RENDICONTI DELLA SQCIET.A ITAUANA DI MINERALOGIA E PETROLOGIA, 1988, \bI. 4), pp. 10)·119

Fluids in the Monte Pulehiana intrusion (N Sardinia):results on fluid inclusion study

MARIA LUCE FREZZO'ITI

Dipartimemo di Scienu della Terra, Via deUe: Cerchia 3, ~Bl00 Siena, Italy

CLAUDlO GHEZZODiputimcnlo di Scie:nze ddla Terra, Via ddk Cerchia 3, '3100 Siena, hdy

)ACQUES TOURETInstituul voor AudwnetUChappcn. Vri;e Univenitdt. De Bodelun, 1007 MC Am.sterdl.llt. 1nc: Ne:thc:rland5

A8S1'RAcr. - The: Monte: Pukhianllleuoogranite is one:of the youngest posl-tectonic shallow level Hercynianintrusions in Sardinia, The granite is a medium to coarse:grained rock, consisting of quartZ, microperthitk K·feldspar, sodk plagioclasc: and minor biotlte. Within the:plulOn, especially in the- bonier zones, SOOle aplitc dykC$and pegmatitic pods are present, Sometimes lhe: IlIIermay contain c:ubedral quarlz crystals in open vugs andcavities. Microscopic and microthermomelricinvestigllllons on fluid inclusions in quarlz fromreptesenlalh'e samples of granile and in onc euhedralquartlt from a pc:gmatitic VU8 gi~ evidence of !WO fluid·rock ime:tKIion episodes. T ....,o main types of fluids.~rCCO$nizc:d; the first onc is presem only in the grtnitc,conSISting of a highly saline aqueous fluid, Extremel)'variable NaCl coniem SUF$CStS ovetsaturarion 11 the time:of the final inclusion evolution (average salinity estimatearound 30 wt.% NaCl; Th .. 200°·3 10°C, in absenceof boiling); and the second, a much less saline aqueousfluid (:5 wt. % NaCI; Th. 1000·260°C), probably ofmeteoric origin, is present in the: granite as well lIS inthe pegmatitc.

These: observations are consistenl wilh the: cxistence:,in subsolidus conditions, of a firSI NaCI ovcrsatu~ted

brine, probably of initial magmatic origin, Absencc ofboiling imposes a pressure gn:atcr than 1.2 kb for thegranitc's depth of emplacement. AI a "Ier stage,homogeneous b.v salinity fluid {probablymct~ ""'I1tcdinvad~ the plulon. A pasl magmalic rooghly isobarict~iectory is proposed naning at the inferred condidonsof the granite cnd of CT)'Slallization (T - 700°C; P ­1.5 kbJ,

K~ lOOms; Fluid inclusions, 1~\IC08raniu:l, Sudinia,brines, P·T path.

Introduction

Fluid inclusions in shallow intrusives havebeen widely studied mainly in order tocharacterize the physico·chemical features ofassociated fluid phases often responsjble ofore deposition, see e.g. WEISBROD, (1981) andROEDDER (1984). However, these techniquesmay also give more information on thepostmagmaric evolution path and the episodesof fluid-rock interaction which have occurredin subsolidus conditions.

The Sardinia-Corsica Hercynian Batholithconsists of several plutonites ranging fromuhramafic rocks and gabbros to leucogranites;these lattet ones CtOp out all over the twoislands, constituting about 25% of allplutonites in Sardinia, Some characteristicfeatures, including the Mo-mineralizationsthat are commonly associated with thesegranitic rocks, are reponed in ORStNI (1980);BRAllA et al., (1981); GHEZZO and ORSlNl(1982); GUASPARRJ et al., (1984a, 1984b) andRossl (1986). In this paper we will discussselected results on fluid inclusion study in theM. Pulchiana leucogranite intrusion, one ofthe largest in northern Sardinia (Fig. 0,

Observations have been made in quartzcrystals of the primary mineral assemblage in

106 M_L FRE.ZZOTTI. C. GHEZZO, J TOUllET

sdected leucogranite samples and in euhedralsmoky quartz crystal from a vug in anassociated pegmatite.

Special attention has been given to thedistribution and composition of fluidinclusions, in order to reconstruct thesequence of fluid-phase/rock interactionswhich affected the plU[onite and areresponsible for its widespread deutericalteration.

Present results allow us to set an upper limitfor the emplacement of the intrusion at about4 km depth: at least two different fluidgenerations have been characterized. Anidealized evolution path at low P and T (wellin subsolidw condition) ispro~ in Fi~9.

Geological setting

The M. Pulchiana leucogranite intrusion isone of the youngest plmons of the intrusivesequence forming the Sardinia - CorsicaHercynian Batholith. This sequena:, mainlylate to post-tectonic, is dominated by twodistinct associations:

- a calcalkaline association dominated byI-type biotite monzogranites andleucogranites, with minor ultramafic rocks,gabbros, tonalites and granodiorites.

- A subalkaline potassic associationcomposed of sienomonzonites, monzogranites,a1askites, sienites, monzonites and dioritescropping out in northern Corsica.

In addition to these two main suites a minorpart of the batholith is constituted by two­mica and/or cordierite bearing granites (S·type), syntectonic strongly foliated gneissicgranites and socHc sienites.

The emplacement of the whole batholithtook place in a time interval between 330 and280 m.y., based on the Rb/Sr and KJArisotopic data (DEL MORO et al., 1975;COCHERIE, 1984).

In recem years, detailed mineralogical andpetrological studies have led to a beuerunderstanding of the sequence ofemplacement and crystallization history ofthese rocks (e.g. 01 S1I\1PUCIO et aI., 1974;ORSINI, 1976, 1980; BRoTZU et al., 1978;BRAllA et al., 1981; GHEZZO and ORSINl,

1982; BROTZU et aI., 1983; GIRAUD, 1983;COCHERIE, 1984; ROSSI, 1986).

The age of the M. Pulchiana pluton has notyet been defined but it must be close to 290m.y., since the emplacement age of theyounger leucogranite suite is referred to as290-28:; m.y. (with a 17Sr/",Srj = 0.7085)(DEL MORO et al., 1975). Its emplacementtook place at shallow levds after the maintectono-metamorphic events and the uplift ofthe whole basement. This emplacement waslikely to occur in extensionaI environment(GHEZzO and ORSINI, 1982). The plutonintrudes older late-tectonic plutonites. Thecontacts are sharp, truncating the magmaticfoliation. The host rocks (Fig. 1) includemedium-grained biotite grnnocliorites rich indark microgranular magmatic xeooliths (at thenorthern contact), biotite granodiorites +biotite-amphibole tonalites (along thesouthern border of the pluton) and pinkishcoarse-grained biotite monzogranitescharacterized by K·feldspar megacrysts. Theyall show discordant contacts toward themetamorphic metapelite country rocks whichhave a migmatitic character in thesillimanite + K-fddspar grade (FRANCESCHEU.Jet al., 1982).

A NE-SW striking fault separates thepluton in two segments and displays a lateralmovemem.

The biotite leucogranite is compositionallyand texturally homogeneous; a coarse-grainedequigranular pinkish granite consisting ofquartz (- 32%), oligoclase plagioclase witha1bitic rim (- 30%), orthoclase + microclinemicroperthites (- 35%) and minor iron-richbiotite (3-4%), with accessory magnetite,apatite, zircon, aIJanite. A Strong pervasivedeuteric alteration ov~rint is always present;and chloritization, albitization, sericilization,argillification are common alterationassemblages, often associated with minorfluorite. A network of microfracrures is alsopresent in the quartz crystals. Most of theseare intercrystalline, but also intracrystaIJinecracks have been observed. Fluid inclusiontrails underline the healed microfractures (DEVIVO et al., 1985). 'This overall pattern issimilar to those described by many authorsin granitic rocks (TIJITLE, 1949; PEcHER. et

FLUIDS I THE MO TE PULCHlANA I TRUSIO I SARDINIA): RESUUS 0 FLUID ETC. 107

+ + + + ++ + + + + +

• + + ++ + +

A

++ +

++++++++++++++++++++++++++++

1::"-.:"'::I Motamorphic CCfr1)Iex

Tona lites andgranodi 0 rites

_ ---- Faults

Biotite monzogranltes

I + + I Biotite teucogronites

Fig. 1. - Sketch map of the M. Pulchiana pJuton llJ'ld location of tbe investigated samples. ClTctes: leucogranitesamples: sqUIJTf!: euhedraJ smoky quaru crystal from a vug in a pegmaOte.

aI., 1985; LESPINA E and PECHER, 1986;KOWALUS et al., 1987).

In the border zones of the plmon bothaplitic dykes and pegmatitic pods are present.They constitute only a minor part of thepluton (no more than 5%).

Apliles occur as irregular, discontinuoudykes or lenses (10 to 100 cm thick) with fine­grained sugary texture. They often sho alayered structure due mainly to differentmineral grain size and biotite content, andcontain small pods with pegmatitic textur .The mineralogy is analogous to that of the

granite itself.The pegmatite pods have an irregular shape,

a 20·150 cm diameter and are often intimatelyassociated with the aplite dykes. The consistof milky quartz, reddish orthoclase-perthite,albite, minor biotite and other phases suchas muscovite and garnet. These pegmatitepods are massi e, hut some of them have vugslined by eubedral smoky quartz crystals(miarolitic pegmatites). ometimes amoderately zoned pattern occurs; movingfrom a border with graphic texture, the grainsize increases (up to 5-15 cm) to a core with

'08 M.L FREzzarn. C. GHEZW. J. TOURET

~gmatitic texture.The following aplite + pegmarite

petrogenesis will be proposed, mainly byreferring to the}AHNS and BURNHAM'S (1969)model:

- the pegmatite pods are the result of aclosed-system crystallization when the granitereached the water-saturation conditions andlocally a hydrous silicate melt crystallized untilreaching its solidus, which is well below thegranite solidus UAHr..'S, 1982).

- 'fht: small quantity of pegmaritesimplies a lau: vapour saturation in thecrystallization history of the granite.

- The tare vugs seem to testify a local«segregation and accumulation of aqueousfluids as crystallization of granitic magmaproceeds ... » (BURNHAM and NEKVASlL, 1986,pg. 260; jAHNS, 1982; loNDON, 1986l.

- The aplite dykes are most likely theresuh of rapid nucleation and crystallizationfrom a residual melt intruded in extensionalfractures during the last stages of solidificationof the pluton. This coincided with a localsudden drop in fluid pressure and hence inthe volatile content of the magma.

Fluid indusions in the Monte PuJc.hianagranite complex

Sample selection and methodology

Chosen among many of similar lookingspecimens, five representadve samples ofleucogranite and one euhedral smoky quartz

crystal (Ienght: more than 10 cm, diameterabout' cm) from a vug in a pegmatite, havebeen selected and studied in detail. Somesamples from the pegmatite itself were alsocollected, but they appeared to be useless forboth the scarcity and the small dimensions offluid inclusions. In all of the samples, doublypolished plates of aoout 2 x 2 cm wereprepared. The thickness was lOO p.m forgranite samples, much higher (200 p.m) for thesmoky quartz crystal which was cutperpendicular to the «Clt axis. These sectionswere first observed with a conventionalmicroscope for the characterization of fluidinclusion dimensions, distribution and relativechronology (basic principles are described inWEISBROD, 1981; TOUR ET, 1977; andROEDDER, 1984: PJ! 1')0 s!Zi-!..).

Microthermometric investigations havebeen perfonned with several heating-freezingstages: Chaixmeca (POTY et al., 1976) at theFree University of Amsterdam; and U.S.G.S.­type (ROEDDER, 1984) at the University ofRome. Calibrations were macle using a numberof organic and inorganic compounds asrecommended in the literature (e.g.HOLUSTER et al., 1981).

Each measurement was repeated at leastonce to ensure reproducibility; the sameobjective (Leitz UTK 50, 32 x long frontaldistance) has been systematically used.

Some fragments of one euhedral quartzcrystal have also been cemented to sampleholders, carbon coated and analyzed(SEM/EDS, University of Siena) in order tocharacterize qualitatively some daughterphases which could have remained in openedfluid inclusions.

Fig. 2. - Fluid inclusions in M. Pukhiana granite complex. a) Primary muscovite bearing (a'TOw) inclusions inthe core zone of the euhedral quarft crystal in the pegmatire 'lug. b) Primary, crystal negative shaped. two-phase(L + V) aqueous inclU$iol1$ in the euhedtal quartz crystal (vug in the pegmatire). c) Ceneral view of fluid inclusiondistribution in the euhedraI quartz cry$lll (\I1Jfl in the pegmatitl:'J. cl) utt' lnils of two-phast' (L + V) aqueous inclusionsin a quanz aystal from rhe Ieucogranite (Gp, set' rext). e) A typical early rhrtt·ph~ (L -+ S + V) hdire bt'aringjnclU$ion (Cl)' (If,",",), in a quartz crystal in the Ieucogranitt'. 0 Large primary indus.ion, negadve cryStal shaped.and muscovitt' bearina (anow), from the euhedraI quartz aystal {vua in pegItlJIUtt'J. a)lnrenectioo of Gu dilferalltrails {$ee rext} in quartz from the leurogranite. In the Iowa right md of the pbotogrtph a group of Gl indU$ionsremaill$visiblt.betwttntwogt.l.It.IlIdons of inrersecting, late Cll trails (00t' dilfllSt'.Ind relatively nrly, several(five) orhers very late and sub-pal"1l1ld). h) Network of two-phase: (L + V) aqueous inclusioll$ (Cu) wirhin aleucogranite quartz crystal. i) SEM micropootosraph of an opened fluid inclU$ion in rhe euhedral smoky quartzcrystal (vug in pegmaritel; assemblage of platy cryuals identified as muscovite on the basis of borh morphologyand composition (set' texr).

Figures from a) to h) were photographed with plane polarized transmitted light and bar scale correspond to20 ,IIm. Bar in phoro i) givt'S scale in ,IIIO.

FUJlDS IN THE MONTE PULCHlANA n-rI1tUSION IN Sr\lU>INW: RESUl1S ON FWlO ETC. 109

.'.,.•

•

f

•

• ••

•• • •,••

•

•

,

•

'.a

•

•

110 M.L FREZZOTTt, C. GHEZZO. J. rouREr

Our results suggest that the recognition ofdawsonite in fluid inclusions when based onlyon opticala~ is doubtful and it shouldbe systematically confirmed by direct analysis(Raman probe, microprobe, etc.).

FIg. 3. - Fluid inclusion distribudon in a quartl crystalfrom the Icucogranite. G l , early inclusions (two-phaseIl'ld three-phase halite bearing); Gll : different trails ofhl.o-phase aqueous inclusions, inu:rsttting each other.1bc illustrated example is pl1I"ticularly demonstrative;solDCtimes there is a partial overl.pping of the twogroups. (X :wne: no inclusions in the present CllSC but,often some inclusions more or leS$ isolated difficult torelate to G j or G u groups).

Fluids in the leucogranite; G j (early, salineaqueous fluids) and GII (late low salinityaqueous fluids)

The distribution of fluid inclusions is verycomplex. As it was not possible to relate themto any growth patterns in the quartz grains,a dearcut distinction betw~n primary andsecondary inclusions can not be made. We canhowever establish a rdative chronology basedon the following criteria (ROEDDER, 1971,1984; TOURET, 1981; HOLUSTER, 1981): a)inclusions isolated or dispersed in clusters areearlier than those clearly related to more or

o0(;>

o

: (i7.. 0';

pe ... t)",.., 00,.

Main types of fluid inclusions

Fluid inclusions are the result of trappingof aqueous solutions, both in quartz from thegranite samples and in the smoky quartz fromthe vug. From the num~r of phases presentat room temperature, several types can ~recognized:

- Two phase aqueous inclusions (L + V):they are by far the most common type of fluidinclusions present in all samples. Theirdimensions vary from a few [0 40-50 p'm, withthe biggest inclusions in the smoky euhedralquartz crystal (Fig. 2b). Liquid/vapour volulm:ratios vary widely but the vapour ph~ neverexceeds 3090 vel. of the whole inclusion.

- Halite bearing aqueous inclusions(L + S + VJ: in the pJuronite. they are oftenpresent in clusters associated with two phaseaqueous inclusions (Fig. 2e). They contain acubic isotropic daughter mineral, which fromin appearance and refractive index (idendcalto w of quartz), is assumed to be halite. Therelative volume of the salt cube is quitevariable. The shape of these three·phaseinclusions is often somewhat irregular but notypical features of «necking down», thincapillary at one end of the cavity (ROEDDER,1981), have been ever observed.

- Afusawite bearing inclusions a.. +s+ V);in some fluid inclusions of the smoky quartzcrystal, radial aggregates of hairlike cryStals,strongly anisotropic, have been observed (Fig.2 a, 0. From the data in the literature andtheir distinctive appearance, they were firstassumed (FREZzOTI1 et al., 1986) to bedawsonite (NaA1(CO,)(OH);J (COVENEY andKELlY, 1971). However, despite greattechnical difficulties, some crystals could bephotographed and qualitatively analyzed with.he SEM/EDS (Fig. 2i).

The following proce<lure proved to be successful;I) brWcing small fntgrnents (about 2 mm in diameter)

of the doubly polished quartZ wafer, after miCfO$COpicobserv.tion.

2) Fixing vfitically the broken piece on the S.E.M.s.mpk hokkr, so that the broken face, 200,.m thick,can be dinctly observed IrIdtt the S.E.M. 1bree or fourchips can be assembled in • single holder.

Most cavities are empty but workablecrystals occur in some of them (less than 5%).

Qualitative analysis (major K, Al, Si) and theappearance of the crystal indicate muscovite.

FunDS IN TI-lE MONTE PULCHlANA INTRUSION eN SARDINIA), RESUlIS ON FUnD ETC. III

less healed fractures. b) Same types offractures are more or less coeval. c)Intracrystalline inclusion trails may be olderthan those which crosscut grain boundaries(i ntercrystaIline).

According to these criteria, inclusions in theleucogranite may have bean formed during atleast two different episodes (Fig. 2g and 3):

- Type G I : primary or early secondaryinclusions in small groups with no preferentialorientation; two phase (L -+- V) and three phase(L -+- S -+- V) halite bearing inclusions (Fig. 2e).

- Type G II : clearly secondary inclusionsconfined in healed fractures; constantly two­phase (L -+- V), no daughter phase is everpresent (Fig. 2d, h).

Melting dt1ta: G j and GII , two separate fluidgenerations. All Gn secondary inclusionsshow a range in temperatures of ice finalmelting (Tm) between -10 and O°C,corresponding to salinities from 05 to 8 wt.%NaCI equivalent (PoTIER et al., 1978).Despite the small size of the studiedinclusions, some temperatures of first meltingcould be measured between -20.8 and-22.8°C. These values are dose to theeutectics of the system HzO -+- NaCl at-2t.loC and the systemH20 -+- KCI-+- NaCI at _22.9°C (ROEDDER,

1971). Together with the final melting values,they suggest that the dissolved ions weremainly Na· with some K· but no Ca2 • andMgl' (TeHO-+-MgClz-+-NaCI:-35°C;TeH20 -+- CaCfz -+- NaCI: _52°C) (CRAWfORD,1981).

The composition of fluid G[ is moredifficult to establish. The presence of NaCIdaughter mineral in many G, inclusionsindicates an oversaturated brine at roomtemperature. From the size of the halite cube,the composition may be estimated to varybetween 26.3 and 55 wt.% NaC!.Microthermometric measurements wereattempted for two-phase G\,indusions closeto the halite bearing ones, ut they provedto be very difficult because of the small sizeof the inclusions. In most cases freezing couldnOl be detected, even at temperatures as lowas -100°C. For three of these inclusions, thefinal melting could be observed between -3.2

and _1.2°C. From the proximity and theobvious contemporaneity with halite bearinginclusions, we infer that these values do notcorrespond to the melting of the ice, bur ofhydrohalite (NaCI.2HzO). If ourinterpretation is correct, we would be on theNaCl rich side of the NaCl.HzO system,and a Tm of _3.2°C would correspond toa salinity of 24 wt.% NaCI eq. A lower limitat 24-25 wt. % NaCl eq. is tentativelyaccepted as an estimate of the order ofmagnitude. Unfortunately, the small size ofthe inclusions prevented the use of opticalcriteria, such as higher refractive index andthe difficulty to coalesce into single crystals,(SHEPHERD et al., 1985) which could be usedto positively identify hydrohalite.

In conclusion, the strong salinity contrastbetween G

land G[I indicates that the two

fluids are fundamentally different, and thatthey must correspond to separate episodes inthe fluid/rock interaction history.

Homogenization temperatures: 32 G Jinclusions of both two-phase (L -+- V) andthree-phase (L -+- S -+- V) were studied at hightemperatures (Fig. 4). In halite bearinginclusions two homogenization temperatureswere recorded: Th (disappearence of the gas

•

•"

'01' '40 '1/1 no Ni> _ ~t

Fig. 4. _ HiuogrlJIlS of homogeniz.ation lempertlura(oQ for inclusions in~ granite and in the: smoky quam:iD pegmatite vug (all homoReniutions occurred in the:liquid phase ThL). p, euhedraI quarlz crySlal from \'118(Iow salinity aqueous fluKh): GII , low salinity aqueousfluids in the granite (c:hemimy identical 10 p). Cn... andGUB correspond to the tWO peaks (180° and BOoerespectively as indicaled by arrows). G l, euly salinefluids in the granite (ThL for both the two-phase (.) andthe three·phase aqueous inclusions). Homogenization inhalite bearing inclusions occurred always before NaCIdissolution (ThL < Ts).

112 M.L FREZZOTIl. C. GHEZzo, J. roURET

Fig. 5. _ Different P, T evolution (isochores) followedby fluid inclusions with the same Nael content but withdifferent lb. (I), (2) and (3) correspond to an increasingdensity characterized by: (1) Ts < Th; (2) TI. Th; (J)T5 > Th. Thl, ru, Th3-homogenizalion temperaturesobserved in the: different inclusions (.fter WElS8ROI>.1984 and ROEDDu, 1984). Note that all the threecorrespond 10 the same TI. All C1Ise5~ in theM. Pulchiana (Gl fluids) c:orrespond to the i$OChorc (J).H _ halite; L • liquid; V .. vapour.

bubble, homogenization temperature, alwaysto the liquid) and Ts (dissolution of the saltcube), ThL is always lower than Ts, acharacteristic of relatively high density fluids(Fig. 5). Only four NaCl dissolutiontemperatures could be measured from 380°Cto 463°C, indicating salinities from 45 to 53wt. % NaCl equivalent. These results fit withthe visual estimates from the size of the saltcube, which suggest salinities from 35 to 55wt.% NaCl. Homogenization temperatures(ThL) for G1 (both two-phase and three­phase) range from 200 to 300°C. The absenceof vapour rich inclusions suggests that noboiling phenomena could have occurred.

Homogenization temperatures for G llinclusions are significantly lower than thosefor G" Th histograms (Fig. 4) show a looselydefined distribution with two maxima at 130and 180°C (respectively G UA and GliB)'There is a slight overlap between the highestTh in G}I (Fig. 4) and the lowest Th in G l ,

This is Significant in that there is always thepossibility that some biphase G, for whichno melting temperatures have been recorded,

Fluids in the smoky quartz of the pegmatite(<<P fluid»)

In thick section the crystal consists of asmoky rim of about 3-5 mm and an internalcolourless part of about 4 cm (Fig. 7). Thepresence of well defined growth zones makesit possible to recognize primary,pseudosecondary and secondary inclu'ions(RoEDDER, 1984). Microthermometricinvestigations have been performed onprimary inclusions related to successive steps

could infact belong to G II ·

Regarding the relative chronology, It'sfundamental to know precisely the evolutionof Th within Gll inclusions. Generally,temperature dispersion within a single trail isin the order of tenths of degrees; much lowerthan the overall variation in the histogram.This observation, common in most deepseated rocks (TOURET. 198U,Ieads to a pictureof successive fracture openings, trappingrelatively homogeneous fluids of changingdensity (Fig, 6). The example described inFigure 6, shows that the relative age of thedifferent trails is very difficult to establish.However, there is a general trend, which canbe observed in most comparable rocks,indicating a decrease in density for olderinclusions (lower Tb:: GliB in figure 8 forrelatively younger inclusions).

/38.7 ~ /45.6 f!JI9W~ "'-./ <> /25~ €J

<C> ",l22fJ c:£) ~138.1113.1, ~

- ~fJ-'; - -7 &r8/.6 " - ->-'-----< U" I

I • I

I refJ6 II 195E I

I • I

16'1SUfJ II I

I

I /!XJJrY,'I

Fig. 6. _ Homogenization temperatures in twOintersecting mms <fluid Gu, quartz in the leucogran.ile).All homogenizations in the liquid (ThU. ThL '"' numbersindicated near each inclusions (Oe).

,",'"''"

l· H t2<Pl

p

FUJIDS IN THE MONTE PULCHLANA INTRUSION (N SARDINW: RfSUUS ON FU1ID ETC. 113

a . b,c, d

x

y

Comparison of fluids in the granite (GlandG I) and in the pq;natite (P)

1be graph of homogenization temperatureS(111) versus salinity (Fig. 8) shows distincttrends:

G1

(saline fluids in the granite)corresponds to a relatively high Th whichpartially overlaps the Th values of the «Pfluid. in the pegmatite with a distinct gap insalinities. The wide range of salinitiesindicates an heterogeneity of the brines and

granite. The overall pattern therd'ore reflectsthe local oscillations and variations in acomplex system of hydrothermal circulationduring the late cooling stages.

-d~

Fig. 7. _ A~ schematic drawing of fluid ioclusiondistribution in a Ihick section (.1. to «c» axis) of thesmoky quartz crystal in the pegmalile vug. Only primaryand pseudosccondary inclusions are reported. (fIatchtd(UN correspoods to the IIlsI small smoky episode of quarugrowth). Below: topoklgka.! diuribution of Th valuesvenus diuarJCe from the con zone. Five diffcrenlpreferenlial distribulion zones (a-b-c-d-e) of fluidinclusions an: praml in I: inlemal zone and al least one(0 in y smoky rim. (l). (2), ()) and (4): representativeThL values sdccled for isochore interpretation (AI toA~, Fig. 10).

A~----,i-~x.-------+-Ih~ S'Th

240

200

160

120L--_-----:<---=-_--"----'

of crystal growth and they follow, in a moreor less precise pattern, some growth zones.Hence, the relative chronology is essentiallybas«J on the distance from the core of thecrystal toward its border. We have alreadyreported that some inclusions contain a solidphase (muscovite Fig. 2 f, n. These are onlya few, in the most internal part, among widelydominant two-phase (L + V) inclusions (Fig.2c). All inclusions in the quartz crystal areconsidered to represent a fluid phase that wewill call «P fluid •.

Me/ting and homogenization data: no saltcu~ has ever bttn o~rved. A number offinal melting temperatures could bedetermined between -4 and O°e. Therefore,the P fluid is assumed to be low salinity water(between 6 and 0 wt. % NaCl), very similarto the GIl fluid in the granite. The results ofTh determinations are plotted in Figure 4.They are scattered with no preferentialdistribution between 140 and 260°C. Figure7 shows the distribution of individual Thvalues labeled according to the successivegrowth zones of the quartz crystal.

A first important separation boundary islocated between the colourless internal zone(x) and the smoky rim (y). In the internal partat least five different fluid pulses haveoccurred (Fig. 7, a-e) with a general trend ofdecreasing temperature during the growth ofthe crystal core. Nevertheless, several smallreversals~ superimposed; in any given zone,the temperature jump is variable (up tohundred degrees). 1be zoning of the crystalthus reflects deposition by discrete fluid pulsesand a general d«rease in temperature withtime, but at the smoky border (y), this generalevolution changes abruptly and the last fluidpulse (0 reoches temperatures similar to thosefound in the core.

The variation between the x and y zonesis consistent with the hypothesis of somephysico<hernical discontinuity during the laststages of growth. However, the two fluidsmost likely have the same overall origin, asTm variation is very small and the only realdifference is indicated by the absence ofmuscovite in inclusions at the smoky border.Both types represent some variation in a fluidwhich otherwise is very similar to Gu in the

114 M.L FREZZOTn, C. GHEZlO. J. roURET

Th·c

400

a local NaCl oversaturation at the time oftrapping. The absence of vapour richinclusions points out that this oversarurationcannot be due to a boiling effect. From thisit follows that the dissolution temperaturesof halite (Ts) are not to be consideredminimum trapping values of the fluid, as isgenerally assumed for homogeneous fluidinclusions homogenizing cby halitedisappearencelt (Ts > Th) (BOONAR. 1982;ROEDDER, 1984). The trapping temperaturemust be below the maximum Ts NaClmeasured (Ts = 480°Cl, which in any case, iswell below any magmatic temperature. Thus,it is very difficult to know the originalcomposition of the circulating fluids.However I considering the salinity range from

.,,'"/ </G.J ()I' ~'ll

/ <.:?t .'/~O6~q,

24 to " wt. % NaCI eq., we can propose anaverage value of about 30 wt. %.

The situarion is more complicated for G)l

and P. In the Th-salinity diagram both fluidsare well separated, bU[ with a partial overlapbetween low salinity, low density fluids in thegranite (G

UA) and the highest density fluid

in the pegmatite (Pl, (Fig. 8). These P fluids,being in an euhedral quartz from a vug, areobviously the latest ones in the pegmatite.

Considering their identical salinity, it can betentatively assumed that G1( and P belong tothe same generation, but that fluids in thepegmatite present lower density (higherhomogenization temparature than in thegranite) .

•• ••

-~7>-d---------~ ----,. .. ./_............

•• - G_' -- I----

,.-, p, ,"": \" \,,1"1"" I

~-,,~~:"~/o 00 \ G>'.,'/0000 \ II A

. .-- .........~ .........~. I'6" 00 I G.: __~__ .... lIB

100

200

300

o 10 20 30 40 50 60NaCI cq. wt.OJo

Fig. 8. - Homogemzalion temperatures versus composition (wt.9b NaCI cq.1 in the syslrm Hp. NaCI.: Cl(.): three-phase ha.li.te bearing il1(:lusions; ClI (0): two-phase aqueous incllUions; P {xl: two-phase aqueousinclusions. Cl and C Il + P il1(:!usions are situated in different areas (Jolt~d linn); Cl distribution gives evidenceof a large salinity range with values from 24 to " wt.9& NaCI eq. which can sugaeS! ovenaturalion aI time oftrapping. As for G I biphase ioclusions the salinity rould nOl be determined precisely, Th measurements can onlybe represented as a field bound between NIClsltuution curve and the lowest qualitative estimate of Cl salinity(/NztclNd affll). P and GlI (in the low salimly fidd) arc distributed in such a way that at least the GJI inclusionshomogenizing at highest temperatures (Cu.J ca.n be considered to belong to the SarDC fluid episode than P. BolhG UA and P reprcsc:nt the SarDC fluid/rock inleractkln episode, the forma in the Icucogranite, the lalter in thepcglJlJl!ite euhedraI quartz C'J)'Sul. Hp + NaCl saturation and critic:aJ curves from PoTIEa and BlOWPI data (19771.

fUJlDS IN ruE MONTE PULCHlANA INTRUSION (N SARDINW: RESULTS ON FUJID En:. lIS

Fluid composition and P, T evolution

The interpretation offluid data wilt be basedon the following premises:

- in the granite, two fluids of verycontrasting salinity:

1) G] (24 to 55 eq. wt.% NaG) with anaverage representative composition of 30 eq.wt.% NaCl; Th between 200 and 310°C witha mean value around 260°C.

-~

fluids

O:P fluid» in the pegmatite, are proposed, aswell as a possible P, T evolution of theplutonite. In an earlier work on theleucogranite plutons GuEZZO and ORSINI,

(1982) have reported the low water contentand emplacement at shallow depths. RAMBozet al., (1982) have shown that any inclusionfluid which has a Ts significantly higher thanTh cannot have been trapped at theliquid + vapour boundary (Fig. 5), andtherefore cannot represent the liquid phaseof an unmixed fluid. These statements suppo"our findings (e.g. absence of vapour richinclusions) and confirm that the fluidevolution did occur in the miscibility field of!he H

20 + NaCI system without any boiling

phenomena. This separation betweenmagmatic aqueous fluids and melt in the fluidmiscibility field (for the system ~O + NaCIabove 1.2 kb at 700°C) gives an upper limit

j

<.11,0

meteoric water

P{kb)

2

1

2) G I1 10w salinity, (5 wt.% NaCI); Thbetween 100 and 220°C with two maximaaround 130°C (GnJ and 180°C (GIIA).

- In the pegmatite, one low salinity fluid(P) very similar to G]. with Th between 140and 260°C and no ~e£inite maxima at anytemperature.

In Figure 9, the isochores for G l and G uin the granite and the area, limited by theisochores for the lowest and highest Th, for

40 4b

100 300 5CX) 700 T{<'C)Fig. 9. - P, T interpretation of f)uid~ G1, Gu and P. (I), (21, 0), {in C/fC/n): representative selected isochoresfrom Th histograms (Fig. 4), corresponding respectively to Gr [magmatic, high salinity fluids), GIlA (udy, lowsalinity fluids), GliB (Iale low salinity fluids) in the leucogranite. {4a} and (4b): Minimum and muimum Th valuesfor P (Iow salinity fluids) in the cuhedral smoky quartz crystal in pegmalitic vugs. Respectively: (1): JO wt.9bNaCl.her WALTlIEJI. (1981); (2:) and (J):' wt.~ NaCl.fter RO£DDEJ. (l984); (4a) and (4b): J wt.~ NaCl afterROEDDU (1984). Amno: model of po5t-masmatic P, T evolution. Critical curve fO'l HJO + NaCl system afterPoTIu. et al.• (I977). Granite JOlidU$ after TI1TTt£ adn BoWEN (l~8). P, T conditions of granite empl~nt(gnry ./'U) discussed in lext; are. below the crirkal curve: P, T conditions of immiscibility for the HJO + NaCIsystem.

116 M_L. FREZW1Tl, C. GHEZzo. J. TOURET

of emplacemem for the p!utonite at a depthof about 4 km.

Probable magmatic origin for G1

It is well known that the first fluids whichevolved after the solidification of a granjremagma are generally highly saline (fluidimmiscibility between silicate and saline melt,e.g. Ascension Island, ROEDDER and COOMBS,1967; CATI-lEUNEAU et al., 1987). For thisreason, we propose that G j fluids areinitially of magmatic origin, butmicrothermometric data indicate a lateevolution and clear posunagmatic condilionsat the time of trapping; Th in fact correspondto isochores that would reach magmaticconditions (about 700°C) at 7 kb (about 23km depth considering lithostatic pressureregime), which is a contradiction for this kindof shallow intrusives. Magmatic immiscibilitybetween brines and silicate melt took placewhen the melt became water-saturated. Dueto the low initial H 20 content of themagma, this could only occur at a high degreeof crystallization, in which all of the primaryminerals, including quartz, had alreadycrystallized (these leucogranites are close tothe minimum eutectic composition, GHEZZO

and ORSINI, 1982). No evidence of the earlymagmatic stages hllS been preserved in therock, and this may be due to continous quartzrecrystallization at high temperatUtts. Finaltrapping has occurred at much lowertemperatures « 480°C) well after the endof crystallization.

1...IJte fluids: interaction with meteoric water(GII and P)

At a late stage (T lOO-300°C), the plutonis invaded by homogeneous low salinity fluids(GIl) that a.rt= also involved in the formationof the euhedral quartz crystals in thepegmatite vugs (P). The small differences inTh and salinity betw~n G1 and P can berelated to different time relationships andmechanisms of trapping. The structuralrelations of the inclusions, their low salinities(about 5 wt.9b NaCl) and low Th are

inconsistent with a pure magmatic origin. Inthe present state of our research, a meteoricorigin is much more probable, but it is notpossibile to state whether only meteoric watersare involved or if Gu result from a dilutionof magmatic with meteoric fluids. Furtherstudies on stable oxygen isotopes in fluidinclusions and in their host minerals arerequired in order to give a more definitiveanswer.

The comprehensive evolution of thehydrothermal circulation at these relatively«low It temperatures (Th from 100 up to260°C) must be linked to:

- the presence of a widespread systemof joints and microfractures, related to brittledeformation.

_ the cooling pattern of the pluton andcountry rocks which canrroUed the wholeconvective circulation svstem.

These factors allowed for the invasion ofthe pluton by meteoric low salinity water. Thejoint system, crosscutting also the vugs in thepegmatite pods, is likely to have connectedthem with hydrothermal circulating fluids. Inany case the local conditions were verycomplicated; this is revealed in the granite bythe presence of different generations ofmicro£racrures, characterized by different Thvalues in fluid inclusions, and in the euhedralquartz crystal by the complex distributionpattern of fluid inclusions reflecting asequence of discrete fluid pulses.

The euhedral crystals that line the vugs insome pegmatites record a considerable partof this fluid evolution during the late stagesof cooling.

Model of P, T paths

A precise delimitation of the P, T pathfollowed by the inclusions would require theknowledge of the pressure at the time oftrapping. This is presently very poorlyconstrained:

a) by the absence of boiling (P. T pathabove the H

10 + NaO critical curve, Fig.

9);b) by a possible maximum press~

correponding to P1hlwo';C at the time .ofgranite crystallization labout 1 to 1.5 kb, FIg.9).

FWIDS IN TIlE MONTE PULCHIANA L,<TRUSION (N SARDINIA): RESUlJS ON FWIO ETC. 117

PUlar) I ~ ,

~,...-

'."

'.V/'

T("CI.,..~

Fig. 10. - Detailed u:nmive interpretation of P,Tconditions during the growth of the euhedraI. quart:l: inthe pegmatite vug. 1,2,3,4: rq>rescmative iJochorf:;$selecled from topological dislribution of ThL in thequarl:l: (Fig. 7). Constant NaCI content (J eq. wr.q(,).AI> A 2 , A) and A.: P and T conditions at the growthof respective zones in quartz cry51al (Fig. 7). Unspecifiedpressure and temperature axes, correspondi"ll roughlyto the domain indicated in Figure 9.

The absence of discontinuity in Th valuesbetween Gp Gu and P suggests a cominuousevolution and, in the present state of ourresearch, the inferred model corresponds tothe arrow indicated in Figure 9 (isobariccooling trajectory around the pluton). In thismodd, the average fluidp~ has remainedfor a long period roughly equivalent to thelithostatic pressure at the time of granitecrystallization. Locally the situation mighthave been much more complicated, asillustrated by successive fluid pulses duringthe growth of euhedral quartz (Fig. 10).

The transparent quartz (core of the crystal,Fig. 7) corresponds to the cooling of our initialfluid at a roughly constant pressure (A

I-A

2•

Fig. 10). This epidose was not unique, butoccurred .several times. The last episode of thegrowth (smoky quartz y, Fig. 7) correspondsto the trajectory Aj-A. in Figure 10;probably at a lower pressure than A

t-A

2, but

by an unknown amoum.

Conclusions

In the investigated samples, two different

aqueous fluids have been characterized. Thesalinity gap between G I on the one hand,and G lI and P on the other, suggests adifferent origin. A magmatic origin isproposed for G1 and a meteoric one for G I

and P. In this hypothesis, highly saline fluidsmay occur in shallow inrrusive without boilingby direct magmatic immiscibility (ROEDDERand COOMBS, 1967). Similar results have beenobtained by a number of other researchers,e.g. SALEMINK (1985). HARRlS (1986),CATI-IEUNEAU et al. (1987). Confirmationwould evidently be obtained by stable isotopestudies; however this is not straightforward.The meteoric signature of P (vug in thepegmatite) is in principle not difficult toestablish, but the situation is entirely differentfor the granite. The G1 and GII inclusionsare so intimately intenningled that only singleindusion analysis could allow the precisedetermination of the isotopic signature of eachgroup.

Aclmowledgem~lS. - We wish to thank CA Ricci forthe stimulating discussions; I. Memmi and C Balddlifor the invaluable hdp at the SEM and microprobe, andP. Lattan:zi for extensive and conslructive remarks onan arlier veo:ion of the drafl 01 the manuscript. M.L.F.acknowkdgl:' B. De Vivo for his suggl:'uons whenworking at the: U.S.G.S. heating frc:e:zing Slage in Romeand also the Vakg.roep E.P.M. (Vrije Universiteit) fOl'their help and supporl <bing her stay in Amslerdam.WOI'k supported by CN.R. grtnU n. 8'0098'10' (toe.G.) and An. 20/2.7 Regal. 26/1/67 Posi:l:. 121.0'720Prot. 067884 (to M.L.F.).

REFERENCES

BoONAIl R.j. (l982)· Fluid inclusions in porphyry ty~

deposits. Course notes, Mineral. Deposits ResearchReview for industry. Penn. State Univ. April 6-9, RB!·RB 2'.

BRAUA A., GItEZZO C, GUASPAItItI G., SABAnNt G.(1981) . As~tti genetici tk/ />(Itolite sardo-eol'W. Rend.Soc. It. Min. Petr. J8, n, 701·764.

BIt017.U P., MORBIDELU L.. TRAvaSA G. (1978) •UTrltJci~firi e dJimici tk/k sieniti ~/ SarrabJIsMeridi01fllle tJ-rdegna SIJd~en/4.le). Period. Mineral.•47. 83·98.

BRaTZU P., Fuaoo V.• MAss U. (1983)· Slilbk.iJl)to~

I!«hemistrJ 0/ Herqnuln Gnmitoid ffKh from TheSamlbus M.ssi/ (Southustem Sardinill. fury). IsotopeGeosc., I, 77-90.

BUItNlIAM CW., No::vA.slL H. (1986) . Equilibriumproperties 0/ granite pegmatite magmas. Am.Mineralogist 71, 239·263.

118 M.L FREZZOTn. C. GHEZZO, J. roURET

C"TIIEUNEAU M., DUBESSY)., M"ItIGNAC c., ParY B.,WF.IS8ROD A. (1987) - Fluids in granitic ~vironmen1S.

(AbsIT.). Soc. It, Min. Petr., Meeting on .. Granitesand their surroundigu, Verb.nia (Italy), 20.9/3.10.

COCHERlE A. (1984) .]nfeJtKtion mllntMU<W){j~:JQn 'Ok.la"s la gmN4'au«U1!ion plutoniquntllko-ll/ulinn,contraintn ciot:himiqun (tlhnmn m t1IUd d iwwpntiu stroIttium d tU J'cnrit.end. 1nCsc d'Elat, UniVffSitede Rennes I. 24S pp.

COVENEY R.M., KEU.Y w.e. {I97l} • Dawroniu asdaught~ minn-al in bydrothnma/ fluid inclusions.Contrib. Mineral. Petrol., 32, )34-342.

CRAWPORD M.L (1981) -p~ tijui/ibria in aqUl:Omf/uiJindusions. In: Fluid indusions: Applications toPctroloxy. Holli$1er L.S. and Crtwford M.L. Eds.Minttaf. Assoc. Canada Soon CO\J.fSe Handbook, 6,7HOO.

DEL MOllO A., DJ SL"lPUCIO P., GuEZZO C., GUASP.uJ.1G., RrrA F., SAllATINI G. (197}j· &diomc:tm utaand int7Jlsive ~qumu in tlM S4rdinia Batho/ith. N. Jb.Miner. Abh., 126, 1,28·44.

DE VIVO B., FREZZOTIl M.L., GIlEZZO C. (I98~)· FluidincluJionJ from M. PulchiaNl H~niJIn kucogTaniticinl7uJion. (Abm.). Gruppi cl lavoro CN.R. Pakozoicoe EvoIuzione Magmalica della Crosta Fanerozoica.Meeting on «Evoluzione Stratigrafic., Tenonic.,Metamortic. e Sll'1Jlturale del Paleozoico haliaoo.,Siena, 13·14 Detcmber.

DI StMPUCIO P., FEKKARA G., GuEZZO C., GUASPAKRIG., PI'.I.lJZZER R., Rica CA., RrrA F., SASATINI G.(1974) . /l m(/fJmorjiJmo (it magmatiJmo pakozoicotklUt Smr.kgna. Rend. Soc. h. Min. Pelt., 30,979·1068.

FJ..\NCESCUEUl M., MDof.'u I., Rico CA. (982) •Zonft>fTllfU mntlmo/ic4 tk,,,, Sanlq,n. Mt/nltrioJl4k.In: Guida alIa Gcologia del Palcozoico sardo. GuideGcologiche Regionali Soc. Gcol. h., 137·149.

FREZZOITI M.L., GuEZZO C, TOURET j,R.L. (1986)·Fluid incluJiom as pOJt magmatic traj«tory tTaCf!r:application to M. Pukhitma Pluron (N Sardinia).(Abm.). Soc. h. Min. Pw. Mcclingon «RouIo deinuidi neUa petrogenesi •. Siena. 3·4 June:.

CIIEZZO C., QulJl.1 j.B. (1982) . UnUlllmti Jlruttu",/i~ cOlllpouzioNJ/i ut btltolit~ m:inico utJo.corw inSilnUznll. In: Guida alia Gcologia del PalcozoicoSardo. Guide GcoIogichc Regionali. Mem. Soc. CcoJ.11" 24, (suppl. Bl, 16~·181.

GIRAUll L. (I983) . Contribution aUt connlliuanu usgrilnitoiks COffO·UlrtkJ. Etud~ Ntl'Ogrllphi'lu( (tStructura/t' ik1 VlInitoikJ M '" rigiOll Zicaw-Siln/fJLucill-Jj·TlIiulno (Co~ du Slid). These d'Etil, Ail[·Mat$Cille, 297 pp.

GUASPAUI G., RICCOBONO F., SABATlNI G. (1984.) .Uucoguni~ofIM StmJinitl &.thoIith:P~ ilJP«lJIInd thrir 1't"~lJlInu to mdIJ//og(11~il. Pcriodico diMineralogia, ~3, 17·~2.

GUASPARRI G., ROCCABQNO F., SASATINl G., (l984b)·Hrn;ynian Mo·minerali1.JltionJ of porthyry·sJyk in t/,(Sardinian btltbolith. A discuuion on t/,( gm~iJ IIndcomf)flriJOll with ohkT tkposits ofIM fantiq. Rend. Sot.h., hUn. Pelt., 39, 11, 629·648.

HUllS C. (1986) . A 'JUiIntilillit't: Jtlllly of MIIfl1lJlticinduJionJ in t« plutonit: ~tT:1lI ofhtncion IsUtnd. j.

Pellol., 27, 2~1·276.

HOUISTEIl L.S. (1981) ·lllformtltion intrimictrtly avaiUtbkfrom fluid incfusior/J. In: Fluid inclusions: Applications10 Petrology. Hollister L.S. and Crawford M.L. Eds.Mineral. AJsoc. Canada Shorl Course Handbook, 6,1·12.

HOL.LIST£lI L.S., Buuuss R.C, ROEDDER E., SF'OONEaE.T.C, TOU1IET j.L.R. (l9811 . Praclical aspects ofmicrothcrmomctry. In: Auid inclusions: Appl.icaliofl5to Petrology. Hollisu:r L.S. and Crawford M.L. Eds.Mineral. Assoc. Canada Short Course Handbook, 6,22~·2H

jAN1IS R.H.(1982l . I/ltcm<J! ~ll<lion of{J(gmlltit( bodi~.In: Granitic pegmatites in science and induSlry. CcmyP. Ed. Mineral AJsoc. Canada Short CourseHandbook, 8, 293·327.

JANUS R.H., BuV.1IAM C.W. (1986) . ExpnimN//fJIJludi~ OJI' pqptllht( J,ntms: I. A fttOfklfor tk Jnrv.tionIInd cryJllIlli1.Jlt;on ofgnmitic pq,ntlltit(s. Ec. GeoI.., 64,843-864.

KOWAU.IS B.}., WANG H.F., Bo-ANjANG (1987)· Ht!akdmicrocrack ori(n/fJtiOnJ in gmnit(from IIIinoiJ borehokUPf-/.} and tbm reUttiomhip to tk rock'J Jtrt!SJ byJlory.TccloniphysKs, 13~, 297·306.

WPfNASSE M., PEotEK A. (1986) . MicrofisJlmng and"fjonlll stTmfi~lJ: W CUllllP/t' of tk Mll1t1N gTrIni~,

MaJJiICmtml, Frrlnu.}. Suuct. Gcol., 8, 169·180.loNDON D. (1986) • Fomustion of touml4lin( rich gon

pockm in millrolitic Mmati~. Am. Mineralogist, 71,396·40~.

ORS1Nl j.B. (1976) . Us granitoiJ~herqniens cono-Jartks..m;~ en ftJirkna Ik deux assa:iJlhOnJ magmatiqu~. Bull.Soc. Gcol. Fr., 18, 1203·1206.

OIlSINI j.B. (1980)· u btlthouk rorw-Jilnk: un f!1<4I11P/t'd~ blIthofj~ hneyffi~n (Jtructure, composition,CltfIInillltion d'mJmIbk). Sa pm dtmJ Ut cbain~ tYtrisq~

uI'Euro~",oym~. These Ail(·Marscille UT, 370 pp.Pf.CIlF.R A., LESPINASSE M., LEROY j. (198') - ReUttionJ

b(twun fluid incluJion trails and stl't'JS field.. a too! forfluid chrollO/ogy . An =mpfcofan intro?/'ilnitic uroniumort! upout (northwm MaJJif Ctwtral. Franu). Lithos.18,229·237.

PoTru R.W., BROWN D.L. (1977) . Tk vo/llmnricptopnti~ of tl'JU«JIIJ chlOlTU from OD to JOOOC IJtprmltTr up to 2000 him bmd OJI' a rr-grmion of_iUthkdalll in t/,( litt!rllturt!. U.S. (;eo!. SUr\'. Bull., 1421·C

Pm'n,R R.W., CLYNNE M.A., BROWN D.L. (1978) .Freezing point d(prt'JSion of lI'luoom Jodium chloriJ(solution. Ec. Gcol. n, 284·28~.

Pan B., LEROY J..jAauMoWlcz L. (1976)· Un nouvclapparril pour '" mt!Jllrt! Us tcnplmtllm SOIlJ krnicroJropr. {'inJIlII"'tion U mkrotht!rmornitrj( CbtziJtMica. Bull. Sot. Frarv;. Mi~ral. CrislalIogr., 99,182·186.

RA1\IBOZ C, PlatAVANT M., WEISBROD A (1982)· Fluidimmiscibility in natuml prrJCt!JSt!S. U~ and mi$u" offluidinclusion data. 11. IlIterpretation of fluid inclusions intnmJ of immisdbitity. Chem. Gcol., 37, 29·48.

ROEDOER E. (1971) . FlJIid incluJion Jludi~ on tkporphyry.ty~ ore UpoJitJ at Binghllm, UliIh, BUlk,Montilnil, aNiC/imllX, Co/otuJQ. £C. GcoI., 66, 98-120.

RoEooD. E. (1981)· Origin offluid incbniomtlJll1 chtlntpthtlt occur IJ/tt!r trapping. In: Fluid inclusions:

FLUIDS IN THE MONTE PULCHIANA INTRUSION (N SARDINIA): RESULTS ON FLUID ETC. 119

Applications to Penology. Hollister L.S. andCrawford M.L. £ds. Mineral. Assoc. Canada ShortCOUI"K Handbook, 6, 101-137.

ROF.DDEI. E. (1984) - fluid InduJiom. P.H. Ribbe Ed.Mineral. Soc. Am. Reviews in Minenl., 12,644 pp.

RO£DDEI. E., CooMBS D.S. (1967) . Immiscibilil1 in~~_~~~w~~iJ~~mm~w

gnrnitk bltxh from AJcmtion Iu."J. J. Pcotrol., 8,417-4~1.

Rossl PII. (1986) . Orgpni:twion et vniw d'lln gl'll11Jbatboli~ O1Of!in;Qw k batho5~uft:o....blink Ut CcMt.These d'Elil, Univcol"Sile de Toulouse, 292 pp.

SALI!MINI{). (l98'j) . steams at 5mjos (Gn'«t'J. ThesisUtrecht. GeologiClil Ultuicclina. 220 pp.

SHfJ'IIEllD T.)., RANKlN A.H., AwEl.TON D.H.M. (l98'l. A PrrIttical Guik to Fluid Inclusion Studin. Blackico£d. 239 pp.

TOUIlET ).L.R. (1977) - T~ signijicanu offluid intlusionIill metamorphit rocks. In: Thcormodynamics inGeology. D.G. Fraser Ed., D. Reidel Pub!. Co.,Dordrcochl, Thco Nelherlands, 20)·228.

TouRET J.L.R. (1981) - Fluid indusionJ in high grakmt!lilmorphic rocks. In: Fluid inclusions: Applicalions10 Pcouology. HoIliSler L.S. and Crawford M.L. £Ps.Mineral. Assoc. Canada Shorl Course Handbook, 6,182·208.

TlllTLE O.F. (1949)· Shuttumipm%f:1 ofp/Ilnn ofjluiJim:lwionJ). Geol., 'j7, HI-)'j6.

TUTT1..£ O.F., BoWEN N.L. (l9~8)· Origin ofVfUli~ intiN Light 0/ t!%{Jt!rimmlill Studi{!S in tbt! rystnnNIZA/Sip,.KA/Si,o,.SiO)·Hp. Gcol. Soc. Am.Mcom., 74.

\'(IALTIIER) (1981) - Fluitk Eim:hlMSSl! im apatit tkscamonatitI IJOm K4ismtuhJ (Okrrbt!igra~). Din.Karlsruhc:, 19'j pp.

\'(IEISIJROD A. (1981) - Fluid induJi011J in shal/QwintruJivn. In: Fluid inclusions: Applications 10Pcolrology. HolliSlcor L.S. and Crawford M.L. Eds.Mineral. Assoc. Canada Shot! CourSe' Handbook, 6.

WEISIlI.OD A. (1984) . Utilization des inclusionI jIuid~en geothermobarontftrit!. In: Th~mometrjco COlbarometric Geologiques. lagache M. Ed. Soc. Fr.Min. CriSI., 41'j·4~9.