher in Zn than their host rocks.
The andradite component is low in all the analyses despite the fact that andradite
garnet is stable to high pressures and temperatures.
The Mn content of the garnets is also low, partly due to the poverty of their host
rocks in Mn and partly to high-grade metamorphic conditions. Although
spessartine is stable at high P and T (Mottana, 1974), Saxena (1968) and Hsu
(1968) assert that the composition of the host rock plays a role in determining the
proportion of spessartine in garnet, and a number of authors have found that low
grade garnets are enriched in MnO content (Barth, 1936; Miyashiro & Shido,
1973). *
Although the grandite component of garnets in eclogitic rocks is considered to be
rom http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015
TABLE 5
Analyses of the garnets from the granulites ofJijal
SI 297b SI 283 SI 270 SI320E SI 104b SI 292 SI 338 SI 290 SI 291 SI 289
SiO2 40-30 40-41 40 02 4115 39-32 39-25 39-44 39-36 3900 39-44
TiO2 019 0-25 0-21 0-09 0-18 0-15 0-26 0-23 0-22 0-30
A12O, 22-82 23-02 22-50 22-19 21-33 22-09 22-04 22-18 21-60 21-72
Fe 2 O, 0-78 0-72 1-02 — — 119 1-42 0-51 0-87 0-70
FeO 16-22 1310 15-73 19-73 19-26 18-21 19-21 19-94 20-20 18-72 2
MnO 0-49 0-27 0-46 0-70 008 0-62 0-49 0-48 0-68 0-73
MgO 12-33 12-22 1101 10-93 10-36 10-32 8-99 8-48 8-25 7-48 O
CaO 7-10 1001 9-31 6-36 8-99 8-07 8-39 8-96 8-86 1115
Na2O 003 002 0-02 002 001 001 0-02 003 005 0-05 >
K2O 001 001 000 001 000 001 0-01 000 000 002 z
P 2 O, 000 000 000 — — 000 000 000 001 000 >
TOTAL 100-27 10003 100-28 100-18 99-53 99-92 100-27 100-17 99-74 100-31 z
n 1-753 1-747 1-758 1-762 1-765 1-767 1-769 1-766
D 3-814 3-773 3-772 3-858 3-873 3-822 3-895 3-782
a (A) 11-570 11-583 11-594 11-583 11-599 11-605 11-597 11-622
X
Number of ions on the basis of 24 (0)
Si 5-957 604 5-9751 5-963 6 0 0 o
Al 600
0-043 6 0 0 0-0251 6 00 0-037 600 3S
Al 3-905 3-926 3-796] 3-891
F e 3+
4 0 6 0115 4-04 — >3-94 — >3-82 0-162 408 4-00
Ti 0023 0010 0-02 V 0-030
Mg 2-443 2-444 2-346"! 2-0261
Fe2+ 1-958 2-475 2-448 2-429
Mn 0058 0-063L
5-92 1-485 •5-95 1-022 04 o-oiol •6-27
0088
01 6-00
Ca 1-4641 1-359)
Na 0-005 0-006 0-003 0-005
K 0 002 0-002J
o-oooJ
Mol. per cent end-members
Almandine 33-9 27-3 32-9 410 39-1 38-3 41-3 42-4 430 39- 7
Andradite 30 2-7 3-5 — 3-8 4-9 2-1 31 2- 9 3G
Grossular 161 240 21-5 17-0 23-3 17-9 18-2 22-3 211 27- 5
Pyrope 460 45-4 41-1 40-5 37-4 38-7 34-5 32-2 31-3 28- 3 >-
Spessartine 10 0-6 10 1-5 0-2 1-3 11 10 1-5 1- 6
PPM r
Co 60 79 55 — 58 62 63 54 42 n
Cr 223 125 845 410 480 136 64 36 46 64
Cu 12 43 14 — — 14 18 18 15 18
Li 1 1 1 — — 1 1 1 1 1 r
Ni 20 24 21 — — 20 21 20 19 19
rS
Pb 40 5 38 — 5 26 29 5 35
V 166 84 160 — — 218 220 248 190 190
Zn 78 84 53 — — 67 189 175 128 62 >
SI 320E and 104b were analysed by microprobe by Ian Young and Cavit Mahmut, respectively. The remainder were analysed by M.Q.J. Total iron is expressed
as FeO in the microprobe analyses.
SI 320E and SI 104b are both garnet websteritic granulites from near the main contact between the granulites and ultramafic rocks along the Karakoram Highway;
they are plagioclase-free but 32OE contains (? secondary) hornblende. For other specimens see Fig. 1 and legend to Table 1.
dependent upon the normative an/(ol + hy) ratio of their host rocks and the
composition and proportion of the associated minerals (Church, 1968), increasing
pressure also favours the accommodation of grossular relative to almandine in
Fe2+—Mg garnets. This is strongly suggested by the higher grossular content in the
garnets of glaucophane schists, eclogites (Coleman et al., 1965; Dobretsov et al.,
1972), kyanite eclogite and grospydite xenoliths in kimberlites (Sobolev et al.,
1968), and by experimental work (Green & Ringwood, 1967; Akella & Kennedy,
1971; Hensen, 1976).
The Jijal granulite garnets are richer in grossular, and, especially, pyrope than

most garnets of the granulite facies. However, Scharbert & Kuart (1974) have
reported up to 40 per cent grossular in garnets of Moldanubian granulites, and
Nemec (1967) considers that up to 50 per cent grossular can be accommodated in
the almandinic garnets of granulites. A comparison of the Jijal garnets with those
from various parageneses (Dobretsov et al., 1972, table 32) shows that they are
most similar to the garnets of eclogites from metamorphic complexes. They have
lower almandine than occurs in the garnets from granulites and higher pyrope than
in those from diopside-plagioclase rocks. The pyrope — (almandine + spessartine)
— (grossular + andradite) contents of the garnets were plotted on the triangular
diagram (Fig. 4) of Dobretsov et al. (1972). Eight plot in the overlapping fields of
eclogite and granulite and two in the eclogite garnet field. One of the latter two is
from the possibly metasomatized rock (289) which is mineralogically non-eclogitic
(garnetiferous hornblende rock). The experimental evidence cited above (see also
Green & Ringwood, 1972) suggests that the garnet-clinopyroxene-plagioclase
assemblage (the most common in the Jijal granulites) developed at lower pressures
than typical eclogites.
Epidote. In addition to retrograde epidote after plagioclase and garnet, and in the
epidote-rich granulites (assemblage 16 and 17), certain Jijal granulites contain
minor amounts of epidote which seems to be in stable co-existence with other
minerals, including fresh plagioclase. Two analyses of epidotes give pistacite, 100
Fe 3+ /(Fe 3+ + Al), contents of 14-6 (SI 291) and 19-3 (SI 290). Epidote has not
been found in any local granulite containing orthopyroxene.
Although the epidote minerals are generally regarded as being stable only up to
about the middle of the amphibolite facies, there is some evidence that under
favourable conditions they may be stable even in granulite facies environments.
Den Tex & Vogel (1962) suggested the recognition of an epidote granulite
sub-facies in the granulites of NW Spain, considered to have formed between 630
and 690 °C and 8 to 10 kb (Den Tex et al., 1972). In the basic granulites around
Saltora, W. Bengal, considered to have formed at 750-830 °C and 6-8 kb, Manna
& Sen (1974) noted an unusual variety of garnetiferous-pyroxene granulite
containing a large amount of 'prograde (?) epidote'. The experimental work of Liou
(1973) suggests that epidote is stable under high Pmo at temperatures of the
granulite facies and, while studying the brown hornblende mylonite from the St.
Paul Rocks, Millhollen & Wyllie (1974) noted its formation at more than 1100 °C.
Thus there is a likelihood that in some Jijal granulites epidote is a primary
(prograde) metamorphic mineral. Typically this occurs in large grains with sharp
boundaries against other minerals. However, the possibility should not be ruled out
that in some rocks in which epidote occurs garnet granulites completely
re-equilibrated at lesser depths, and hence lower P—T conditions, during upward
transport and are not isofacial with the remaining granulites. This, at least, seems to
be the case with the well-formed paragonite, 100 Na/(Na + K) = 96, found in an
epidote-bearing granulite (291). Such retrograde epidote is found replacing other
minerals along fractures, as abundant needles in plagioclase, and as large to small
grains found in abundance (up to 50 per cent) in a few rocks where it may show
radial growth and cut across large grains of prograde epidote.

Orthopyroxene. Orthopyroxene is a common mineral in the ultramafic rocks of
Jijal, but in the garnet granulites it is restricted to the less abundant, plagioclase-free
websteritic rocks. However, unstable relics of hypersthene have also been noted in a
plagioclase-bearing granulite. Analyses of five orthopyroxenes are given in Table 6.
Two are from ultramafic rocks (275 and 295) and the remaining three from
granulites; their compositions have been plotted in Fig. 5. Four of the pyroxenes
are bronzites whilst that from harzburgite 275 is enstatite; their enstatite contents
are dependent upon the Mg/Fe 2+ ratios of their host rocks, and their refractive
indices and density increase with iron content. The bronzites of the Jijal granulites,
like the orthopyroxenes of the granulite facies (Howie, 1965), are strongly pleo-
chroic with a pink to red, /? yellow, and y green.
Of the two orthopyroxenes from the ultramafic rocks, the one from harzburgite
275 is very similar to those from alpine ultramafic rocks (Jackson & Thayer, 1962;
Medaris, 1972). The orthopyroxene from websterite 295, however, has a
considerably lower enstatite content, reflecting the lower mol. ratio MgO/(MgO +
FeO total + MnO) of its host rock compared with alpine peridotites. Compared to
the orthopyroxenes of harzburgites from Burro Mountains (Loney et al., 1971),
orthopyroxene 275 is much lower in Cr and V while the values for Co, Cu, and Ni
are fairly close.
The three bronzites from the websteritic members of the granulites (297a, 104b,
320E) are significantly different from the bronzite of the ultramafic websterite
(295). Apart from being more strongly pleochroic and having lower enstatite
contents, the granulite bronzites are poorer in Ti (except 104b), Cr, and richer in Al
and, in the case of 297a, in Fe 3+ , Co, Cu, V, and Pb than the ultramafic bronzite.
Clinopyroxene. Analyses of four clinopyroxenes from the ultramafic rocks and
ten from the garnet granulites are presented in Table 7. The former are diopsides
and have light- to grass-green powders, colourless to very faint green in thin
section; the latter are diopsidic (similar in colour to those in the ultramafic rocks)
to salitic (dark green powders, green in thin section). The a and y of the pyroxenes
generally plot a little above the trend established by Deer et al. (1978, p. 252) but
the two most aluminous (also the most Fe3+-rich) pyroxenes show a significant
departure. The Ca-Mg-Fe ratios obtained from plotting 2V against /? are consider-
ably different from those derived from the analyses and suggest that the Hess
(1949) and Muir (1951) diagrams are not suitable for composition determinations
of certain diopsidic-salitic pyroxenes (? those with high Aliv + Fe 3+ ).
The essential variation in the chemistry of the analysed clinopyroxenes is in the
TABLE 6
Orthopyroxenes from the Jijal complex
SI 275 SI 295 SI 297a SI320E SI 104b
SiO 2 56-72 55-20 53-27 53-75 51-79

TiO 2 0-27 0-24 009 0-07 0-32
A12O3 0-79 1-36 3-45 2-85 5-48
1-27 0-94 1-94 — —
FeO ' 4-90 10-50 11-46 15-52 15-84

MnO 016 0-30 0-30 018 0-30
MgO 34-99 30-28 28-37 26-48 25-18
CaO 0-29 0-57 0-43 0-34 0-60
Na 2 O 002 000 001 0-01 004
K2O 001 000 000 001 000
H 2 O+ — — 0-27 — —
P2O5 000 0-00 000 — —
TOTAL 99-42 99-39 99-59 99-21 99-55

a 1-660 "1=671 1-675 — —
P 1-664 1-678 1-683 — —
y 1-672 1-684 1-690 — —
D 3-270 3-351 3-370 — —
Number of ions on the basis of 6 oxygens
Si 1-963 1-960 1-907 1-947 1-878
Al 0032 0040 0093 0053 0122
Al 0000 0017 0053 0069 0112
Ti 0 007 0006 0 002 0002 0 009
Fe3+ 0033 0025 0-052
F e 2+ 0142 0-312 0-344 0-470 0-481
Mn 0005 0009 0009 0006 0 009
Mg 1-805 1-602 1-514 1-429 1-361
Ca 0011 0022 0-016 0013 0023
Na 0001 0000 0001 0001 0002
K 0000 0-000 0-000 0000 0000
Z 1-995 2 000 2000 2000 2-000
XY 2-004 1-993 1-991 1-990 1-997
100 Mg
(Mg + Fe 3 + + 1=e 2+ + M n )
90-9 82-2 78-9 750 73-5
Mg 90-5 81-3 78-2 74-5 72-8
Fe 90 17 6 21-0 24-8 26-1
Ca 0-5 1-1 0-8 0-7 1-1
PPM
Co 47 79 96
Cr 886 1482 438 205 410
Cu 15 22 51 — —
Li 1 1 1 — —
Ni 563 266 238 — —
Pb 6 6 11 — —
V 39 84 172 — —
Zn 40 69 — — —
Analyses 275, 295 (from ultramafic rocks) and 297 (granulite) by M.Q.J.
Analyses 320E amd 104b (granulites) by microprobe by Ian Young and Cavit
Mahmut, respectively, express total iron as FeO.
quantity of Mg, Fe 2+ , and to a lesser extent, in Si and Ca, while those from the
granulites also show substantial variations in Al, Fe 3+ , and Na. With increase in
Fe, the amounts of Si, Ca, Cr, and Ni fall and V rises. Co, Cu, and Pb are generally
higher in the granulite clinopyroxenes than in those from the ultramafic rocks.
10 Fe 2+ +Fe 3 *+Mn
FIG. 5. Plots of the pyroxene compositions for the ultramafic rocks (A) and garnet granulites (A) on a portion
of the pyroxene quadrilateral. Tie-lines are drawn between the analysed pairs.
Seventeen clinopyroxene compositions, three of them from the ultramafic rocks

determined optically by Qureshi & Jan (1977), have been plotted on the pyroxene
quadrilateral (Fig. 5); those from the granulites have a distinctly different trend
than those from the ultramafic rocks. The cation proportions of Fe and Mg and the
A12O3 contents of the ultramafic clinopyroxenes are well within the range displayed
TABLE7
Clinopyroxenesfrom the Jijal complex
SI 275 SI 275a SI 272 SI 295 SI 297a SI 283 SI 297b SI 297c SI320E SI 104b SI 270 SI 292 SI 338 SI 290
SiO 2
TiO 2
5400
0-40
54-32
0-16
54-35
010
53-77
0-22
51-82
0-32
51-58
0-38
51-21
0-74
51-60
0-27
52-47
0-25
49-68
0-54
49-51
0-71
49-80
0-71
48-33
0-86
48-73
0-88
s
r\J
A12O3 0-58 0-50 0-86 1-14 3-47 5-26 410 5-14 2-94 5-46 5-90 6-02 9-56 9-88
Fe 2 O 3 0-47 0-29 0-27 0-21 119 0-80 1-37 — — — 2-22 2-11 4-97 4-46 <—»
•w
FeO 1-34 1-93 1-85 301 3-24 2-15 2-97 3-77 5-22 5-04 3 06 4-82 ' 4-13 4-71
MnO 008 009 009 013 014 002 004 005 007 010 0-04 0-06 0-04 0-04 >
MgO 18-36 17-87 17-68 17-13 15-40 14-88 I486 14-80 14-43 14-24 13-94 12-91 1002 9-69
CaO 24-21 23-82 23-73 23-27 23-74 23-98 24-50 25-22 23-41 24-02 23-39 22-03 19-73 19-26 zD
Na 2 O 0-20 0-24 0-20 017 0-22 0-64 0-26 006 0-50 0-25 0-50 1-111 1-90 1-82
K2O 001 001 0-02 0-02 0-00 000 0-00 — 0-01 000 001 001 0-03 001
P2O5 0-00 0-00 000 000 0-00 000 000 — — — 000 000 0-00 0-00 >
TOTAL 99-65 99-23 99-15 99-07 99-54 99-69 100-05 100-91 99-30 99-33 99-28 99-58 99-57 99-48 X
a 1-668 1-671 1-676 1-672 1-675 1-678 1-681 1-695 1-697 o
P 1-675 1-678 1-683 1-680 1-682 1-690 1-691 1-704 1-706
1-695 1-698 1-705 1-703 1-707 1-712 1-710 1-719 1-721 m
y
50° 52° 57° 57° 60°
D 3-271 3-284 3-291 3-302 3-322 3-300 3-320 3-335 3-300 3-369 3-358
Number of ions on the basis of six oxygens
Si 1-965 1-985 1-985 1-975 1-907 1-884 1-879 1-876 1-942 1-846 1-834 1-848 1-791 1-804
Al 0025 0015 0015 0025 0-093 0-116 0-121 0-124 0-058 0154 0166 0-152 0-209 0-196
Al — 0-007 0022 0-024 0058 0111 0056 0096 0070 0085 0098 0111 0-209 0-235
Ti 0011 0004 0003 0006 0009 0010 0-020 0-007 0 007 0015 0-020 0020 0-024 0 025
Fe 3+ 0-013 0008 0007 0006 0-033 0-022 0-038 — — — 0062 0059 0-139 0124
Fe 2+ 0041 0059 0-057 0-092 0100 0-066 0091 0-115 0-162 0-157 0095 0-150 0128 0146
Mn 0-003 0-003 0-003 0-004 0004 0-001 0001 0-001 0-002 0003 0-001 0-002 0001 0-001
Mg 0-996 0-973 0-962 0-938 0-845 0-810 0-813 0-802 0-796 0-789 0-770 0-714 0-554 0-535
Ca 0-944 0-933 0-929 0-916 0-936 0-939 0-963 0-982 0-929 0-956 0-929 0-876 0-784 0-764 a
Na 0-014 0-017 0-014 0-012 0016 0-045 0-019 0-002 0-036 0-018 0036 0 080 0-137 0131 rn
K 0000 0000 0001 0001 0000 0000 0-000 — 0-000 0000 0-000 0 000 0001 0-000
X +Y 2-022 2004 1-998 1-999 2-001 2-004 2001 2-005 2-002 2-023 2011 2012 1-977 1-961
Mg 49-9 49-3 49-2 48-0 44-0 44-1 42-6 42-2 42-1 41-4 41-5 39-7 34-5 34-0 r
Fe 2-8 3-5 3-4 5-2 7-3 4-8 6-8 6-1 8-7 8-4 8-5 11-7 16-7 17-3 o/~\
Ca 47-3 47-2 47-4 46-8 48-7 51-1 50-6 51-7 49-2 50-2 50-0 48-6 48-8 48-7
s
•0
PPM
Co 19 20 21 27 36 40 39 79 36 35 19 41 rm
Cr 1642 1710 2120 1642 614 222 222 137 273 684 520 160 78 39
Cu 13 10 6 10 32 24 12 — — — 7 11 162 84 13
Li 2 4 1 2 2 3 2 2 3 — 6 >
Ni 318 229 165 147 119 316 117 157 — — 231 99 74 70
Pb 14 9 12 12 20 9 22 — — — 5 12 — 24
2
VJ
V 68 65 87 132 310 200 472 224 — — 547 608 470 375

Zn 33 — 264 — — 95 — — — — 84 98 220 — Z
Samples 297b, 297c and 320E were analysed by microprobe by R. F. Symes, Ian Young, and Cavit Mahmut, respectively, and express total iron as FeO. The
remaining analyses are by M.Q.J. Not included in the totals are 0-13 per cent H 2 O+ in 275a, 0-08 per cent H 2 O+ in 295, 0-03 per cent H 2 O in 272, 0-16 per cent
H 2 O+ in 297a and 0-13 per cent H 2 O+ in 270.
Clinopyroxenes 275, 275a, 272, and 295 are from ultramafic rocks; the remainder are from granulites.
by the clinopyroxenes from harzburgite-subtype ultramafic rocks; their A12O3

contents are much lower than those from lherzolite subtype alpine complexes
(Jackson & Thayer, 1972; Green, 1964). The Mg/(Mg + Fe) ratios (0-91-0-95) of
the diopsides from the ultramafic rocks are also similar to those of the alpine
peridotites although slightly lower values have been reported from the Lizard area
(Green, 1964). The Cr content of the diopsides is slightly lower than in those of the
Oregon peridotites (Himmelberg & Loney, 1973), California (Loney et al, 1971),
and the Lizard. Ni, V, and Pb also are lower than in the Lizard.
The clinopyroxenes from the granulites (including those of the websteritic

members) are different from those of the ultramafic rocks. The former contain
more A12O3 (3-5-9-9 per cent), Cu, Co, V, less Cr, Ni, and have higher Fe 2 O 3 /FeO
(0-31-1-20) and Wo/(En + Fs) ratios than the latter. The Mn content of the
granulite clinopyroxenes also is generally lower, partly due to the low Mn content
of their host rocks and partly because of the abundance of the associated garnet
which Mn occupies preferentially over clinopyroxene. With increase in Fe, the
granulite clinopyroxenes show a general increase in Fe 3+ /Fe 2+ , Al, Na, and Ti, and
a decrease in Si and Ca. Traces of potassium are present in those five which have
higher Na contents.
The granulite clinopyroxenes are not omphacites, having insufficient Si and Na.
They have broad similarities to clinopyroxenes from some ultramafic xenoliths in
basalts (Morgan, 1974; Wilkinson, 1973, 1974), some granulite facies clino-
pyroxenes and, interestingly, with those of the Adirondack anorthosite series
(Buddington, 1950). The higher A12O3 of the four clinopyroxenes from pyroxenites
(297a, 270, 104b, 32OE) is similar to those from high P/T peridotites but much
higher than those of the 'low pressure' peridotite pyroxenes (Green, 1964; Challis,
1965; Medaris & Dott, 1970; Medaris, 1972).
The variations in silica and alumina in the Jijal granulites do not seem to have a
determinative influence on the A12O3 contents of the clinopyroxenes; the two most
aluminous clinopyroxenes (290 and 338) are from the rocks with the highest silica
but not alumina. The proportion and composition of the coexisting phases appear
to have a greater influence over the A12O3 content of the clinopyroxenes.
The Na content of the granulite clinopyroxenes is not high and is similar to those
from many other granulites. In six of them the low Na is in direct response to the
low Na 2 O of their host rocks despite the total disappearance of plagioclase.
However, rocks 290 and 338, with considerably higher Na2O (2-4 and 3-1 per
cent, respectively) also do not contain Na-rich clinopyroxenes. It is most likely that
the operating pressures were such (lower than in the case of typical eclogites) that
more Na could not be accommodated in the two clinopyroxenes as jadeite solid
solution, resulting in the presence of plagioclase. This view is supported by the
work of Green & Ringwood (1967).
White (1964), Sobolev et al. (1968), Dobretsov et al. (1972), and Deer et al.
(1978) have noted that alumina in omphacites from eclogites is concentrated in the
octahedral site. In the granulite facies clinopyroxene, Al is approximately equally
distributed in Z and Y (White, 1964). Aoki & Kushiro (1968) and Francis (1976)
suggest that the Al vi : Aliv ratio in clinopyroxenes from eclogites is more than 2:1
and in those from granulites less than 2:1 (generally about 1:1). The Alvi to Aliv
ratio in the Jijal granulite clinopyroxenes ranges from 0-5 to 1-2 (Fig. 6),
supporting the observations of these authors.
No hypothetical end-members have been calculated for the clinopyroxenes
because of the criticism of Edgar et al. (1969) and Thompson (1974) concerning
the different methods (at least fourteen) of calculation. In most of the Jijal granulite
clinopyroxene analyses there is sufficient Fe 3+ to consume most or all of the Na in
forming the acmite component preferentially over jadeite. This 'acmite first'
approach has been strongly criticized by Thompson (1974), although the positive

30
2-5
20
^ 1-5
10
05 10 V5 20 2-5
Al'V x 10
FIG. 6. Granulite clinopyroxenes—octahedral Al versus tetrahedral Al. Fields of eclogite and granulite
clinopyroxenes after White (1964).
correlation between Na and Fe 3+ in the clinopyroxene analyses justifies it.

However, even if jadeite is calculated before acmite, the quantity of the former is
significantly less than the Tschermak's component. The low jadeite and high
Tschermak's components in clinopyroxenes have been considered to be typical of
the granulite facies (White, 1964; Green & Ringwood, 1967; Kushiro, 1969;
Dobretsov et al, 1972).
To summarize this discussion, it is concluded that the substantial amounts of
Alvi and low Na in the Jijal clinopyroxenes are suggestive of their formation at high
pressures but below those of the eclogite facies.
Amphibole. The amphibole of the granulites is generally green or brownish green,
but in some rocks it is faintly coloured or bluish green; a few have a purplish (?)
Ti-rich hue. Bluish green amphibole is also a common secondary product after
garnet and, more rarely, after clinopyroxene. Whether or not all of the bluish
amphibole is secondary, at least in the 'garnetites', is difficult to assess.
Compositions of five primary amphiboles analysed range from alumino-

magnesio-hornblende to pargasite and alumino-tschermakite, following the I.M.A.
nomenclature (Leake, 1978). None is exceptional, but their high A12O3 and
Alvi/Aliv are suggestive of high-pressure conditions (Leake, 1965, 1971; Raase,
1974). Details of the amphiboles, epidotes, and paragonite will be reported later.
Fe-Mg DISTRIBUTION BETWEEN THE COEXISTING MINERALS

The Fe-Mg distribution in coexisting minerals has been investigated by a
number of workers. The distribution coefficient (KD) for the Fe-Mg exchange

reaction is defined as
where XMg refers to mole fraction Mg/(Mg + Fe 2+ ) in minerals A and B. Thus
Pyroxene pairs. Despite the chemical factors (Binns, 1962; Ray & Sen, 1970;
Maxey & Vogel, 1974; Lindh, 1975) and non-ideal mixing in the iron-rich
pyroxenes (Davidson, 1968; Virgo & Hafner, 1970) that can influence KD, it is
commonly accepted that crystallization conditions have significant control over KD
(Kretz, 1963; Scharbert & Kuart, 1974; Sen & Manna, 1976); see also Wells
(1977) and Herzberg (1978). Although there is a scatter in the KD values, the
igneous pyroxenes have significantly higher KD Mg-Fe values (near 0-73) than
those of metamorphic pyroxenes with typical values of 0-54 {e.g. Lindh, 1975). The
KD Mg-Fe for the pyroxene-pairs from the Jijal complex are 0-52 (SI 275) and
0-50 (SI 295) for the ultramafic rocks, and 0-52 (SI 297a), 0-62 (SI 320E) and
0-56 (SI 104b) for the granulites. These values are close to those of metamorphic
pyroxenes and suggest that even the ultramafic rocks may have re-equilibrated at
lower than magmatic temperatures.
Hornblende—pyroxene pairs. It has been noted in a number of areas that KD
Fe—Mg has a narrow range for hornblende-hypersthene pairs and a much broader
range for hornblende-clinopyroxene pairs. The value for the hornblende-ortho-
pyroxene pair in SI 104b is 0-97; and those for hornblende—clinopyroxene are 1-65
(SI 283), 1-62 (SI 297b), and 1-72 (SI 104b), all belonging to the garnet granulites
and falling within the range shown by Swedish charnockites (Saxena & Hollander,
1969) and W. Bengal granulites (Sen & Manna, 1976) but distinctly lower than
those of the Quairading granulites (Davidson, 1971). The values for the
hornblende-clinopyroxene pairs are more similar to those of orthopyroxene
eclogites formed in the transitional granulite facies of Colton, U.S.A., than to those
from other types (Lappin & Smith, 1978).
Hornblende-garnet pairs. The KD Fe-Mg values for these pairs (0-22 in SI 283;
0-27 in 289; 0-34 in 291; 0-25 in 297b; and 0-33 in SI 104b) in granulites cover
the same range as in the Halmstadt charnockites, whereas those from Varberg have
a greater range (Saxena & Hollander, 1969). Rao & Satyanarayana (1971) have
reported distinctly higher values for the amphibolite pairs in Mysore, India.
It is worth noting that mineral pairs other than pyroxenes are considered to be
more strongly influenced by the composition of the phases in question, especially
Al (particularly Aliv) in hornblende, Fe 3+ in clinopyroxene and Mn and/or Ca in
garnet (Kretz, 1960; Saxena & Hollander, 1969; Davidson, 1971). However, the
extent of the influence of these constituents over KD is not fully understood.
Garnet-clinopyroxene pairs. Partitioning of elements during the gabbro-eclogite
transition is commonly considered to be sensitive to temperature and pressure
(Loomis, 1977). Variation in KD Fe2+—Mg between co-existing garnet and
clinopyroxene is thought to be mainly in response to T, the influence of P and

composition being small to negligible (Banno, 1970; Mori & Banno, 1973; Irving,
1974; Raheim & Green, 1975; Wood, 1976).
Irving's (1974) study of more than 300 garnet-clinopyroxene pairs from
eclogites and garnet pyroxenites shows that the pairs from glaucophanitic terrains
have the highest KD Fe 2+ -Mg values (generally >17), whereas those from
inclusions in basalts and from 'alpine-type' peridotites have the lowest values
(<2-7). The values from the Jijal granulites are: 4-73 (SI 292), 4-84 (290), 4-98
(320E), 5-18 (338), 5-24 (104b), 6-50 (270), 6-58 (297b) and 7-42 (283),
averaging 5-68. These values and their lower range, ignoring the rather high value
in 283, suggests equilibration of the garnet-clinopyroxenes under not greatly
varying granulite facies conditions.
Enstatite-olivine pair. The KD Fe-Mg for this mineral pair in harzburgite SI 275
is 0-91. However, although Frost (1975) found a similar value (0-91) for this pair
from 'ordinary' and high-pressure, high-temperature peridotites, ultramafic nodules
in basalts, and contact metamorphosed ultramafic rocks, he suggested that the
Fe-Mg distribution in the pair is rather insensitive to temperature changes (cf.
Springer, 1974).
Olivine-diopside pair. The Fe-Mg distribution between this pair seems to be
temperature sensitive (Medaris, 1972; Frost, 1975). The KD Fe-Mg (=0-47) for
the pair from harzburgite 275 is between those of the high-temperature peridotites
and the olivine diopside pairs formed under greenschist facies conditions (Frost,
1975).
TEMPERATURE-PRESSURE ESTIMATES FOR THE JIJAL ROCKS

The attainment of equilibrium, suggested by texture, narrow range in the KD
Fe-Mg values between the coexisting mafic mineral pairs, and their general
similarity to those from other high grade metamorphic rocks, together with the
systematic partitioning of minor and trace elements (Jan, 1977) and the apparent
lack of compositional zoning in the analysed minerals, make the Jijal rocks suitable
for estimation of temperature and pressure. Space does not permit details of the
methods followed and only a brief account is given below. Some of the models
followed here have been discussed by O'Hara & Yarwood (1978), Ernst (1978)
and Lappin & Smith (1978).
The garnet granulites. Table 8 presents the temperature estimates deduced by
different methods. If the few higher (based on two-pyroxenes) and lower values
(based on the garnet-clinopyroxene model of Perchuk, 1968) are discarded, the
temperature estimates range from 650 to 800 °C. (The temperature values given by
Ca in the Gar-Cpx method are not always realistic; Perchuk, 1968.) Dobretsov et
al. (1975) have criticized the garnet-amphibole and clinopyroxene-amphibole
geothermometers of Perchuk, but in the present case they yield consistent results.
Of particular interest is the rather narrow range of temperature estimates (at an
assumed pressure of 12 kb) given by the method of Raheim & Green (1975). Wood
(1977) has shown that this latter method is applicable to most natural rock
TABLE 8

Temperature estimates (°C)for the garnet granulites
1 2 3 4 5 6 7 5 9 10 11
270 680° 650° 780°

283 655° 625° 650° 750° 790°
289 700°
290 760° 610° 834°
291 750°
292 770° 640° 850°
297b 680° 725° 650° 640° 735°
338 730° 630° 710°
104b 730° 700° 675° 740° 768° 810° 725° 815° 700° <S8OO° 724°
32OE 700° 720° 805° 830° 720° 750° ^800° 790°
297a 888° 820° 750° «800°
1. K°££%, method (Raheim & Green, 1975).

2. Ca/(Ca + Mg + Fe + Mn + Na + K) garnet vs Ca/(Ca + Na + K.) clinopyroxene (Perchuk, 1968).
3. K°^-_AF^ (Perchuk, 1967, 1969).
4 K
- Z,~-*?Z ( p erchuk, 1969).
5. Two-pyroxene geothermometer (Wood & Banno, 1973).
6. Two-pyroxene geothermometer (Perchuk & Vaganov, 1977).
7 K
- D"7°F' MH (Perchuk, 1968).
8 K
- ZT4X (Perchuk, 1969).
9. Approximate estimates from the composition of clinopyroxene coexisting with orthopyroxene (Fleet,
1974).
10. Very approximate estimates from Ca/(Ca + Mg) in clinopyroxene coexisting with orthopyroxene
(Mysen, 1976), by extrapolation.
11. Garnet-clinopyroxene geothermometer of Saxena (1979), at an assumed pressure of 12 kb.
assemblages, and in the case of the South Harris granulites (considered by Wood
to have formed at 825 °C, 13 kb) gave results that are in excellent agreement with
those derived from a number of other methods; see also Saxena (1979).
The estimated 12 kb of pressure seems reasonable. Based on the diagram of
Raheim & Green (1975), minimum estimated pressures for the plagioclase-free
granulites are about 12 kb for 270, 283 and 297b, 13 kb for 104b, and 14 kb for
292 and 32OE. Similarly maximum pressures for the plagioclase-bearing granulites
(290 and 338) could not have been more than 13-5 kb. One of the plagioclase-free
garnet-clinopyroxene rocks of Jijal (SI 299, Table 1) is chemically similar to the
garnet pyroxenite (R 392) experimentally investigated by Irving (1974; table 6).
For SI 299 rock, a pressure of 12 kb at 750 °C is sufficient for the formation of a
plagioclase-free garnet-clinopyroxene assemblage.
Application of the KD Mg-Fe against KD Ca in garnet and clinopyroxene

(Perchuk, 1968, 1969) gives the following pressure estimates: SI 270 = 12 kb,
SI 283 = 8 kb, SI 290 = 14 kb, SI 292 = 13 kb, SI 297b - 11 kb, SI 338 = 16 kb,
SI 320E = 10 kb, and SI 104b = 9 kb. The temperature estimates by this method
are generally lower than those deduced by other methods. Assuming a temperature
of 750 °C, SI 297a yields a pressure estimate of 10 kb, using Ca/(Ca + Mg) in
orthopyroxene (Mysen, 1976), and 320E and 104b give 22 and 12 kb,
respectively, using A12O3 in orthopyroxene coexisting with garnet and clino-
pyroxene (Akella, 1976), the three values having been derived by extrapolation.

Unstable relics of hypersthene in rock 328 suggest a pressure in excess of 8-10 kb
at a temperature of 700 °C according to the reaction Opx + Pig = Gar + Cpx +
Qz (Green & Ringwood, 1967; Winkler, 1974). Since olivine and orthopyroxene
have not been found in any of the rocks, except in websterites, despite their
abundance in the norms, it is assumed that they disappeared by reacting with
plagioclase at pressure of more than 10 kb at about 700 °C. Thus it seems likely
that the Jijal rocks have equilibrated at 650-800 °C and 10-15 kb (possibly
between 670 and 790 °C and 12 to 14 kb).
Petrographic, chemical and mineralogical evidence supports the idea that the
plagioclase-bearing and plagioclase-free granulites are the result of the same
processes. The presence or absence of the plagioclase is the consequence of the
different chemistries of the two types rather than different pressures of formation.
The control of Mg/(Mg + Fe), normative Ab/An, and Na2O and SiO2 contents of
the bulk chemistry over the disappearance of plagioclase in the gabbro to eclogite
transition has been emphasized by Green & Ringwood (1967) and Kushiro &
Yoder (1966). The plagioclase-bearing rocks of Jijal are richer in Na2O and have
much higher normative Ab/An ratios than the plagioclase-free granulites, and
whilst the latter are olivine (±nepheline)-normative, the former are hypersthene +
quartz-normative. Thus in the more basic assemblages the plagioclase disappeared
more readily at pressures under which it could exist in the less basic assemblages.
Yoder & Tilley (1962) concluded that for eclogites formed under similar
conditions, a continuous shift in the tie-lines between the coexisting minerals is a
function of bulk composition. Rotating tie lines, on the other hand, were thought to
suggest T—P variations. Although this conclusion is considered by some workers to
be of a limited interpretative value (cf. Morgan, 1970), the remarkably parallel
trends of the tie-lines between the coexisting minerals, especially garnet-
clinopyroxene pairs, in the Jijal granulites (Fig. 7) may be suggestive of limited
variations in temperature and pressure.
Equilibrium texture in the hornblende-bearing rocks, systematic distribution of
elements, and consistency of temperature estimates involving rocks containing
variable amounts of hornblende suggest that relatively high PHi0 during the
granulite facies metamorphism was responsible for the formation of hornblende
(and, in some cases, epidote). It is possible, as suggested by Leake (personal
communication to M.Q.J.), that patches of the terrain were dehydrated and the
released water moved into structural 'sink' areas as well as moving outwards and
upwards to be lost, causing the development of hornblende in the rocks through
118 M. Q . J A N A N D R. A . H O W I F
Ca

FIG. 7. Mg — (Fe 2+ + Fe 3+ + Mn) — Ca plots of the coexisting minerals from the garnet granulites of Jijal.
Solid circles—clinopyroxenes, open circles—orthopyroxenes, solid triangles—garnets, solid squares—
amphiboles. Tie-lines have been omitted in some cases for the sake of clarity. The parallel trend of the
garnet-clinopyroxene tie-lines may be suggestive of similar metamorphic conditions (Yoder & Tilley, 1962).
which it passed. The abundance of the hornblende rocks, and epidote, in the
southern marginal part of the granulites lends some support to this idea, whereas
the mixtures of hornblende patches, garnetites, and plagioclase-granulites may in
part be due to severe structural disturbances. The chemical nature of the
hornblende rocks, characterized by a high Na/K ratio like the rest of the granulites,
may also be an indication that the fluids had an internal origin. Less likely, during
their upward transport the granulites were completely recrystallized locally under
high PHJ0 (water being of an external origin), such that the assemblages with
hornblende and epidote are not isofacial with the rest.
The Jijal granulites were called hornblende eclogite by Desio (1974). However,
evidence presented above, including the sections on garnets and clinopyroxenes,
and the abnormally large extent of the Jijal rocks, do not favour this view.
The ultramafic rocks. Temperature estimates for the ultramafic rocks are based
on a limited number of mineral analyses, i.e. chromite, olivine and two pyroxenes
from sample 275, two pyroxenes from 295, and the clinopyroxene from 272
(coexisting stably with enstatite which has not been analysed). Because of the
highly magnesian nature of the pyroxenes, experimental data on Ca-Mg pyroxene
solid solution can be applied to them. The results of the various geothermometers
are given in Table 9.
The plot of Fe 2+ /Mg for the olivine—clinopyroxene pair from 275 falls slightly on
the lower temperature side of the curve for chondrites considered to have
recrystallized at about 800 °C (Medaris, 1972). On O'Hara's (1967) grid, 295 and
275 give different T-P values and the pressures are so high that the rocks fall in the
field of garnet peridotites. Pressure estimates using A12O3 (Mori & Green, 1976)
and Ca/(Ca + Mg) (Mysen, 1976) in orthopyroxene in 275 and 295 range from 10
to more than 20 kb. Mysen & Boettcher (1975) found that the Cr-Al vi distribution
TABLE 9
Temperature estimates (°C)for the ultramafic rocks
; 2 3 4 5 6 7 8 9
275 974° 850° 900° 910° 930° 900° 800° 900° 600°
272 900° 890° 910° 900° 800°
295 971° 835° 875° 870° 890° 900° 800°
Based on:

Two-pyroxene geothermometers: (1) Wood & Banno (1973); (2) Perchuk & Vaganov (1977).
(3) Ca:Mg:Fe clinopyroxene coexisting with orthopyroxene (Fleet, 1974). Ca/(Ca + Mg) of clinopyroxene
in coexistence with orthopyroxene: (4) Davis & Boyd (1966); (5) Mysen (1976); (6) Warner & Luth (1974);
(7) Lindsley & Dixon (1976), and Mori & Green (1976).
Irvine-Jackson Cr-spinel-olivine method: (8) after Medaris (1975); (9) after Fuji (1977).
in coexisting pyroxenes yields a pressure-independent geothermometer which seems

superior to that based on Ca/(Ca + Mg) in orthopyroxene coexisting with
clinopyroxene. Estimated temperatures according to this method are more than
1100 °C for the two ultramafic rocks as well as for the websteritic granulite 297a.
These relatively higher temperatures might record the closure of these pyroxenes to
redistribution of the trivalent cations (cf. O'Hara, 1977) during the earlier history
of the rocks.
From the estimates given in Table 9, it seems likely that the Jijal ultramafic rocks
equilibrated finally in the neighbourhood of 800-850 °C. Experimental work
suggests that spinel peridotites form at intermediate P between plagioclase- and
garnet-peridotites. A pressure range of 6-12 kb was suggested by the work of
Obata (1976) and 8-16 kb by that of O'Hara & Yarwood (1978) for the formation
of spinel peridotites at 800-850 °C, whereas the experimental results of Jenkins &
Newton (1979) give a possible interval for this transition of 9-15-12-5 kb at
900 °C. Pressures of 8-12 kb at these temperatures suggest that the rocks
equilibrated finally in an oceanic/suboceanic geothermal regime (Clark &
Ringwood, 1964).
SUMMARY AND CONCLUSIONS
The Jijal complex is a 150 sq. km wedge of garnet granulites with less
voluminous ultramafic rocks in the southern part. The latter are not in structural
harmony with the granulites, and petrographic and geochemical evidence suggests
that the two types of rocks are not comagmatic. There is a likelihood that the
ultramafic rocks were intruded in the granulites as solid material capable of plastic
flow, after both had been metamorphosed (? at different sites) but before the entire
complex was tectonically carried into its present position. To the north of the Jijal
complex (Fig. 1) are rocks of the amphibolite facies ( r < 700 °C; P 4-6 kb),
apparently over-riding the granulites. To the south are metasediments and granitic
gneisses in the greenschist to amphibolite facies. The garnet granulites are
chemically quite unlike the amphibolites (generally banded); the amphibolites have
sparse garnet. The contact zone between the amphibolites and granulites is
sheared and faulted; the fault to the south is undoubtedly regional and is considered
to represent the northern boundary of the Indian plate, onto which the Jijal
complex was obducted as the bottom of the island arc (see, for example, Tahirkheli
etai, 1979).
The granulites are characterized by an abundance of garnet and can be classified
into plagioclase-bearing (basic to intermediate) and plagioclase-free (ultrabasic to
basic) types. Out of the 17 assemblages identified, plagioclase + garnet +
clinopyroxene + quartz + rutile is the most abundant; it may be joined by
hornblende and/or epidote, in some to the exclusion of the clinopyroxene. The
plagioclase-free granulites are less abundant and are represented by rocks rich to

very rich in either garnet, hornblende, clinopyroxene, or epidote. They may form
mappable units, especially in the south, or may be in confused intimate mixtures
with the plagioclase-bearing type.
The granulites follow igneous rather than sedimentary trends on Niggli-type
diagrams, the departures being due to bulk mineralogy rather than to metasomat-
ism, except in the hornblende-rich rocks. Variation diagrams split them into at least
two series and it is not certain if the plagioclase-bearing and plagioclase-free rocks
are genetically related. However, chemical 'breaks' between basic to intermediate
rocks and their comagmatic ultrabasic rocks are not totally unexpected. The
intimate association and other aspects of the rocks can be better explained if they
are considered to be comagmatic rather than being the products of two or more
unrelated magmas. Since the intermediate rocks, with high Ca/(Na + K), and
ultrabasic rocks do not have chemical analogues among the common volcanic
rocks, it is considered that the entire range of the granulites represents a
metamorphosed differentiated plutonic sequence of norites, with minor
hypersthene-quartz diorites, olivine gabbros, troctolites, pyroxenites and
feldspathic allivalites.
Garnets in the granulites are essentially composed of almandine (27—43 per
cent), pyrope (28-46 per cent), and grossular (16-28 per cent). The latter two
components are higher than those found in garnets of most granulites and are
indicative of high-pressure conditions. High pressure is also suggested by the large
amount of A12O3 in the clinopyroxenes (up to 9-9 per cent), amphibole (up to 16-3
per cent), and orthopyroxene (up to 5-5 per cent), supplemented by the high
Alvi/Aliv ratios (up to 0-68) of their amphiboles, the common presence of rutile, the
abundance of garnet, and the high density (3-2—3-5) of the rocks compared with
that of the neighbouring rocks (<3-0). However, a number of factors, especially the
Alvi/Aliv ratios (about 1:1) and the presence of higher Tschermak's component
and/or acmite and lower jadeite components in the clinopyroxene suggest that
pressures were not those of the eclogite facies. It is estimated that the rocks
equilibrated within the ranges of 11-15 kb and 650-800 °C (perhaps 670-790 °C
and 12-14 kb).
The presence or absence of plagioclase is a consequence of bulk chemistry rather
than differences in pressure. Plagioclase disappeared more readily in the more
basic, Na-poor rocks at the same pressures under which it was stable in the less
basic, Na-rich rocks. There is a strong possibility that much of the amphibole and
some epidote are of primary metamorphic origin formed under elevated P H o , the
necessary water being provided internally when most of the granulites were being
dehydrated. However, much, possibly all, of the epidote in the epidote-rich rocks
may be of retrograde origin.
The Jijal granulites are poor in K, Rb, Ni, and Cr. Although some K, Rb, (? and
Na) may have been expelled to upper levels of the Earth's crust during
metamorphism, the magma may well have been depleted in the four elements. The
Jijal rocks have analogues in the calc-alkaline series but their Na/K ratios are
closer to those of oceanic tholeiites (Engel et al., 1965; Bryan et al., 1976) than to
other volcanic rocks. They may have been derived from a tholeiitic magma of

oceanic affinity or they may be genetically related to the pyroxene granulites of
Swat considered to have originally crystallized from an andesitic basalt
(calc-alkaline) magma of island arc/continental margin affinity. The Jijal rocks
may have originally crystallized in the Old Tethyan crust upper mantle or an
ancient continental margin and then later subducted to great depth (~40 km)
during the collision of the Indian-Asian landmasses, metamorphosed to granulites,
and brought up during later Himalayan orogenic episode(s).
In a road cut at Patan, a room-sized relic block of noritic rock traversed by veins
of garnet granulite lends support to the view that the garnet granulites may
represent high-pressure equivalents of norites and related rocks, i.e. they may
represent a separate (perhaps basal) part of the huge Swat pyroxene granulite
'lopolith', with the plagioclase-free garnet granulites representing a cumulate
portion.
The ultramafic rocks are mostly represented by diopsidites; peridotites, dunites
and websterites together make less than 50 per cent by area of the outcrops.
Serpentinization is mild and selective, except along shear zones, and texture and
petrography suggest that they are alpine-type in nature. The mole per cent
MgO/(MgO + Fe total + MnO) and A12O3 contents of the rocks, and the
composition of clinopyroxene (Mg49.5-4gFe2.8-6) and olivine (F092-89), are similar to
those of the harzburgite subtype ultramafic complexes (Jackson & Thayer, 1972).
The lack of plagioclase and garnet and the low A12O3 content of their pyroxenes
indicate intermediate pressures for the Jijal ultramafic rocks. Their present
mineralogy also is suggestive of equilibrium conditions in the granulite facies
(800-850 °C and 8-12 kb). The rocks are depleted in alkalis and alumina, and are
thus not capable of producting basaltic magmas on partial melting.
The abundance of diopsidites is in marked contrast to other alpine complexes
where olivine-rich rocks are more abundant. It is tempting to connect the Jijal
ultramafic rocks and granulites magmatically with the Swat noritic granulites (Jan
& Howie, 1980) found to the north. However, the distinct chemical and
mineralogical differences, especially in the pyroxenite members from the three
suites, raise serious objections to such an interpretation (Jan, 1977). The possibility
of Ca and Si metasomatism for the production of abundant diopside during or
before the granulite facies metamorphism should not be ruled out. At least the thin
diopsidite veins in the peridotites seem to have formed by Ca + Si metasomatism
above 500 °C.
It is difficult to decide whether the Jijal ultramafic rocks represent a faulted
122 M . Q. J A N A N D R. A . H O W I E
suboceanic crust/upper mantle slab, mantle diapirs emplaced into deep orogenic
roots, or dismembered ultramafic rocks differentiated from a basaltic magma. The
T—P estimates for their final equilibration suggest an oceanic or suboceanic rather
than continental thermal regime. The complexity of the origin of alpine peridotites
has been illustrated by Wyllie (1967, 1969). Like most other alpine peridotites
(Den Tex, 1969), the Jijal ultramafic rocks have a complex history. One of the
earlier events in their evolution is possibly recorded by the closure of their
pyroxenes to the redistribution of Cr (? and other trivalent cations) at temperatures
greater than 1100 °C. Amongst the latest events is their low grade and selective

'metasomatism' (development of serpentine, talc, tremolite) which may be
contemporaneous with, and post-date, the upward migration of the Jijal tectonic
wedge.
The Jijal garnet granulites may represent a portion of the high-pressure part of a
paired metamorphic belt in the southern part of Kohistan (the lower pressure belt
being represented to the north of the Jijal belt by amphibolites and pyroxene
granulites).
ACKNOWLEDGEMENTS
Critical reading of the manuscript by Dr D. R. C. Kempe, the useful suggestions
of Dr J. N. Walsh during the chemical analyses, the microprobe analyses by
Messrs R. F. Symes, Ian Young, and Cavit Mahmut, and the 2V determinations by
Dr M. Majid are gratefully acknowledged. Professor M. J. O'Hara and Dr L. L.
Perchuk kindly supplied pre-prints of their papers. M.Q.J. is indebted to the British
Council for the award of a grant which enabled him to study in England for more
than three years.
REFERENCES
AHMED, Z., 1978. Chromite from Sakhakot-Qila area, Malakand Agency, Pakistan. Miner. Mag. 42, 155-7.
AKELLA, J., 1976. Garnet pyroxene equilibria in the system CaSiO 3 -MgSiO 3 -Al 2 O 3 and in a natural mineral
mixture. Am. Miner. 61, 589-98.
& KENNEDY, G. C , 1971. Studies on anorthite + diopside-hedenbergite50 at high pressures and
temperatures. Am. J. Sci. 270, 155-65.
AOKI, K., & KUSHIRO, I., 1968. Some clinopyroxenes from ultramafic inclusions from Hinome-gata, Japan. J.
Jap. Ass. Miner. Petrol, econ. Geol. 68, 303-10.
BANNO, S., 1970. Classification of eclogites in terms of physical conditions of their origin. Phys. Earth planet.
Interiors, 3,405-21.
BARTH, T. F. W., 1936. Structural and petrological studies in Dutchess County, New York. Part II: Petrology
and metamorphism of the Paleozoic rocks. Bull. geol. Soc. Am. 47, 775-850.
1962. Theoretical Petrology. New York: Wiley.
BILGRAMI, S. A., 1968. Geology and chemical mineralogy of the Zhob valley chromite deposits, West
Pakistan. Amer. Miner. 54, 134-48.
BINNS, R. A., 1962. Metamorphic pyroxenes from the Broken Hill district, New South Wales. Miner. Mag. 33,
320-38.
BRYAN, W. B., THOMPSON, G., FREY. F. A., & DICKEY, J. S., 1976. Inferred geologic settings and
differentiations in basalts from the Deep-Sea Drilling Project. J. geophys. Res. 81, 4285-304.
BUDDINGTON, A. F., 1950. Composition and genesis of pyroxene related to Adirondack anorthosite and
anorthosite-marble contact zone. Am. Miner. 35, 659-70.
1952. Chemical petrology of some metamorphosed Adirondack gabbroic, syenitic and quartz syenitic
rocks. Am. J. Sci. Bowen vol. 37-84.
CHALLIS, G. A., 1965. The origin of New Zealand ultramafic intrusions. / . Petrology, 6, 322-364.
CHURCH, W. R., 1968. Eclogites. In: HESS, H. H., & POLDERVAART, A. (eds.), Basalts. New York: Wiley
Interscience, 755-98.
CLARK, S. P., & RINGWOOD, A. E., 1964. Density distribution and constitution of the mantle. Rev. Geophys. 2,
35-88.
COLEMAN, R. G., 1971. Plate tectonic emplacement of upper mantle peridotites along continental edges. J.
geophys.Res. 76, 1212-22.
1972. Blueschist metamorphism and plate tectonics. 24th Int. geol. Congr. 2, 19-26.
& KEITH, T. E., 1971. A chemical study of serpentinization—Burro Mountain, California. J. Petrology,
12,311-28.
LEE, D. E., BEATTY, L. B., & BRANNOCK, W. W., 1965. Eclogites and eclogites: their differences and
similarities. Bull. geol. Soc. Am. 76, 483-508.
DAVIDSON, L. R., 1968. Variations in ferrous iron-magnesium distribution coefficients of metamorphic
pyroxenes from Quairading, Western Australia. Contr. Miner. Petrol. 19, 239-59.
1971. Metamorphic hornblendes from basic granulites of Quairading district, Western Australia. Neues
Jb. Miner. Mh. 344-59.
DAVIS, B. C. T., & BOYD, F. R., 1966. The join Mg2Si2O6-CaMgSi2O6 at 30 kilobars pressure and its

application to pyroxenes from kimberlites. J. geoph. Res. 71, 3567-76.
DEER, W. A., HOWIE, R. A., & ZUSSMAN, J., 1962a. Rock-forming minerals, 1. London: Longman.
19626. Rock-forming Minerals, 5. London: Longman.
1978. Rock-forming Minerals, 2A (2nd ed.). London: Longman.
DEN TEX, E., 1969. Origin of ultramafic rocks, their tectonic setting and history: a contribution to the
discussion of the paper 'The origin of ultramafic and ultrabasic rocks' by P. J. Wyllie. Tectonophysics, 7,
457-88.
& VOGEL, D. E., 1962. A 'granulitgebirge' at Cabo Ortegal (N.W. Spain). Geol. Rdsch. 52, 95-112.
ENGELS, J. P., & VOGEL, D. E., 1972. A high-pressure intermediate-temperature facies series in the
Precambrian at Cabo Ortegal (northwest Spain). 24th Int. geol. Congr. 2, 64-73.
DESIO, A., 1974. Geological reconnaissance in the middle Indus valley between Chilas and Besham Qila
(Pakistan). Boll. Soc. geol. Ital. 93, 348-68.
DOBRETSOV, L. N., KHLESTOV, V. V., REVERDATTO, V. V., SOBOLEV, N. V., & SOBOLEV, V. S., 1972. The
facies of metamorphism (transl. by D. A. Brown). Canberra: Australian National University.
EDGAR, A. D., MOTTANA, A., & MACRAE, N. D., 1969. The chemistry and cell parameters of omphacite and
related pyroxenes. Miner. Mag. 37, 61-74.
EHRENBERG, S. N., 1975. Feather River ultramafic body, northern Sierra Nevada, California. Bull. geol. Soc.
Am. 86, 1235-43.
ENGEL, A. E. J., ENGEL, C. G., & HAVENS, R. G., 1965. Chemical characteristics of oceanic basalts and the
upper mantle. Ibid. 76, 917-34.
ERNST, W. G., 1978. Petrochemical study of lherzolitic rocks from the western Alps. / . Petrology, 19, 341-92.
FLEET, M. E., 1974. Mg-Fe 2+ site occupancies in coexisting pyroxenes. Contr. Miner. Petrol. 47, 207-14.
FRANCIS, D. M., 1976. The origin of amphibole in lherzolite xenoliths from Ninivak Island, Alaska. J.
Petrology, 17, 357-78.
FROST, D. F., 1975. Contact metamorphism of serpentinite, chloritic blackwall and rodingite at
Paddy-Go-Easy Pass, Central Cascades, Washington. Ibid. 16, 272-313.
FUJI, T., 1977. Fe-Mg partitioning between olivine and spinel. Yb. Carnegie Instn. Wash. 76, 563-9.
GANSSER, A., 1974. The ophiolite melange, a world-wide problem on Tethyan examples. Eclog. geol. Helv. 67,
479-507.
GREEN, D. H., 1964. The petrogenesis of the high-temperature peridotite intrusion in the Lizard area,
Cornwall. J. Petrology, 5, 134-88.
& RINGWOOD, A. E., 1967. An experimental study of the gabbro to eclogite transformation and its
petrological applications. Geochim. cosmochim. Ada, 31, 767-833.
1972. A comparison of recent experimental data on the gabbro-garnetgranulite-eclogite transition.
J. Geol. 80, 277-88.
HALL, A., & WALSH, J. N., 1969. A rapid method for the determination of fluorine in silicate rocks and
minerals. Analyt. Chim. Ada, 45, 341-2.
HAMLYN, P. R., 1975. Chromite alteration in the Panton Sill, East Kimberley region, Western Australia.
Miner. Mag. 40, 181-92.
HENSEN, B. J., 1976. The stability of pyrope-grossular garnet with excess silica. Contr. Miner. Petrol. 55,
279-92.
HERZBERG, C. T., 1978. Pyroxene geothermometry and geobarometry: experimental and thermodynamic
evaluation of some subsolidus phase relations involving pyroxenes in the system CaO-MgO-
AljOj-SiOj. Geochim. cosmochim. Ada, 42, 945-57.
HESS, H. H., 1949. Chemical composition and optical properties of common clinopyroxenes. Am. Miner. 34,
621-66.
HIMMELBERG, G. R., & LONEY, R. A., 1973. Petrology of the Vulcan Peak alpine-type peridotite, southwestern
Oregon. Bull. geol. Soc. Am. 84, 1585-1600.
HOWIE, R. A., 1965. The pyroxenes of metamorphic rocks. In: PITCHER, W. S., & FLINN, G. W. (eds.),
Controls of Metamorphism. Edinburgh: Oliver & Boyd, 319-26.
& SUBRAMANIAM, A. D., 1957. The paragenesis of garnet in charnockite, enderbite, and related
granulites. Miner. Mag. 31, 565-86.
124 M. Q . J A N A N D R. A. H O W I E
Hsu, L. C , 1968. Selected phase relations in the system Al-Mn-Fe-Si-O-H: a model for garnet equilibria. J.
Petrology, 9,40-83.
IL'VITSKIY, M. I., & KOLBANTSEV, R. V., 1968. Paragenetic types of olivine and statistical analysis of their
chemistry. Dokl. Acad. Sci. USSR, Earth Sci. Sec. 179, 135-7.
IRVINE, T. N., 1967. Chromium spinel as a petrogenetic indicator. Part 2. Petrologic applications. Can. J.
Earth Sci. 4, 71-103.
IRVING, A. J., 1974. Geochemical and high pressure experimental studies of garnet pyroxenite and pyroxene
granulite xenoliths from the Delegate basaltic pipes, Australia. / . Petrology, 15, 1-40.
JACKSON, E. D., & THAYER, T. P., 1972. Some criteria for distinguishing stratiform, concentric and alpine
peridotite-gabbro complexes. 24th Int. geol. Congr. 2, 289-96.
JAN, M. Q., 1977. The mineralogy, geochemistry, and petrology of Swat Kohistan, NW Pakistan. Unpublished

Ph.D. thesis, University of London.
1979. Petrography of the Jijal complex, Kohistan. Geol. Bull. Univ. Peshawar, 11, 31-50.
& HOWIE, R. A., 1980. Ortho- and clinopyroxenes from the pyroxene granulites of Swat Kohistan,
northern Pakistan. Miner. Mag. 43, 715-26.
& KEMPE, D. R. C , 1973. The petrology of the basic and intermediate rocks of Upper Swat, Pakistan.
Geol. Mag. 110,285-300.
& TAHIRKHELI, R. A. K., 1969. The geology of the lower part of Indus Kohistan, Swat. Geol. Bull. Univ.
Peshawar, 4, 1-13.
JENKINS, D. M , & NEWTON, R. C , 1979. Experimental determination of the spinel peridotite to garnet
peridotite inversion at 900 °C and 1000 °C in the system CaO-MgO-AI 2 O 3 -SiO 2 , and at 900 °C with
natural garnet and olivine. Contr. Miner. Petrol. 68, 407-19.
KRETZ, R., 1960. The distribution of certain elements among coexisting calcic pyroxenes, calcic amphiboles
and biotites in skarn. Geochim. cosmochim. Ada, 20, 161-91.
1963. Distribution of magnesium and iron between orthopyroxene and calcic pyroxene in natural mineral
assemblages./. Geol. 71, 773-85.
KUNO, H., 1968. Differentiation of basaltic magmas. In: HESS, H. H., & POLDERVAART, A. (eds.), Basalts.
New York: Wiley, 623-88.
KUSHIRO, I., 1969. Clinopyroxene solid solutions formed by reactions between diopside and plagioclase at high
pressures. Spec. Pap. mineralog. Soc.Am. 2, 179-91.
& YODER, H. S., 1966. Anorthite-forsterite and anorthite-enstatite reactions and their bearing on the
basalt-eclogite transformation. / . Petrology, 7, 337-62.
LAPPIN, M. A., & SMITH, D. C , 1978. Mantle-equilibrated orthopyroxene eclogite pods from the basal gneisses
in the Selje district, western Norway. Ibid. 19, 530-84.
LEAKE, B. E., 1964. The chemical distinction between ortho- and para-amphibolites. Ibid. 5, 238-54.
1965. The relationship between tetrahedral aluminium and the maximum possible aluminium in natural
calciferous and subcalciferous amphiboles. Am. Miner. 50, 43-51.
1971. On aluminous and edenitic hornblendes. Miner. Mag. 38, 389-407.
1978. Nomenclature of amphiboles. Ibid. 42, 533-63.
HENDRY, G. L., KEMP, A., PLANT, A. G., HARVEY, P. K., WILSON, J. R., COATS, J. S., AUCOTT, J. W.(
LUNEL, T., & HOWARTH, R. J., 1969. The chemical analysis of rock powders by automatic X-ray
fluorescence. Chem. Geol. 5, 7-86.
LINDH, A., 1975. Coexisting pyroxenes—a multivariate statistical approach. Lithos, 8, 151-61.
LINDSLEY, D. H., & DIXON, S. A., 1976. Diopside-enstatite equilibria at 815 ° to 1400 °C, 5 to 35 kb. Am. J.
Sci. 276,1285-1301.
Liou, J. G., 1973. Synthesis and stability relations of epidote, Ca 2 Al 2 FeSi 3 O l2 (OH). J. Petrology, 14, 381-413.
LONEY, R. A., HIMMELBERG, G. R., & COLEMAN, R. G., 1971. Structure and petrology of alpine-type
peridotite at Burro Mountain, California, U.S.A. Ibid. 12, 245-309.
LOOMIS, T. P., 1977. Kinetics of a garnet granulite reaction. Contr. Miner. Petrol. 62, 1-22.
MANNA, S. S., & SEN, S. K., 1974. Origin of garnet in the basic granulites around Saltora, W. Bengal, India.
Ibid. 44, 195-218.
MAXEY, L. R., & VOGEL, T. A., 1974. Compositional dependence of the coexisting pyroxene iron-magnesium
distribution coefficient. Ibid. 43, 295-306.
MEDARIS, L. R., 1972. High-pressure peridotites in southwestern Oregon. Bull. geol. Soc. Am. 83,41-58.
1975. Coexisting spinel and silicates in alpine peridotites of the granulite facies. Geochim. cosmochim.
Ada, 39,947-58.
& DOTT, R. H., 1970. Mantle derived peridotites in southwestern Oregon: relation to plate tectonics.
Science, 169,971-4.
MILLHOLLEN, G. L., & WYLLIE, P. J., 1974. Melting relations of brown-hornblende mylonite from St. Paul's
rocks under water-saturated and water-undersaturated conditions to 30 kilobars. J. Geol. 82, 589-606.
MIYASHIRO, A., & SHIDO, F., 1973. Progressive compositional change of garnet in metapelite. Lithos, 6,
13-20.
MOODY, J. B., 1976. Serpentinization: a review. Ibid. 9, 125-38.
MOORES, E.M., 1969. Petrology and structure of the Vourinos ophiolite complex, northern Greece. Spec. Pap.
geol. Soc.Am. 118.
1973. Geotectonic significance of ultramafic rocks. Earth Sci. Rev. 9, 241-58.

MORGAN, B. A., 1970. Petrology and mineralogy of eclogite and garnet amphibolite from Puerto Cabello,
Venezuela. J. Petrology, 11, 101-45.
1974. Chemistry and mineralogy of garnet pyroxenites from Sabah, Malaysia. Contr. Miner. Petrol. 48,
301-14.
MORI, T., & BANNO, S., 1973. Petrology of peridotite and garnet clinopyroxenite of the Mt. Higasi-Akaisi
mass, Central Sikoku, Japan—subsolidus relation of anhydrous phases. Ibid. 41, 301-23.
& GREEN, D. H., 1976. Subsolidus equilibria between pyroxenes in the CaO-MgO-SiO 2 system at
high pressures and temperatures. Am. Miner. 61, 616-25.
MOTTANA, A., 1974. Melting of spessartine at high pressures. Neues. Jb. Miner. Mh. 256-71.
Mum, I. D., 1951. The clinopyroxenes of the Skaergaard intrusion, eastern Greenland. Miner. Mag. 29,
690-714.

MYSEN, B. O., 1976. Experimental determination of some geochemical parameters related to conditions of
equilibration of peridotite in the upper mantle. Am. Miner. 61, 677-83.
& BOETTCHER, A. L., 1975. Melting of a hydrous mantle: II. Petrology, 16, 549-93.
NEMEC, D., 1967. The miscibility of pyralspite and grandite molecules in garnets. Miner. Mag. 36, 389-402.
OBATA, M., 1976. The solubility of A12O3 in orthopyroxenes in spinel and plagioclase peridotites and spinel
pyroxenite. Am. Miner. 61, 804-16.
O'HARA, M. J., 1967. Mineral parageneses in ultrabasic rocks. In: WYLLIE, P. J. (ed.), Ultramafic and Related
Rocks. New York: Wiley, 393-403.
1977. Thermal history of excavation of Archaean gneisses from the base of the continental crust. / . geol.
Soc.Lond. 134, 185-200.
& YARWOOD, G., 1978. High pressure-temperature point on an Archaean geotherm, implied magma
genesis by crustal anatexis, and consequences for garnet-pyroxene thermometry and barometry. Phil.
Trans. R. Soc. Lond. A288,441-56.
PERCHUK, L. L., 1967. Analysis of thermodynamic conditions of mineral equilibria in amphibole-garnet rocks.
Izv. Akad. Nauk SSSR. Ser. Geol. 3, 57-83 (in Russian).
1968. Pyroxene-garnet equilibria and the depth facies of eclogites. Int. Geol. Rev. 10, 280-318.
1969. The effect of temperature and pressure on the equilibrium of natural iron-magnesium minerals.
Ibid. 11,875-901.
& VAGANOV, V. I., 1977. Temperature regimes of formation of continental volcanic series. Preprint, Inst.
Exper. Miner., Acad. Sci. USSR.
PINUS, G. V., AGAFONOV, L. V., & VELINSKIY, V. V., 1970. Eclogite-like rocks of the Anadyr-Koryak folded
systems. Pacif. Geol. 2, 81-91.
QURESHI, A. A., & JAN, M. Q., 1977. The ultramafic and high-grade metamorphic rocks near Jijal, Swat, NW
Pakistan. AMR. Accad. naz. Lincei, 63, 259-68.
RAASE, P., 1974. Al and Ti contents of hornblende, indicators of pressure and temperature of regional
metamorphism. Contr. Miner. Petrol. 45, 231-6.
RAHEIM, A., & GREEN, D. H., 1975. P, T path of natural eclogites during metamorphism—a record of
subduction. Lithos, 8, 318-28.
RAO, V. D., & SATYANARAYANA, K., 1971. Co-existing garnet and hornblende from the amphibolites of
Mysore State, India. Bull. Natn. Geophys. Inst. 9, 123-32.
RAY, S., & SEN, S. K., 1970. Partitioning of major exchangeable cations among orthopyroxene, calcic
pyroxene and hornblende in basic granulites from Madras. Neues. Jb. Miner. Abh. 114, 61-88.
RILEY, J. P., 1958. The rapid analysis of silicate rocks and minerals. Analyt. Chim. Ada, 7, 189-202.
RODGERS, K. A., 1973. Chrome spinels from the Massif du Sud, northern New Caledonia. Miner. Mag. 39,
326-39.
Ross, C. S., FOSTER, M. D., & MYERS, A. T., 1954. Origin of dunite and olivine-rich inclusions in basaltic
rocks. Am. Miner. 39, 393-437.
SAXENA, S. K., 1968. Distribution of elements between coexisting minerals and the nature of solid solution in
garnet./AW. 53, 994-1014.
1979. Garnet-clinopyroxene geothermometer. Contr. Miner. Petrol. 70, 229-35.
& HOLLANDER, N. B., 1969. Distribution of iron and magnesium in charnockitic rocks, Halmstadt,
Sweden. Neues. Jb. Miner. Mh. 85-92.
SCHARBERT, H. G., & KUART, G., 1974. Distribution of some elements between coexisting ferromagnesian
minerals in Moldanubian granulite facies rocks, Lower Austria, Austria. Tschermaks Miner. Petrogr.
Mitt.2\, 110-34.
SEN, S. K., & MANNA, S. S., 1976. Patterns of cation fractionation among pyroxenes, hornblende and garnet in
the basic granulite of Saltora, West Bengal. Indian J. Earth Sci. 3, 117-28.
SIGHINOLFI, G. P., 1971. Investigations into deep crustal levels: fractionation effects and geochemical trends
related to high-grade metamorphism. Geochim. cosmochim. Ada, 35, 1005-21.
SOBOLEV, N. V., KUZNETSOVA, I. K., & ZYUZIN, N. I., 1968. The petrology of grospydite xenoliths from the
Zagadochnaya kimberlite pipe in Yakutia. J. Petrology, 9, 253-80.
SPRINGER, R. K., 1974. Contact metamorphosed ultramafic rocks in the Western Sierra Nevada Foothills,
California. Ibid. 15, 160-95.
126 M. Q. J A N A N D R. A. H O W I E
SUBRAMANIAM, A. P., 1962. Pyroxenes and garnets from charnockites and associated granulites. In: ENGEL,
A. E. J. el al. (eds.), Petrologic Studies. A volume in honor of A. F. Buddington. Geological Society of
America, 21-36.
SWEATMAN, T. R., & LONG, J. V. P., 1969. Quantitative electron-probe microanalysis of rock-forming
minerals. J. Petrology, 10,332-79.
TAHIRKHELI, R. A. K., MATTAUER, M., PROUST, F., & TAPPONNIER, P., 1979. The India Eurasia suture zone
in northern Pakistan: synthesis and interpretation of recent data at plate scale. In: FARAH, A., & D E
JONG, K. A. (eds.), Geodynamics of Pakistan. Quetta: Geological Survey of Pakistan, 125-30.
THAYER, T. P., 1960. Some critical differences between alpine-type and stratiform peridotite-gabbro
complexes. 21st Int. geol. Congr. 13, 247-59.
1967. Chemical and structural relations of ultramafic and feldspathic rocks in alpine intrusive complexes.
In: WYLLIE, P. J. (ed.), Ultramafic and Related Rocks. New York: Wiley, 222-39.

1970. Chromite segregations as petrogenetic indicators. Spec. Publ. geol. Soc. S. Afr. 1, 380-90.
THOMPSON, R. N., 1974. Some high-pressure pyroxenes. Miner. Mag. 39, 768-87.
THORNTON, C. P., & TUTTLE, O. F., 1960. Chemistry of igneous rocks. 1. Differentiation index. Am. J. Sci.
258, 664-84.
VELINSKIY, V. V., PINUS, G. V., LESNOV, F. P., & VASIL'YEV, Y. R., 1970. Chemical characteristics of
ultramafic rocks belonging to different magmatic associations and their petrogenetic significance. Dokl.
Acad. Sci. USSR, Earth Sci. Sect. 191, 141-3.
VINOGRADOV, A. P., 1962. Average contents of chemical elements in the principal types of igneous rocks of the
Earth's crust. Geochemistry, 641-64.
VIRGO, D., & HAFNER, S. S., 1970. Fe 2+ , Mg order-disorder in natural orthopyroxenes. Am. Miner. 55,
201-23.
WARNER, R. D., & LUTH, W. C , 1974. The diopside-orthoenstatite two-phase region in the system
CaMgSi 2 O 6 -Mg 2 Si 2 O 6 . Ibid. 59, 98-109.
WEEDON, D. §., 1970. The ultrabasic/basic igneous rocks of the Huntly region. Scott. J. Geol. 6, 26-40.
WELLS, P. R. A., 1977. Pyroxene thermometry in simple and complex systems. Contr. Miner. Petrol. 62,
129-39.
WHITE, A. J. R., 1964. Clinopyroxenes from eclogites and basic granulites. Am. Miner. 49, 883-8.
WILKINSON, J. F. G., 1973. Pyroxenite xenoliths from an alkali trachybasalt in the Glen Innes area,
northeastern New South Wales. Contr. Miner. Petrol. 42, 15-32.
1974. Garnet clinopyroxenite inclusions from diatremes in the Gloucester area, New South Wales,
Australia. Ibid. 46,275-99.
WILSON, A. D., 1955. A rapid method for the determination of ferrous iron in rocks. Bull. geol. Surv. Gt. Br.
9, 56-8.
WINKLER, H. G. F., 1974. Petrogenesis of Metamorphic Rocks. New York: Springer Verlag.
WOOD, B. J., 1976. The partitioning of iron and magnesium between garnet and clinopyroxene. Yb. Carnegie
Instn. Wash. 75,571-4.
1977. The activities of components in clinopyroxene and garnet solid solutions and their application to
rocks. Phil. Trans. R. Soc. Lond. A281, 101-11.
& BANNO, S., 1973. Garnet-orthopyroxene and orthopyroxene-clinopyroxene relationships in simple
and complex systems. Contr. Miner. Petrol. 42, 109-24.
WYLLIE, P. J. (ed.), 1967. Ultramafic and related rocks. New York: Wiley.
1969. The origin of ultramafic and ultrabasic rocks. Tectonophysics, 7, 437-55.
YODER, H. S., & TILLEY, C. E., 1962. Origin of basalt magmas: an experimental study of natural and synthetic
rock systems. J. Petrology, 3, 342-532.
ZOLOYEV, K. K.., 1970. Chemical basis of the quantitative mineralogic classification of alpine-type ultramafic
rocks. Dokl. Acad. Sci. USSR, Earth Sci. Sect. 192,417-20.

1980 Qasim Jan

Uploaded by

Copyright:

Available Formats

1980 Qasim Jan

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1980 Qasim Jan

Uploaded by

Copyright:

Available Formats

The Mineralogy and Geochemistry of the

Metamorphosed Basic and Ultrabasic Rocks of the

by M. QASIM JAN 1 and R. A. HOWIE 2

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

cases to ±0-03. Refractive indices, measured at room temperature, are accurate to

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

The garnet granulites

Assemblages Mineral phases (+ =: present, 0 = absent)

Typel No Plag Gar Cpx Hbl Epi (etc)

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Type 2 Garnet-rich types + rutile and other opaque minerals

(Clino)zoisite-rich types, with minor rutile (sphene) and/or opaque minerals

The 17 mineral assemblages identified in the granulites are listed in Table 1,

Comparison with other areas

garnet-clinopyroxene-plagioclase assemblage in that area is less extensive. The

The ultramafic rocks

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

The garnet granulites

•O — ' T ' T ' — © — f - 00 II

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

relationship between the plagioclase-bearing basic and intermediate rocks (close

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

own compounds, is preferentially accommodated in pyroxenes, whilst Ni enters

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

The ultramafic rocks

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

proportions and, to a lesser extent, the composition of the three principal

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

SiO2 40-90 003 007 Co 140 441 705 630

99-91 98-84 98-30 99-07

Number of ions on the basis of four (anal. I) or thirty-two oxygens

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Orthopyroxenes from the Jijal complex

SI 275 SI 295 SI 297a SI320E SI 104b

SiO 2 56-72 55-20 53-27 53-75 51-79

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

TOTAL 99-42 99-39 99-59 99-21 99-55

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Seventeen clinopyroxene compositions, three of them from the ultramafic rocks

V 68 65 87 132 310 200 472 224 — — 547 608 470 375

by the clinopyroxenes from harzburgite-subtype ultramafic rocks; their A12O3

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

correlation between Na and Fe 3+ in the clinopyroxene analyses justifies it.

Compositions of five primary amphiboles analysed range from alumino-

Fe-Mg DISTRIBUTION BETWEEN THE COEXISTING MINERALS

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

where XMg refers to mole fraction Mg/(Mg + Fe 2+ ) in minerals A and B. Thus

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015

TEMPERATURE-PRESSURE ESTIMATES FOR THE JIJAL ROCKS

Downloaded from http://petrology.oxfordjournals.org/ at University of California, San Fransisco on April 7, 2015