GEOL. C ROAT. SOil 7 - IS 3 Figs. 3 Tabs. ZAGREB 1997 

Compositional Zoning in Amphibole from Amphibole Bearing Parageneses 
of West Psunj (Croatia): Evidence for Progressive Metamorphism? 

Orazen BALEN, Vladimir BERMANEC and Vera MARCl 

Key words: Amphibo le, Amphibolite , Geot hcrmomc­
try, Gcobaromctry, PSlInj . Croatia. 

Abstract 
Amphibo le bc(u"ing paragcncscs from the wes tern par I of ML 

Psunj (CroaTia) record evidence of prograding metamorphism. Opti­
cal and mic roprobe anal yses, together with thcrm obarornetric evalua­
ti ons 0 11 amphibole beari ng paragcncscs, show a zonal ion wit h Si­
and Mg-conccll lralions decreas ing from core to ri m al ong wilh 
increasi ng A I-, N (I-. and Ti -contenl. Changes in the chem ical compo­
si tion of am phibo le g rai ns ;I f e interpreted thro ugh coupled su bstiw­
lions. and reac ti ons w ith co-ex isting minerals duri ng an increase in 
mctamorphic conditions from greenschi st 10 amphibo lite facies . The 
change in P-T conditions recorded in the growth of amphibole grains 
(gcncral prograding patt crn) togcthcr with ch:mgcs in thc modal com­
positions in rclated paragcncscs could bc intcrprcted in a model of a 
subduction zone. 

1. INTRODUCTION 

Minerals or the amphibole group are common rock­
formin g minera ls in metamorphic rocks. They are sta­
ble over a wide range of pressu res and temperatures i.e. 
in various grades of metamorphism. The associa tions of 
amphibole with plagioclase, epidote, clinozoisite, chlo­
rite, sphene, e tc. arc common in metamorphosed mafic 
rocks of the grccnschi sl facics, epidote-amp hibol ite 
facies and amphibol ite facies. 

Many authors recognize similar assemblages, usual­
ly called "commo n assemblages" (LAIR D & ALBEE, 
198 1) for mafi c schis ts of greenschist to amphibo lite 
fac ies all over the world. In contrast to mafic schi sts , in 
metape litic rocks eve n small in cremental changes in 
pressure and temperature can be dete rmined by the sim­
ple recognition of key minerals. Changes of metamor­
phic grade in mafic schists are less obviolls. Dillcrenccs 
between metamorphic facies may be recorded through 
continuous reac tions wh ich produce " fin gerp rint s" in 
the chemi ca l con1pos it ion of phases and the modal 
composition of minera ls. These, less obvious changes 
may be follo wed through changes in opti cal properties 
and the information recorded in the zone o f" growth. In 

Uni ve rsi ty of Zagreb, r acuJty o f Scicnces, Department of Geology, 
Institu tc for M iner<lJ ogy and Pctrology, Horvatov<lC bb. HR-l0000 
Zagrcb, Croatia. 

general the appearance or disappearance of phases (i.e. 
first appearance of hornblende in epidote-amphiboli te) 
a re only revealed when great changes in P-T conditions 
occur and small cha nges of P-T cond itio ns is easy to 
overlook, especially in the field. 

The assemblage stabl e in the greenschi st zone con­
sists of albite + epidote + chlorite + acti nolite + qU<:ll"tz + 
sph ene (temperatu re below 370°C). Tn the trans ition 
zone an assemblage appears wh ich includes periste rite 
pairs + epidote + chlorite + Ca-amphibole (usua lly both 
ac tinolite and hornbl ende) + quartz + sphene and the 
temperature varies be tween 370-420°C. The amph ibo­
li te facies is characterized by Ca-plagioclase (A n2o.50) + 
Cn-amphibole (usually homblende) + chl orite + sphene 
+ il menite (MARUYAMA et aI. , 1982). 

Geologic cv idence sugges ts that the transition from 
the greenschi st to the epidot e-amphibo li te fa c ies and 

from the epidote-amphibolite to amphibolite fac ies may 
represent a P-T fi e ld in which a number of the c ritical 
equi li brium curves are c losely spaced and inte rsect each 
other (APTED & LlOU , 1983). Therefore the recogn i­
tion of P-T condi tions in maf"ie schists is a difficult task 
but in favorable circumstances oscillations of metamor­
phism could be recorded and the quant ita tive value of 
mctamorphic variablcs cou ld be approximatcd. 

Petrolog ical study of the ultramafic rocks lOge ther 
with known data from associated amphibolitcs in the 
area of the Slavonian mountains suggests that the inves­
tigated rocks belong to an ophiolite com plex (pAMIC 
& MARCI, 1990; PAM IC & LANPHERE, 199 1) 

The purpose of this study is to de termine the mi neral 
asscmblages that coexist in equilib ri um, and to es timate 
(th rough chemical data of" the amph iboles) the li mits of 
tcmperature and pressure dur ing metamorphism of the 
amphibole-bearing rocks of wcstern PSLlnj. Thc growth 
of amphibole under changing P-T condi tions lllay serve 
as an indi cator of the geotectonic evolut ion of thc inves­
tigated area. 

2. HISTORICAL REVIEW 

The re arc numerou s pape rs published on th e c rys­
tal line rocks from the Slavonian mountains (for a com ­
prehens ive li s t of references see PAM Ie & LAN­
PHER E, 1991). KISPATI C (1 892) produ ced the first 
petrograph ic and geologic data for various metamorphic 



Geologia Croatica 501l 

O·~ _____ ~,5 km 

0DD[SgE] 

Fig. I Generalized geological map of 
Mt. Psunj from PAMle & LAN­
PHERE (1991). Legend: 0 - study 
area; I) Tertiary and Quaternary 
sediments; 2) Tertiary volcanic 
rocks; 3) Mesozo ic sediments; 4) 
metamorphic complex with meta­
basites; 5) metamorphic complex 
without metabasites; 6) migma­
tites; 7) S-type granitoids; 8) 
amphibolite facies; 9) greenschist 
facies; 10) amphibolite interlay­
ers; II) I-type granites; 12) COll­
tactlinc;13) horizontal fault; 14) 
norma! fault; 15) reverse fault; 
16) unconformity; 17) isoclinal 
folds; 18) foliatioll. 

6 7 8 9 10 II 

,~, 
]~ ]5 16 

rocks and indicated the importance of index minerals 

and assemblages. MARCI (1965) and TAJDER (1969) 

focused their altention on the western part of Mt. Psunj 
and on the mineralogical composition of amphibole­
bearing rocks. Recently PAMIC & MARCT (1990) and 

PAMIC & LANPHERE (1991) examined the amphibo­

lites petrographically and showed that the amphibolite 

of the Slavonian mountains is regular member of the 

Hercynian progressive metamorphic complex. The 

chemical composition of the amphibolitic rocks corre­
lates with oceanic tholeiitc. Nevertheless, thc geotec­

tonic setting is not reliably determined. 

3. STUDY AREA 

The study area is delimited by the creeks Rasaska 
and Jezeriea 10 the south, Sumetljica to the north, Bre­

zovo Polje peak in the east, and Ihe granite body of 

Omanovac to the west (Fig. 1). 

Amphibolite and amphibole schists of west Psunj 
occur as lenses up to 10-20 m in size. There arc also 

occurrences of small bodies of metagabbro in the study 
area. The investigated rocks (amphibolite, amphibole 

schist, metagabbro) occur in discordant position to the 
surrounding rocks (granite, sediments), and are proba­

bly Precambrian in age (JAMICIC et aI., 1989). Later 
intrusion of the granite body resulted in retrograde 
altcration of the amphibole bearing rocks. 

-#%-
17 

"-
18 

4. MATERIALS AND METHODS 

The mineral parageneses of amphibole bearing 
schists, metagabbro and amphibolite were examined by 
standard polarizing microscope and detailed optical 
measurements were made using a Carl Zeiss Jcna 
microscope equipped with a five axes universal stage 
and conoscopic illumination. 

Selected grains werc chcmically analyzed by elec­

Iron microprobe at the Univers ity of Bern. The electron 
microprobe used was a CAMECA SX 50 instrument 
equipped with WDS spectrometers with LiF, PET and 
TAP diffraction crystals. A beam current of 20 nA, 
accelerating voltage 15 kY, peak counting 20 sec, and 
beam diameter less than 1 ~m were the operating COIl­

ditions. Synthetic standards used included fayalite (Fe), 
spinel (Mg), tephroite (Mn), ilmenite (Ti), eskolaite 
(Cr) together with natural standards: orthochlase (Si, 
K), albite (Na), anorthite (Ca, AI). Instrument control 

and on-line data reduction were performed using the 
CAMECA version of the PAP procedure (POUCI-IOU 
& PICHOIR. 1984). 

The amphibole formulae were generated with the 
MINFILE software (AFIFI & ESSENE, 1988) on the 
basis of 230 and 13 cations + Ca + Na + K. 



1l:l ICll, Bermanec & r\'1:1rci: Compositional Zoning in Amphibole from Amphibole Bearing Paragencscs ... 

S. RESULTS 

5.1. PETROGRAPHY 

Four rock types we re recognized in the amphibole 
bearin g rocks of western MI. Psunj on the bas is of 
mic roscopi c study , mine ral content and tex tural fea­
tures, and are listed as follows: 

A) e pidote-amphibole schi st: 

amphibole (act inol ite ancl hornblende), epidote, qua­

rt z, oli goclase, albite ± ilmenite, sphene, chlorite; 

B) amphibolite: 

amphibo le (hornblende), oligoclase ± cl inozoisitc, 
chlorite, ilme nite , sphene , quartz; 

C) metagabbro: 

amphibo le (hornbl e nd e), oligoclase, clinozo is itc, 
ilme nite, sphene ± e pidote, chlorite. cli nopyroxene, 
quartz; 

D) "granitized" amphibo lite: 

amphibole (hornb le nde) , oligoclase, c linozo isite, 
biotite, ilmcn ite , sphene ± quartz, e pidote, micro­

c line. 

Thc mine rals are lis ted in order of abundance. 
Some minerals (c hlorite, c linozois ite, ... ) which are 

commo n in the rock types arc produc ts of late r retro­
grade me tamorphism and do not represent an equi li bri­
um state with the other mine rals. Also, clinopyroxene­
amphibole-ch lorite +clinozoisite alterations imply a ret­

rog rade trend. These re trograde changes are allribut ed 
to c hanges of the metamorphic conditions during intru­
sion o f the granite body. 

Amphibole 

Amphibole occurs as xenoblasts in all the paragenc­
ses. In asscmblage A it a lso OCCUI'S as needle like aggre­
gat es with weak green-ye llow pleochroi sm. During 
growth, amphibole replaced pyroxene and encompassed 

o the r mine ral grains (plagioclase, ilme nitc, sphe ne). 
Within thc cracks ins ide the amph ibol e grains aggre­
ga tes o r chlorite and cl inozoisi te were fOfmed. Amphi­
bo les in epidote-amphibo le schists of the ne ighbouring 
g ranit e body often show a lteration to biotit e. This is 
c harac ter ized by the changc rrom a grcen coloured 

amphibole at on e e nd or a grain toward th e brown 
colour or biotite at the o the r end of the same grain (in 
sections parallc llo the c-axis or amphibole). 

Mcasurements or the optical parame ters of selected 
gra in s of amphibol e indi cated s ix different types of 
amphibole: 

a) Needle like actinolit e with a maximum ex tinc tion 
angle Z"c = 12° and optical axial angle 2V = _74°; 
pale pleochroi sm: X = Y = yel low , Z = green. It was 
observed in type A assemblage. 

b) Hornblende with a maximum extinction angle Z"c;;;; 

19° and optical axial angle 2V = _64°; pleochroi sm: 
X = yellow, Y = green , Z = bluish-green, observed 
in type A and B assemblages. 

c) Hornble nde with a maximum extinction angle ZAC ;;;; 
16° and optical axial angle 2V ;;;; _72°; pleochroism: 
X ;;;; yellow, Y = green, Z = dark green, observed in 

type C assemblage. 

d) Hornblende with maximum ex tinction angle Z"c = 
30° and optical axial angle 2V ;;;; +87°; pleochroism: 

X ;;;; yellow, Y ;;;; grecn, Z = dark green, observed in 

type C assemblage. 

e) Hornble nde with maxirnum eX linction angle ZAe ;;;; 

18° and optical axial angle 2V = _89°; pleochroism: 

X = yellow, Y = green, Z ;;;; da rk green, observed in 
type C assemblage. 

f) Hornbl ende with maximum ex tinc tion angle ZAC = 
go and opt ical axial angle 2V = -85°; pleochroism: 
X ;;;; brown, Y ;;;; yel lowish-green, Z = green, 

obscrved in type D assemblage. 

Plagioclase (oligoc lase) 

Plagioclase occurs as hipicliomorphic e longated 

grains ancl alot riomorphic cquidimensional grains. Usu­
ally . the hipidiomorphic grains display pol ysyn the tic 
twin lamc llae according to a lbite law B in ;;;; -1(010). 

Metamorphic processes caused inte nsive sericitization 
and saussuriti za lion. The Miche l-Levy method and U­

stage measurements using the Pedo rov method de te r­
mined the plagioclase as oligoclase with a compositi on 
or An 26-30 mol. %. Oligocl ase appears in all asse rl1 -
blages , 

Albite and mie roc linc 

Albite occurs as fre sh xenob las ts, obvious ly as a 
lale mine ral in the assemblage. It occurs in assemblage 
A. Microeline is observed in the type D assemblage and 

is a product of metasomatic reactions caused by the lat ­
er intrusion of the granite body. 

Epidote 

Epidote is common in type A assemblage and a 
minor com pone nt in both C and D assembl ages. It 
appears as irregular, weakly pleochroic grains, pale ye l­
low to yellow colors; 2V = -75° (pistaeite) and is ofte n 

associated with needle like actinolite. 

Clinozoisi te 

C linozois ite appears as irregular grains, which origi­
nat ed rrom primary basic plagioe lases; 2V = +85°. It 
occurs in the B, C and 0 asscmblages . 

Quartz 

Quartz is present in all assemblage types and often 
shows evidence o f cataclastic processes. 



~ ~ :::. ~ 
o ~ :3 V; 

g- :3' :3 ~ -j 
(") CJl (') ...., --

en 0 0.. & ;;i' 
V; -, 0- (") -, 
i:: "0 '< ,..... CJl 

co -::r' tol < -j tol 

~ ~, ~ g- "0 
'J} 0 ~ ::!, 
'J} (") (") => 
~~3'g2 
~p(1)5i''''': 

2: 2. ~ 0. :3 
(") _, (D 0 to) 

(")=>e::'""!':o~ 
c52~u:3 
::: 0 '< tol 
0- -, 0 ...., ;::::, 
(') ~, >-; ~ (") 

on:: (1) :3 
0- ". ;:j ::J -, 1./l5 u (1)=> 
~ 0.. ::r' t.) ~ < _. - . 
(1) 0 g 0.. 2-
0..5-0""0 ::;:: 
~'C;::: 

n 
:T 

C -<; 
C 
n 
n 
co 
Ul 
=> 

to) ::E 0.3. g. 

[~~ ~ 
..... ' ,..... 0 0 o -- tol 
::J ~ q;' ;;l 

'U "" ~ ~ 

o " g ::!, b:J 
~ - -CJl (') 0 
o 0" = 
~ 0 -_a." 
;- ,::, in 
n - 0 8 ...., ~ ...., .g re 
~ - ~ 

:3' E ri 

g 
15 

"V 
'< 
(3 
~ 

" ~ 

-< -<D."2-
"0 () "0 to) tol to 
(') 2: ('):3 g -, 
>- s; tJ ~ 0.. §-. 

- ::.: tol:::::, 0" ..-to 0 (/l 0",< (") 

0; 

=> a. 
n 
0; 
~ 
~ 

" 3 
v-
0; 

"" " !:" 

=> 

~ 2- 0 
;:j (') ::r' 
CTo. -
;; i:: g 

Oq (') :=: 
" - " ·0' 

::l ~ 

~ ~ 
~ => 
2 g 
~ ~ 

" -::, ::r' 

" " 

v-
s-
"" -" '" 3 

" 
" ~ 
" ~ 
~ 

" 
" ~ '" !". 3 
0 " => => ;:;' 
<3 
~ 

'U 
:T 

" => 
" 
=> 

" 
" ~ ~ 
" 3 , 

&:; :3 
a. -. 
0' a > 
'U "0 " CJl ...., (") 
_. 0 0 
g-o-::! 
. " -;:::< 0.. :3' 
_, tol (rQ 

(/l 5i -t 

O. 0 
g:;.o 
s:; in'"'£ < - . 
" "0 co 
0.. ~ 2-
_, 0 (1) 
=> ~ ~ .... (') ~ 

~ ~ :3 
(') ::: n ~ no tol 
D C ::­
~ _ 0 
I./l 0.. ::J 
r? Ci'" I./l 
;:j 0 (") 
0" (") 0 
S" ::r' :3 

0::; :: 0-o ~ ..... ' 
• (") ::l - " " a. ~ _. ~ 

N < " -. a.:;' 

Oxides A1 

Si02 45.71 (1.43) 
Ti02 1.09 (0.32) 
AI20 3 7.31 (097) 
Fe

2
0

3
' 3.27 (0.39) 

FeO 15.39 (0.76) 
MnO 0.42 (0.04) 
MgO 10.12 (0.78) 
CaO 11.85 (0.16) 
Na20 0.91 0.13) 
Kp 0.71 (0.18) 
Hp· 1.98 (0.02) 
Sum 98.75 (0.62) 

#Si 'v 6.91 
#AI 'V 1.09 
T site 8.00 

#Alv 1 0.22 
#Fe3+ 0.37 
#Ti 0.12 
#Mg 2.28 
#Fe2+ 1.95 
#Mn 0.05 

M 1,2,3 5.00 

#Ca 1.92 
#Na 0.08 
M4 site 2.00 

#Na 0.19 
#K 0.14 
A site 0.33 

#0 22 
#OH 2 
No anal. 9 

g 
~~§.[~~~~~.: 
~~2,~ ~ §~::i 
+ ;:. N ~ ~ :::::. :::l :::l :::. r.l 
AS' W So g ~~~ 3 g 

or. 00 , "" 

A2 

50.84 
0.04 
2.34 
1.38 

16.40 
0.25 
11.75 
12.29 
0.24 
0.11 
1.99 

97.62 

7.68 
0.32 
8.00 

0.09 
0.16 
0.00 
2.64 
2.07 
0.03 
5.00 

1.99 
0.01 
2.00 

0.06 
0.02 
0.08 

22 
2 

81 C1 C2 C3 

44.54 (0.33) 42.96 (0.74) 46.74 46.12 (0.24) 
0.95 (0.04) 2.73 (0.46) 1.86 0.95 (0.14 
10.32 (0.28) 11.53 (1 .02) 9.19 8.99 (0.19) 
4.41 (0.22) 6.51 (1.97) 2.49 3.30 (0.50) 
12.69 (0.16) 8.45 (1.74) 9.54 12.16 (0.40) 
0.29 (0.03) 0.21 (0.03) 0.27 0.22 (0.03) 
10.90 (0.14) 11.70 (0.44) 13.66 11.79 (0.16) 
12.02 (0.12) 10.93 (0.47) 12.27 12.020.10) 
1.44 (0.08) 1.61 (0.29) 1.00 1.05 (0.02) 
0.47 (0.06) 0.49 (0.14) 0.27 0.47 (0.02) 
2.02 (0.01) 2.03 (0.00) 2.06 2.03 (0.00) 

100.05 (0.59) 9915 (0.07) 99.35 99.11 (0.17) 

6.59 6.33 6.80 6.83 
1.41 1.67 1.20 1.1 7 
8.00 8.00 8.00 8.00 

0.39 0.34 0.37 0.39 
0.49 0.72 0.27 0.37 
0.11 0.30 0.20 0.11 
2040 2.57 2.96 2.60 
1.57 1.04 1.1 6 1.50 
0.04 0.03 0.03 0.03 
5.00 5.00 5.00 5.00 

1.91 1.72 1.91 1.90 
0.09 0.28 0.09 0.10 
2.00 2.00 2.00 2.00 

0.32 0.18 0.19 0.21 
0.09 0.09 0.05 0.09 
0041 0.27 0.24 0.30 

22 22 22 22 
2 2 2 2 
5 6 5 

C4 D1 

49.33 (0.57) 47.52 (0.80) 
0.66 (0.05) 0.98 (0.22) 
6.44 (0.19) 7.77 (0.54) 
4.34 (0.58) 3.55 (0.74) 
10.88 (0.55) 13.17 (0.85) 
0.23 (0.04) 0.40 (0.09) 
13.28 (0.29) 11.59 (0.60) 
12.02 (0.19) 11 .88 (0.20) 
0.81 (0.09) 1.10 (0.09) 
0.24 (0.03) 0.44 (0.08) 
2.07( 0.01) 2.04 (0.02) 

100.30 (0 .71 ) 100.44 (0.62) 

7.14 6.97 
0.86 1.03 
8.00 8.00 

0.24 0.30 
0.47 0.39 
0.07 0.10 
2.87 2.53 
1.32 1.62 
0.03 0.05 
5.00 5.00 

1.86 1.86 
0.14 0.14 
2.00 2.00 

0.09 0.18 
0.05 0.08 
0.14 0.26 

22 22 
2 2 
4 8 

D2 

45.63 (0.71) 
0.55 (0.08) 
14.43 (0.77) 
3.70 (0.86) 
9.51 (1.18) 
0.22 (0 .08) 
11.12 (0.58) 
10.51 (0.44) 
2.20 (0 .24) 
0.51 (0.07) 
2.09 (0.01) 

100.47 (0.39) 

6.55 
1.45 
8.00 

0.99 
0040 
0.06 
2.38 
1.14 
0.03 
5.00 

1.62 
0.38 
2.00 

0.23 
0.09 
0.32 

22 
2 
10 

o 

o 
8 
0-
~~. 

(") 

g 
[j" 
.~ 

s 



13;l lcn.l3cnnall<.:c &. Marci: Compositional Zoning in Amphibole from Amphibole I3carin!! 1';"'ilgCllcscs ... II 

1 • 

tremolite magnesio-hornblende tschermak.ite 0.9 

• A type 
0_8 £' 

• " LL 
• B type + 
.t. C type en 

• :2 
• D type • '. '§; • . . 

• • t . • • - 0.7 • • :2 • • • • 
• .t. ....... • • .. ."\. , 

• • .. II. II. .... 
" • •• 0.6 . • . • 

" " ". 
" "." 

" 0.5 
8 7.5 7 6.' 6 

Fig. 2 Class ification o f calcic 
amphibole (modifi ed arter 
LEAKE, 1978: HAWTHOR­
NE, 1983: ROCK & LEAKE, 
1984 according to Internal ional 
Mineral ogical Association 
Comm iss ion on New Minera ls 
and M i neral Names ( IM A 

CN fvIMN) Nom enclature of 
A mphiboles (Drart 5, t995)) . 
Diagram parameters: Cae ;::: 
1.50; (Na + K)A < 0.50; Ca,\ < 
0.50. 

Si (apIu) 

the cleavage cracks of amphibole as alte rati ons and as 
indi vidual grains in the matrix. 

Additional minerals 

Garnet occurs as grains in type D assemblage. Cal ­
c ite is a secondary mineral. Apatite and zircon are very 
rare, and are obse rved as solid inclu sions in both 
amphibole and plagioclase. 

5.2. MINERAL CHEMISTRY 

Amphibole 

A represent31ive chemical analys is of selected 
amphibo le sampl es and their chem ical formulae are 
presented in Tabl e I. The Fe2

+ and Fe3
+ content s were 

es tim ated fro m the total o f cations (exc luding Na, K 
anel Ca) normalized to 13. 

Assumption of normali zing cations to 13 + Ca + Na 
+ K excl udes Mn , Fe2

+ and Mg from the M .. site in the 
formulae (ROB INSON el aI., 1982) and produces fairly 
reasonable es timates or FeO and Fe20 3• This normaliza­
tion procedure is also favored by the International Min­
era logical Assoc iation (lMA) amphibole nomenclature 
scheme (L EA KE, 1978; HAWTHORNE, 1983 ; ROCK 
& LEAKE, 1984). In this work, the orig inal c lassifica­
tion scheme aft er Leake was modificd according to the 
Commi ss ion on New Minerals and Minera l Names 
(CNMMN IMA) Nom enclalure of Amphiboles (Drafl 
5 , 1995). Modifi cation s are basically chan ges in the 
values of diagram parameters which result in the disap­
pea rance of the fie lds o f tcherm ak itic and ae tinol ilic 
hornblende. 

All the analyzed amphi bole grains have (Ca+Na)n> 
1.0 and Nau < 0.5, so they arc calc ic amphiboles. A 
graphical presentation o r thc analyses is given in Fig. 2. 

Further subdivi sion of (he amphibol cs 011 the basis 
of the plot Mgj(Mg+Fe2+) against Si shows that most of 
the anal yzed samples occur in the fi e ld o f magnesio­
hornbl ende (Iype A, B and D). Amph ibo les of C Iype 

show a hi gh degree o f variabil ity and projcct into the 
fi e lds o f tschermakite and magnesio-hornble nde. The 
actinolite of type A project into thc " tremolite" fi e ld. 

The te trahedra l va lues o f S i are restri cted be tween 
6. 33 and 7.14 apfu with the exception of ac tin o lite 
(7.68 ap fu S i). 

The octahedral aluminum ranges between 0.22 and 
0. 39 apfu wilh Ihe exeeplioll of aClinolile (0.99 ). 
According 10 ROBINSON el a l. (1982), Ihe 10lal Al v, 
shou ld not exceed 1.4 and averages around 0.4 ap fu. 
Titanium in the octahedral sites varies from 0.06 to 0.30 
apfu (it is not detccted in actino lite). 

The contents of Mg and Fe2
+ in the M t.2.~ sit es are 

compl ementary because these arc two essent ial cations 
in the amphiboles. Variati on of these two cations in the 
amphiboles reflect s the chemi stry of their host rocks. 
Mn is a minor clement in the M 1.2.3 s ite. 

Calc ium is concentra ted in the M4 s ite and varies 
from 1.62 101.99 aprl!. Sodium parlially fill s Ihe M, 
sil e to 2.00 apfu and the remaining Na with K is 
ass ig ned to the A site. T he total Na in both s ites in 
amphibo les from type D assemblage can reach 0.62 
apfu whi ch is an elevat ed concentrat ion in compari son 
with other amphiboles . Potass ium is res tric ted to the A 
sile from 0.02 10 0. 14 , averaged 0.08 apfu. 

5.3. THERMOBAROMETRIC EVALUAT10N Or: 
MET AMORPHISM 

Recent studics indicate rul es which govern changes 
in the composition or amphibol es and plagiocl ases as a 
response to progressivc alteration of the metamorphic 
cOllciiliollS ( HOLLAN D & R1C HARDSO N, 1979; 
SPEAR, 1980, 1981a, b; HY NES, 1982; MA RUYA ­
MA el aI., 1982, 1983; BLU NDY & HOLLA ND, 
1990). 

Amphibo les of assemblage D have th e larges t 
amount s of Na, K and AI. In amphiboles of assemblage 
C conccntrations of Ti are hi gher but this is probabl y 
due to the hi gh contcnt of Ti in the prolho lite (gabbro). 



12 

Trc 
SPEAR BLUNDY & HOLLAND 
(1980) (1990) 

Sample 26% An 30% An 

A1 500 666 675 
A2 350 478 484 
B1 500 730 740 
C1 530 783 795 
C2 500 688 697 
C3 490 682 691 
C4 510 618 626 
D1 510 654 662 
D2 530 738 749 

Table 2 Temperature of metamorphosis on the basis of the SPEAR 
(1980) and BLUNDY & HOLLAND (1990) geothermometcrs. 

A high content of Ti is also found in type D amphi­
boles, The lowest Si contents are found in amphiboles 
of lype C and D assemblages. In amphiboles from A 
assemblage the distribution ofTi, Na, K, AI and S i have 
an opposi te tfend to those amphiboles of assemblage D. 
Amphibo les from assemblage A have the lowest con­
tent or Ti, Na, K and Al and largest amounts of Si. 

Therefore it can be stated that the highest tempera­
tures occurred during growth of amphiboles in assem­
blage D, were lower in assemblages Band C and lowest 
in the assemblage A. 

Equilibrium on the scale of the hand-specimen is 
assumed between obse rved amphibole-plagioclase 
pairs. Because plagioclase grains were not measured by 
microprobe the whole range of plagioclase composi­
tions obtained by optical measurements were used for 
geothermobarometry calculations. 

Tcmperatures of metamorphism were outl ined with 
graphical solulion on the basis of SPEAR's (1980) 
geothcrmometer, and calculated from the equation from 
the B LUNDY & HOLLAND (1990) geolhennometer, 
with an assumed pressure of 5 kbars (5.108 Pa) and pla­
gioclase compos ition range betwcen 26% An and 30% 
An (determined by U-stage measurcments). The results 
arc shown in Table 2. 

There arc differences between the results obtained 
using these two approaches. The "semiquantitative" 
numerica l solut ion (because the chemistry of plagio­
clase had not been determined by electron microprobe) 
obtained from the BLUNDY & HOLLAND (1990) 
geothermometer shows a simi lar trend in temperature 
d istribut ion for all amphibole bearing parageneses, but 
the values seem to be too high for the observed parage­
neses. This geothennomcter is velY sensitive fo r cation 
content and due to thc "semiquantitative approach", a 
possib le discquilibrium (amp hibole rim - plagioclase 
core), and metasomatic effects from intrusion of granite 
body, exact va lues of the thermometric evaluation 
should not be accepted. 

Geologia Croatica 50/1 

Temperatures obtained on the basis of the diagram 
proposed by PL YUSNINA ( 1982) arc in the range of 
520-550 C for the non-actinoli te assemblages (B, C 
and D). Geothennometry evaluat ions are in agreement 
wi th the SPEAR (1980) empirical geothermometer. 

Estjmation of pressures was performed with the aid 
of a diagram proposed by PLYUSNINA (1982) . T he 
calibration of this diagram produced a wide range of 
pressures (depend ing on the AI content in the amph i­
bole) where amphiboles occur in the range from 2 kbars 
to 5 kbars (2 to 5.108 Pal wilh the exception of one 
specimen in assemblage D. This specimen has a P value 
of? kbars (7.108 Pal. 

According to temperatures stated in the introduction 
(MARUY AMA et aI., 1982) amphiboles from assem­
blage A belong to the greenschist facies, while amphi­
boles from assemblages B, C and D are of the amphibo­
lite facies. 

Zonal growth of amphibole grains is a result of mul­
tiple periods of mineral growth. The cores appear to be 
re li cs, preserved because of incomp lctc cquilibrium. 
Assuming that the cores grew before the rims, amphi­
boles can be used as an relative time indicator. More­
over, the distinct compositions of core and rim, indicate 
different metamorphic conditions. Each growth event 
can be characterized by a different metamorphic grade. 
Partial reequi libration wi ll prevent the de termination of 
the core composition at the peak of metamorphism. It is 
also very probab le that the mineral assemblage was 
slightly dirl"crenl when the core crystall ized. or course, 
only larger grains have diffe rent cores from rims, 
because the core in (he sma ller grains was completely 
consumed. Elements necessary to make new rims come 
from alteration of other phases (change in modal and 
chemical compositions). 

Informat ion about the metamorphic conditions dur­
ing crystal growth of amphiboles from type C assem­
blage are recorded in Fig. 3. As minerals are buffered 
by the whole assemblage, the composition of individual 
minerals can be mutually compared, and variations in 
mineral chemistry can be related to differences in meta­
morphic grade or facies series rather than bulk rock 
composition. 

Profiles show the differences in the d istribution of 
cations. These differences are due to changes in meta­
morphic conditions during the mineral growth. The 
cores of amphiboles have higher amounts of Si and Mg 
and lower amounts of AI, Ti, Na, K and Fe than the rim 
of same grain . Increas ing of Fe, Na, Fc/Mg ratio and 
decreasing of Mg and Si from cores to rims of the 
grains also indicates ris ing pressure. Accord ing to 
SPEAR (l981b) this sugges ts that the temperaturc 
increased dur ing mineral growth. 

Therefore we can suggest conditions of relative pro­
grading metamorphism. 

Co-existing amphiboles have (NaA +K) ,; 0.25, NaM, 

,; 0.2, 0.6'; AI 'v ,; 1.2 and 0.6'; (Alvl+Fc3++ T i) ,; 1.2 in 
low-pressure samples. Alternatively co-exist ing amphi­
bole have 0.2'; (NaA+K)'; 0.5, 0.2'; Na'H'; 0.5, 1.2'; 



Balcn, Bcrmancc & Marci: Cumposilioll;11 Zoning in Amphibolc from Amphiboic L3carillg Paragcncscs ... IJ 

rim core rim • .00 

--0- # Si IV 

Si02 46.74 49.70 50.19 48.99 46.61 7.50 

Ti02 0.86 0.54 0.49 0.70 0.95 7,1)(1 -

Alz0 3 8.68 5.80 5.41 6.28 8.51 , 
FeZ0 3 2.33 2.67 2.41 2.99 4.00 ~ 6,50 

FeO 2.92 11.63 11.06 11.50 11.84 
~ 

'.00 

MnO 0.20 0.27 0.29 0.19 0.22 

MgO 11 .82 13.54 13.89 13.22 11.98 
5,50 

CaO 12.08 12.21 12.11 12.12 11.94 '.00 

NazO 1.08 0.81 0.71 0.71 1.06 
rim core rim 

Kp 0.47 0.21 0.17 0.28 0.42 

HP 2.03 2.05 2.05 2.04 2.04 3,50 

-0- # AIIV 
Total 99.21 99.43 98.78 99.02 99.57 

'00 
--0-- # Fe~' 
-6- # Mg 

#Si IV 6.91 7.25 7.34 7.18 6.86 , 
#AI IV 1.09 0.75 0.66 0.82 1.14 

~ <.50 

rn 

T pos. 8.00 8.00 8.00 8.00 
~ 

' .00 8.00 , 
~ 

~ #AI VI 0.42 0.25 0.27 0.27 0.34 ~- 1.50 _ 

#Fe3+ 0.26 0.29 0.26 0.33 0.44 " '.00 
#Ti 0.10 0.06 0.05 0.08 0.11 

#Mg 2.60 2.95 3.03 2.89 2.63 0.00 
rim core rim 

#Fe2+ 1.60 1.42 1.35 1.41 1.46 

#Mn 0.03 0.03 0.04 0.02 0.03 

M I ,2 ,3 5.00 5.00 5.00 5.00 5.00 
o.~ 

-<>- # Ti 
--0-- # Na A 

#Ca 1.91 1.91 1.90 1.90 1.88 0 .25 -& #K 

#Na 0.09 0.09 0.10 0.10 0.12 
~ 0.20 

M4 Pos 2.00 2.00 2.00 2.00 2.00 
, 
1i • ;; 0.15 

#Na 0.22 0.14 0.10 0.11 0.19 .. 
#K 0.09 0.04 0.03 0.05 0.08 ~ 0.10 

A pos. 0.31 0.18 0.13 0.16 0.26 
~ 

0.05 

#0 22.00 22.00 22.00 22.00 22.00 
0.00 

#OH 2.00 2 .00 2.00 2.00 2.00 rim core rim 

Fig. 3 Comparison of chcmical compositions in core and rims from lypc C amphibolc. 

AIIV ~ 1.8 and 1.2 ~ (Alvl+FeJ++Ti) :::; 1.8 in mcdi um ­
pressure samples (LAIRD & ALBEE, 198J). In the 
west Psunj region there are low-pressure amphiboles 
but samples of types C and 0 were probably stabilized 
at medium-pressure. Accurate estimation of pressure is 
difficult and the broad conclusion of pressure estimates 
is that pressures in parageneses from B, C and 0 
assemblages were higher (i n respective order) than in 
assemblage A. 

6, DISCUSSION AND CONCLUSIONS 

Investigated amph iboles can bc regarded as a solid 
solution consisting of 

tremolitc 
pargasite 
edcnitc 
tchermakite 

(Ca,Mg,Sipn (OH),), 
(NaCa2Mg,AISi"AI,on (OH)2)' 
(NaCa,Mg,Si7AIO,,(OH)2)' and 
(Ca2Mg,AI,Si"A l,on ( 0 H),) 

components. The trenlOlite-actinolitc molecule is stable 
in thc grecnsch ist field while the tschermakite molecule 

is stable in the amphibolite facics, and the edcnitc mol­
ecule in the epidote-amphibolite facies (HOLLAND & 
RICHARDSON, 1979). Recalculations of the amphi­
bole compositions to the mole fract ions (CURRIE, 
1991) shows that the tremolite-actinolite and edenite 
mole fractions prevail in amphiboles of assemblage A, 
while tschermakite and edenite are dominant in the oth­
er assemblages (Table 3). 

Amphiboles from assemblage A belong to the 
greenschist facies or to the albite-epidotc zone. With a 
rise of P-T condit ions, the tschermakite mole-fractions 
also increase, so the amphiboles of assemb lages B, C 
and D belong to the transition zone and lower amphibo­
lite facies. 

In this rangc of P-T conditions a change in the com­
posi tion of amphiboles leads to thc transition from acti­
nolite into hornblende through coupled substitutions: 

o (A) + Si(lV) = Na(A) + AI(IV) eden ite substitution 

Mg(VI) + Si(IV) = AI(VI) + AI(IV) tsehermakite sub­
stitution 



14 

A1 A2 B1 C1 C2 

hornblende 51.3 23 .3 9.1 46.5 
barroisite 7.5 1.2 9.0 15 .6 8.1 
tchermakite 8.2 40.6 55.2 20.0 
edenite 32.8 7.9 41.1 18.1 25.0 

gedrite 0.3 0.2 
tara mite 10.8 

tremolite 37.6 

actinolite 29.5 

wh ich togethe r gave 

[J (A) + M g(Y I) + 2Si(lY) = Na(A) + A I(YI) + 
2AI(IV) pargas ilc substitution 

and numerous other reactions with co-existing minerals. 
During reaction s the re arc changes in the c hemical 
compositions of mine ra ls (continuous reac tions) and in 
the modal composition of minera ls, resulting in forma­
tion o f ne w m ine rals and the de trime nt or expi rat ion of 
pre-exis ting mine rals (discontinuolls reactions). So, par­
a lle l wi th the transition o f act inolite into hornblende, 
albit e d isappears, the composition of chloritc changes 
( inc reasing Mg component , de trimental to the chlorite 
modal composit ion , and finall y disappeara nce of c hlo­
rite from the assemblage) . Paralle l 10 thi s process is an 

increase in the AI component in epidole, detrimental of 
the modal compos ition and the final disappearance of 
e pidote. With the c hanges of modal composition of 
c hl ori te and cpidolC, the AI content in thc p lagioclase 
and the Fe content in amphiboles increases . Also with 
mo re basic plag ioc lasc the AI and Na content in the 

amphibolcs increases. These c hanges arc di scontinuous. 
T he s tud ied m atc rial has cons is tc nt pe trographi c 

fea tures with cert ain s tructural, textural, compositional 
and mine ralogical vari ation s. Di rfere nees be l ween the 
amphibolc bearing pm'ageneses of wes t PSlInj are main­
ly the modal compos ition of mi ne rals, and var iat ions in 
the chemical compositions which have only been inves­
ti ga ted in de tail in the amphiboles . Al l paragcneses 
have in common amphibole and plag ioclase with minor 
amounts of qua rt z and ilmenite or sphene. 

Chemica l inves ti gations of the amphiboles indicate 
va ri ations in the features which e na bl e deduc tion of 
genet ic cond it ions and processes. Te mpe rature of for­
mation is se t to be tween 490 an d 550'C (350°C for 
actillo litcs) and pressure between 2 and 5 kbars (2 and 
5 ·IO' Pa). One specim e n (lype D) has a P va lue of 7 
kbars (7' 10' Pal. 

G eoc hemi cal in vestigat ion s (PAMIC & MARCI, 
1990) suggest that there is no rad ical change in the pro­
tholite compos ition (prolholile is thol e iitic basalt) and 
that these me tamorphic processes can occur in the ophi­
olite zone (pAM IC & LANPHERE, 1991 ). Differences 
in assemblages may be a ttribut ed 10 c han ges in P-T 
condit ions. 

Gt,-ologi;t Cromiea 50/ 1 

C3 C4 D1 D2 

43.6 58.4 56.9 
9.0 13.3 13.1 30.4 
17.3 3.5 37. 1 
30.0 13.8 26.3 24.7 

7.7 
Tab le 3 Mole frac ti ons of 

9.8 amphibo le end-members 
4.5 calculated with sort ware 

aftcr CURRI E ( 1991 ). 

Changes in P-T conditions are recorded in the 
growth or the amphibole grain s (gene ral prograding 
me tamorphi sm) and c hanges in modal compositions in 
re lated parageneses indi cate that the st ud y area has 
undergone geotectonic change during ils e volution. A 
mode l which may expla in the c hange of conditions 
from greenschis t to amph ibolit e faci es as recorded in 
the an1phibolc of wes t Psunj , may be that o f a subduc­
tion zone. Howeve r to tho roughly validate thi s mode l 
more data are required to confirm espec ia lly the re la­
tionship between the P-T inc re ments and the lime scale. 

Fu rtherm ore, differences between assemblage A and 
0 , C and D could be explain ed by varyi ng c he mical 
composition or or ig inal protholi te (for instance lulTitie 
sediments). The protholite compos ition of assemb lage 
A has not ye t been rel iably de te rmined. 

Acknowledgments 

Use or the electron mic roprobe of the Uni ve rs ity of 
Be rn by Sehweitze ri sehe r Nat ionalfond s (c redit 2 1-
26579.89) is acknowledged togethcr w ith fi nanc ial sup­
por t from the Croati an Minis try of Scie nce a nd Tech­
nology (credi ts 1-09-016 and 1-09-238). 

The a uthors are a lso g raleful 10 ProL PCle r ARKAI 
(H un gari a n Academy or Scie nces) and to an anony ­
mous re vi e wer for the ir improve ment s to th e manu­
script. 

7. REFERENCES 

AFIF I, A. & ESSENE, EJ. ( 1988): M INF IL E: A 
mic rocomputer program Cor storage a nd manipula­
tion of chemical data on mincrals.- Am. Mine ral. , 
73, 446-448. 

APTED , M.J. & L10U, J.G. ( 1983) : Phase re la tions 
among greenschist, epidote-amph ibolite and amphi ­
bolite in a basaltic sys le m. - Am. Jour. Sc i., A, 283, 
328-354. 

BLUNDY, J.D. & HOLLAND , T.J.B. ( 1990): C al c ic 
amphibole equilibria and a new amphibo le-plag io­
c lase geothermome tc r. - COJltrib. M inerai. Pe trol. , 
104 , 208-224. 



l3a1<..:l1. I3cnnallcc & M;)rci: Compos ition;)1 Zoning ill Amphibole from Amphioole Beari ng Paragel1eses ... 15 

CURRIE, K.L. (1991): A simp le quantitative calcula­
tion of mol fractions of amphibole end-members.­
Can. Mincral., 29, 287-299. 

HA WTI-IORNE, F.e. (1983): The crystal chemistry or 
the amphiboles.- Can. Mineral., 21, 173A80. 

I-lYNES, A. (1982): A comparison of amphiboles from 
meclium - and low - pressure metabas ites.- Contrib. 
Mineral. Petrol., 81, 119-125. 

HOLLAND, TJ.B. & RICHARDSON, S.W. (1979): 
Amphibole zonation in metabasites as a guide to the 
evolution of metamorphic conditions.- Contrib. 
Mineral. Petrol., 70,143-148. 

JAMICIC, D., VRAGOV IC, M., & MATTCEC, D. 
(1989): Osnovna geoloska karta 1: 100000, T umac 
za list Daruvar L33-95 (Geology or sheet Daruvar). ­
Geoloski zavod Zagreb (1988), Say . geol. zavod 
Beograd, SSp. 

KISPATIC, M. (1892): Prilog geolozkol11 poznavanju 
Psunja.- Rad JAZU, 109, 124-181, Zagreb. 

LAIRD, J. & ALBEE, A.L. (1981): Pressure, tempera­
ture , and time indicators in mafic schist: thei r appli­
cation to reconstructing the po!ymetamorph ic histo­
ry ofVermonl.- Am . J. Sci., 281,127-175. 

LEAKE, B.E. (1978): Nomenclature or amphiboles.­
Mineral. Mag. , 42, 533-563. 

MARCI, V. (1965): Petrography of the western part of 
Psunj mountain (in Croatian).- Acta Geologica, IV, 
315-322, Zagrcb. 

MARUYAMA, S., L10U, J.G. & SUZUKI, K. (1982): 
The peristerite gap in low-g rade metamorphic 
rocks. - Contrib. Mincral. Petrol., 81, 268-276. 

MARUYAMA, S., SUZUKI, K. & L10U, J.G. (1983): 
Greenschist-amphiboli te transition equil ibria at low 
pressures.- J. Petrol., 24, 583-604. 

PAMIC, J. & LANPHERE, M. (1991): Hercynian gran­
ites and metamorphic rocks from the Mts. Papuk, 
Psunj, Krndija and the surrounding basement of the 

Panonian basin in Siavonija (northern Croatia) (in 
Croatian).- Geologija, 34, 81-253, Ljubljana. 

PAMIC, J. & MARCI, V. (1990): Petrology of amphi ­
bolites from the Slavonian Mountains in Northern 
Croatia (in Croatian). - Geoloski vjesnik, 43, 121 -
133. 

PL YUSNINA, L.P. (1982): Geothermometry and geo­
barometry of plagioclase-hornblende bearing assem­
blages.- Contrib. Mineral. Petrol., 80, 140-146. 

POUCHOU, J.L. & PI CHOIR, F. (1984) : Un nouveau 
modele de caleul pour la microanalyse quantitative 
par spectrometric de rayons X.- La Recherche 
Aerospatiale, 3,167-192. 

ROBINSON, P., SPEAR, r.s., SCHUMACHER, le., 
LAIRD, J. , KLEIN , e., EWANS, B.W. & DO­
OLAN, B .L. (1982): Phase relations of metamor­
phic amphiboles: natural occurrences and theory.­
In: VEBLEN, D.R. & RIBE, P.H. (eels.): Amphi­
boles: petrology and experimental phase relations. 
Reviews in Mineralogy, 9B, MineraL Soc. Am., 1-
228. 

ROCK, N.M.S. & LEAKE, B.E. (1984): The Interna­
tional Mineralogical Association amphibole nomen­
clature scheme: computerization and its conse­
quences. - Mineral. Mag., 48, 211-227. 

SPEAR, F.S . (1980): NaSi ~ CaAI exchange equilibri­
um between plagioclase and amphibole. An empiri­
cal moclel.- Contrib. Mineral. Petrol., 72, 33-4l. 

SPEAR, F.S. (1981 a): Amphibole-plagioclase equi li b­
ria: an empirical model 1'01' the relation albite + 
tremolite:::: cdcnitc + 4 quartz.- Contrib. Mineral. 
Petrol., 77, 355-364. 

SPEAR, F.S. (198Ib): An experimental study of horn­
blende stabi lity and compositional variability in 
amphibolite.- Am. Jour. Sci., 281, 697 -734. 

TAJDER, M. (1969): The petrogenesis of blastopor­
phyritic amphibolite in the region of Omanovac in 
Psunj (in Croat ian) .- Acta Geologica, VI, 5-16, 
Zagreb. 

Manuscript received July 15, 1996. 
Revised manuscript accepted April 28, 1997. 



16 Geologia Croaticl 50/1