Pyroxene Exsolution: an Indicator of High-Pressure Igneous

American Mineralogist, Volume 73, pages261-273, 1988

Pyroxene exsolution:An indicator of high-pressureigneous crystallization of pyroxene-bearingqv rtz syenite gneissfrom the High Peaks region of the Adirondack Mountains Plur- W. Or,r,rr..q. Department of Geology and Geography,Vassar College, Poughkeepsie, New York 12601, U.S.A. How.q.nr W. Jlrrn, Er.rzlnnrs B. J.q.rrn Department of Geology and Geography,University of Massachusetts,Amherst, Massachusetts01003, U.S-A-

Ansrn-q.cr Relict, coarse,high-temperature inverted pigeonite lamellae in host augite and augite lamellae in host inverted pigeonite are locally preservedin metamorphosedigneous rocks from the Adirondack Mountains of New York State. The exsolution history of these pyroxenesconsists of (l) crystallization of high-temperatureaugite or pigeonite, (2) exsolution of coarse "00l" lamellae of augite in host pigeonite and pigeonite in host augite, (3) fautting of lamellae and then of host along (100) as lattice parameterschange during cooling, (4) inversion of pigeonite to orthopyroxene,and (5) decomposition of the lamellae to produce intergrowths ofaugite and orthopyroxene along (100) ofhost grain. Thesetextures have been observedin pyroxenesthat spanthe full rangeofcompositions found in Adirondack anorthositic and syenitic rocks. An inverted pigeonite from a pyroxene-bearingquartz syenite gneisscollected in the Mount Marcy quadrangleyields a rein- tegratedcomposition of Wo,r,EnorFsro, indicating that igneouscrystallization of this rock took place at pressuresgreater than or equal to approximately 9 kbar.

gioclase,pyroxene, hornblende, and garnet, occurs at the INrnorucrroN margins of the massif and in localized shearzones within This paperdescribes pyroxene exsolution textures found the core of the massif. Syenite gneissestypically overlie in metamorphosed igneous rocks from the Adirondack the gabbroic anorthosite gneissfound at the margins of Mountains of New York State.These exsolution textures the anorthosite massif. are petrologically significantin that they allow one to "see Valley and O'Neil (1982)and Valley (1985)presented through" the effects of regional metamorphism and to evidencethat anorthosite in the Adirondacks was intrud- study the igneoushistory of the rocks that contain them. ed at relatively shallow levels (<10 km) and then later Pyroxenesdescribed in this report were found in rocks underwent granulite-facies metamorphism. This was a that range in composition from gabbroic anorthosite to significant finding in that it was the first actual evidence pyroxene-bearingquartz syenitegneiss and were collected that Adirondack anorthosite was not intruded under during our geologic mapping of the Mount Marcy and pressure-temperatureconditions similar to those that oc- Santanoni quadranglesin the High Peaks region of the curred during regional metamorphism. This finding gave Adirondacks. Sample locations are shown on Figure 1, suppolt to the suggestion,made on the basis of analogy and sample descriptionsare given in Appendix 1. with Labrador anorthositemassifs, that anorthositein the The Marcy anorthosite massif crops out in a roughly Adirondacks may have intruded under anorogenicrather heart-shapedpattern and coversapproximately 3000 km'z than synorogenicconditions (Berg, 1977;Emslie, 1978). (Fig. l). It has generallybeen charucterizedas consisting Evidence presentedin this paper suggests,however, that of a core of anorthosite surrounded by gabbroic anor- pyroxene-bearingquartz syenitesclosely associatedwith thosite(Buddington, 1939,1969; Davis, 1971).Mapping and commonly thought to be coeval with the Adirondack by Ollila (1984),however, has shown that there are sig- anorthosite were intruded at depths greater than l0 km nificant amounts of gabbroic anorthosite within the core and perhapsas great as 30 km. The most important evi- regions of the massif. Core rocks are characterizedby denceleading to this conclusionis igneousexsolution tex- igneous textures. Subophitic gabbroic anorthosites and tures in Fe-rich pyroxenesfrom quartz syenites.Low-Ca gabbros in which garnet coronas around mafic minerals pyroxenesthat exhibit these textures are sufrciently Fe- are the only evidenceof regional metamorphism are rel- rich that they only could have crystallized at higft pres- atively common. Gabbroic anorthosite gneiss,character- sures.In order to develop this argument, it is first nec- ized by plagioclaseaugen in a granoblasticmatrix of pla- essaryto briefly describeexsolution in pyroxenes. 0003-004x/88/0304-026I $02.00 26r 262 OLLILA ET AL.: PYROXENEEXSOLUTION

74030' 44ogo' ARANACLAKE LAKEPLACID AU SAELE

,'/ | Bloomingdole /\. | /- t ,, z. /' r -' ll ),1' "H,l )\'):. ' 't/'/ -',t\ /\ \.zt. \ | l:.\ w&/ r /'r// // ^ ),--XV/r l t ,,F i-:)'-Y/ t >>,

-\ -- >W:,'); ,"'iii2)'!)"":i l;L'/,lr'( r-r,r-4{{{ \ ,. I \ E/ /^\ /g-17 /'l\ | ..a/ - t /t I /\ / /ra - / -- .. / \'rr. t

t ,'- ),)\ \ tll/l \ //l PARADOXL ,, r. zl i2Tilta/> l\' \ /'ltf \\ /,r, t \ \/ ./n-{ /

4tP.t5' z4030'

t---l l-vT-Tl l++l

ANORTHOSITE CALC - SILICATE ROCKS GRAN|TtC,SYENtTtC, 8 MIXED GNEISSES

Fig. 1. Generalizedgeologic map of the Mount Marcy anorthosite massif. Locations for samples9-17 , AL-4, BA-5, GB-2, and PO-17are indicatedon the map.

are functions of pressure, temperature, and chemical PynoxnNn ExsoLUTroN composition, the orientation of the best-fit planes is also Optimal phaseboundary theory was used by Robinson a function ofthese variables.The rapid expansionofpi- etal.(1971, 1977), Robinson(1980), andJaffeeral. (1975) geonite as it goes through the P2,/c = C2/c reversible, to explain why pigeonite lamellae in an augite host and nonquenchabletransition is of particular importance in arrgitelamellae in a pigeonitehost are oriented along non- interpreting exsolution temperatures. At temperatures rational planes.The theory statesthat monoclinic pyrox- above the stability of P2,/ c pigeonite,a and c ofpigeonite eneswith identical D dimensions and differing a, c, and are greater than a and c of augite.At temperatureswhere B will have two planesof exactdimensional fit. One plane, P2,/c pigeonite is stable, the reverse is true. Becauseof "001," is closeto (001), its orientationbeing controlled theserelationships, high-temperature lamellae of pigeon- by the diference between the c dimensions of the host ite in host augite or augite lamellae in host pigeonite ori- and lamellaepyroxenes. Another plane,"100," lies close ent in obtuseangle p [c rr "001" < c A (001)];whereas to (100) and is controlled by the differencebetween the c low-temperaturelamellae orient in acute angle0 Fig.2). dimensionsof the pyroxenes.Since the lattice parameters High-temperature lamellae are much coarser (approxi- OLLILA ET AL.: PYROXENE EXSOLUTION 263 mately 10-100 pm thick) and spacedfarther apart than lamellae formed at lower temperatures, which are less than I pm thick. Examples of high-temperature exsolution in low-Ca pyroxenes (inverted pigeonite) are relatively common in plutonic igneousrocks and have been described from many localities. This type of exsolution was originally describedby Hess (1941, 1960) and has been described in Adirondack rocks by Davis (1971), Ashwal (1982), and Jaffe et al. (1983). Robinson et al. ( I 977) describedhigh-temperature ( - 1000'C) exsolution in host augitesfrom the Bushveld complex in South Af- rica and the Nain complex in Labrador, and more recently Livi (1987) has describedsimilar exsolutionfea- tures in pyroxenesfrom the Laramie anorthosite massif. Beforethe work of Ollila et al. (1983, 1984),this type of exsolution had not been recognizedin the Adirondacks. In contrast to the exsolution textures visible in monoclinic pyroxenes, exsolution between orthorhombic and monoclinic pyroxenes is much simpler. The only plane that is at all similar between orthopyroxene and clino- pyroxeneis (100), and lamellaeof clinopyroxenein orthopyroxenehost or orthopyroxenein clinopyroxenehost orient along this plane. As pointed out by Robinson Fig.2. L .oa.r basedon ol."*u,roo, o, orro*Ln"textures (1980),this type of exsolutionis common in metamor- for the exsolutionhistory ofigneous augite in the Adirondacks. (l) of homogeneousaugite; (2) phic pyroxenesand in magnesianigneous pyroxenes that The sequenceis crystallization exsolutionof coarse"001" pigeonitelamellae, which orient in crystallizedat temperaturesbelow the stability of pigeon- obtuseangle of the host augite;(3) faultingof pigeonitela- during the late B ite. This type of exsolution also occurs mellaeduring cooling (Robinson etal., 1977);(4) faultingof the stagesof exsolution of igneouspyroxenes. host augitealong (100), inversion ofpigeonite and decomposi- Exsolution textures in Adirondack rocks are compli- tion of"001" invertedpigeonite lamellae along (100) ofthe host catedby the fact that igneousrocks have been affectedby augite;(5) formation of pyroxene"mesoperthite"-all the ortho- a regional metamorphism. This metamorphism has re- pyroxenethat originallywas in "001" lamellaenow forms(100) sulted in complex exsolution textures that have been lamellaein hostaugite; (6) orthopyroxenelamellae coarsen, and poorly understood. In a few locations, however, igneous fine pigeonitelamellae, which orient in acuteangle B of host (100) rocks were only slightly deformed during regional meta- augite,exsolve from hostaugite. Note that low-temperature as to protrudebeyond morphism, and pyroxenesin rocks from these localities orthopyroxenelamellae may exsolveso the confinesofthe augitehost. retain recognizablehigh-temperature exsolution textures that have proved to be the key in unraveling the exsolution history of Adirondack pyroxenes. weakly foliated and consist of medium- to fine-grained Figure 2 is a schematic diagram showing a model for granoblasticaggregates of garnet, pyroxene, and feldspar, the exsolution history of metamorphosedigneous augite. but also contain coarser (>2 mm) pyroxene grains that StagesI through 3 are basedon descriptions ofunmeta- retain relict igneous exsolution textures. Augite from morphosed igneouspyroxenes (Robinson et al., 1977; sample9-17 showstextures equivalent to stage4 in Fig- Robinson, 1980).Stages 4 through 6 are basedon obser- ure 2. Augite in samplesBA-5 and GB-2 show textures vations of pyroxenes from Adirondack rocks. The dia- intermediate betweenstages 4 and 5 of Figure 2.It should gram is applicable to both augite and pigeonite hosts,but be emphasizedthat these sorts ofexsolution textures are inverted pigeonite that retains coarse "001" augite la- quite rare. Medium- to coarse-grainedaugite in meta- mellae is much more common than host augite with rel- morphosed igneous rocks from the Adirondacks most ict, high-temperature "001" inverted pigeonite lamellae. commonly shows exsolution textures corresponding to Figures 3 through 5 are photomicrographs ofgrains that stage6 of Figure 2. Augite in gabbroic anorthosite gneiss illustrate the exsolution processesoutlined in Figure 2. typically occurs in granoblasticaggregates and only con- Relict, high-temperature exsolution lamellae in host tains fine, low-temperature"001" and "100" pigeonite augite were found initially in a coarse-grainedgabbroic lamellae such as illustrated betweenthe coarserlamellae anorthosite from the central part ofthe Santanoni quad- in part 6 ofFigure 2. rangle(sample 9-17, Fig. 3). This rock is subophiticwith Inverted pigeonite containing coarse "001" augite both augite and inverted pigeonite interstitial to plagio- exsolution lamellae is common in Adirondack anortho- clase.Subsequently, similar exsolutiontextures were found sitic rocks(Davis, 1971;Ashwal, 1982;Jafe et al., 1983). in rocks from the central part of the Mount Marcy quad- Suchis not the casefor syenitic rocks. Davis ( I 97 I ) stated rangle(GB-2 and BA-5, Figs. 4 and 5). Theserocks are that inverted pigeonite was never observed in syenitic 264 OLLILA ET AL.: PYROXENE EXSOLUTION

Fig. 3. Photomicrograph of host augite from sample 9-17. Host augite is at extinction, and relict high-temperaturepigeonite lamellae now inverted to orthopyroxeneare oriented from lower left to upper right. These lamellae are faulted and have partially broken down to produce an intergrowth ofaugite and orthopyroxene along (100) ofhost augite (upper left to lower right). This grain correspondsto stage4 tnFig.2. gneissesfrom the Saint Regisquadrangle. Bohlen and Es- high-temperature exsolution lamellae were found in the sene(1978) describedinverted pigeonitefrom Adiron- northeasterncorner ofthe Santanoni quadrangleand the dack pyroxene-bearingquartz syenite gneiss,but the py- northern part of the Mount Marcy quadrangle (Fig. l). roxenesthey describedlacked the distinctive exsolution Inverted pigeonitewith coarse"001" augite exsolution texturesofinverted pigeonite.Bohlen and Essene'scon- lamellae was found in sample AL-4 (Fig. 6), a medium- clusions were based on the reintegratedcompositions of to coarse-grainedpyroxene-plagioclase-quartz-micro- exsolved grains that contain abundant augite exsolution perthite gneiss.Compositionally similar, but finer-grained lamellaeoriented along (100) of host orthopyroxene.As and more strongly foliated rocks located in the southern can be seenin Figure2, the assumptionthat thesegrains part of the Santanoni quadrangle do not contain relict were originally pigeonite is quite reasonablegiven the high-temperature exsolution lamellae. Exsolution tex- effectthat regional metamorphism has had on exsolution tures are not as well preservedin syenitic rocks from the textures. Mount Marcy quadrangleas in sample AL-4, but sample There is one potential problem, however, in interpret- SC-6 (Fig. 7) contains what are interpreted as epitaxial ing grains that only consist of augite and orthopyroxene intergrowths ofpigeonite (now inverted) and augite, and intergrownalong (100). Rietmeijer (1979), in an exten- both samplesPO-17 (Fig. 8) and SC-6 contain orthopy- sive study ofpyroxene exsolutiontextures in syeniticrocks roxene grains with abundant augite exsolution lamellae from the South Rogaland complex in Norway, pointed orientedalong (100) ofthe host orthopyroxene.Samples out that quite similar texturescould be produced by both SC-6and PO-17 also contain host augitewith relict coarse exsolution processesand by epitaxial overgrowths. Some "001" lamellae. of the criteria Rietmeijer used to distinguish between exsolution lamellae and epitaxial overgrowths included CrrBnrrsrnv the width of the lamellae, the shapeof the lamellae, and Mineral analyseswere obtained on an ETECAutoprobe the presenceor absenceof inclusions along hostJamellae at the University of Massachusetts.Microprobe analyses boundaries.These kinds of observationscannot be made were performed at 15-kV acceleratingpotential and an if all high-temperature lamellae have broken down to aperture current of 0.03 prA.Standards consisted of both produce augite-orthopyroxene"mesoperthite." It is un- synthetic and natural silicate and oxide minerals, and certain how prevalent this phenomena is in Adirondack corrections were made following the procedure of Bence rocks, but this possibility decreasesthe certainty of ig- and Albee (1968)and usingthe correctionfactors of Al- neous crystallization temperatures based on integrated beeand Ray (1970).Analytical resultsare listed in Table compositionsof grainsthat do not show relict high-tem- l. Reintegratedanalyses ofexsolved pyroxeneswere ob- perature-typeexsolution lamellae. tained by a number of techniques. Most reintegrated Syenitic gneissesthat contain pyroxenes with relict, analyseswere obtained by combining microprobe anal- OLLILA ET AL.: PYROXENE EXSOLUTION 265

tioning of the electronbeam. Lamellaecompositions were determinedby comparisonwith host spectra.Results from theseanalyses are listed in Table 1. Optical reintegration techniquesofthe grain shown in Figure 8 were evaluated by scanningcrack-free areas as large as possibleand ana- lyzing for Fe, Mg, Ca, and Si. Results of these analyses are shown in Figure 9. These analysesindicate that the optical reintegration techniquesprovide very reasonable limits on pigeonite compositions in this rock.

Tnrvrpnru.ruREs AND pRESSUREoF CRySTALLIZATION Igneous crystallization temperatures for pyroxenes in anorthositicand syeniticrocks from the Adirondackshave beendetermined previously by Ashwal (1982)and Boh- len and Essene(1978). As first pointed out by Bohlenand Essene(1978), reintegratedpyroxene temperaturesfrom a variety of metamorphosedigneous rocks from the Ad- irondacks yield temperatureswell above regional metamorphic temperatures.Bohlen et al. (1985) estimated maximum regional metamorphic temperaturesto be approximately 800 "C. Figure l0 showsintegrated pyroxene compositionsfrom Bohlen and Essene(1978) and this report. Crystallization temperaturesrange from approximately I150 "C for gabbroicanorthosite (9-17) to I100 'C for metagabbro(GB-2) and 950 to 850 "C for syenitic rocks (BA-5, PO-17, P-10). Pyroxenecrystallized after plagioclase in anorthositic rocks, so liquidus temperatures must have been somewhathigher than the pyroxene temperatureswould indicate. The depth at which anorthositic rocks crystallized is Fig. 4. Photomicrographof hostaugite in sampleGB-2. Fine difficult to determine in the Adirondacks, becauseregion- (100)orthopyroxene exsolution lamellae are nearly vertical.Rel- al metamorphism has overprinted any contact aureole ict high-temperaturepigeonite lamellae are horizontal. This grain assemblagesthat might have been useful in this respect. is intermediatebetween stages 4 and 5 in Fig. 2. The thick or- (197 thopyroxenelamella in the bottomthird ofthe grainis thought Martignole and Schrijver 1, 1973) and Martignole to representan epitaxialovergrowth ofpigeonite (now inverted) (1979), however, presentedarguments in favor ofa deep- on the augitegrain. Note how the augitehost is depletedof seatedemplacement of anorthosite-charnockitesuites in orthopyroxenelamellae near the coarseorthopyroxene lamella. the Grenville province and the Adirondacks. Martignole In othergrains, epitaxial overgrowths such as this onehave bro- and Schrijver's(1971, 1973)and Martignole's(1979) ar- kendown to produce(100) "mesoperthite." guments, however, are largely based on garnet-bearing assemblages.Martignole (1979) left open the possibility that the garnet-bearingassemblages reflect metamorphic yses of host and lamellae with relative volumes deter- recrystallizationof a shallowly intruded igneousrock, but mined from point counts of photomicrographs.Approx- suggestedthat the preservationof ophitic texturesin maf- imately 1000points per grain wereused for this technique. ic charnockitesis most easily explained by intrusion dur- In one case(BA-5), the microprobe beam was widened ing high-grade regional metamorphism. More recently, to about 20 pm, and approximately 100 analyseswere Martignole(1986) has presenteda model that allows for collected on a number of traverses across the grain. In both the shallowemplacement of anorthositeand the deep this way a significant proportion of the grain was ana- emplacementof charnockitic rocks. lyzed.For samplePO-17, mineral analysesof arrgiteand Valley and O'Neil (1982) concluded,on the basis of orthopyroxenehost determinedby Jaffeet al. (1978)were O-isotope studies of the wollastonite deposit at Wills- used in conjunction with point-count data from photo- boro. New York. that anorthositewas intruded at shallow micrographs. The assumption that augite and orthopy- levels(< l0 km) and then was metamorphosedat higher roxene host and lamellae compositions are similar in this pressuresduring regional metamorphism. Valley (1985) rock was checkedat Vassar Collegeby means of energy- strengthenedthis argument with a combined phase-equi- dispersive analysesof host and lamellae on a scanning libria and O-isotope study of Cascadeslide, a monticellite electron microscope equipped with an energy-dispersive marble locality found in the Mount Marcy quadrangle. analyzer. Use of an sEMprovided advantagesin that a According to shallow-intrusion models, the orthofer- backscattered-electronimage allowed more preciseposi- rosilite origrnally described by Jaffe et al. (1978) from 266 OLLILA ET AL.: PYROXENE EXSOLUTION

Fig. 5. Coexistingaugite (right) and inverted pigeonite (left) from sample BA-5. The thick "001" augite lamellae in the inverted pigeonite and c of host augite are nearly vertical. The augite grain is intermediate between stages4 and 5 of Fig. 2, whereasthe inverted pigeonite correspondsto stage3. This differencein textural development is typical in theserocks. Relict igneoustextures are usually better preservedin inverted pigeonite than in host augite. pyroxene-bearing quartz syenite gneiss in the Mount Nain anorthositemassif or by Frost and Lindsley (1981) Marcy quadrangleis indicative only of minimum meta- from the Laramie anorthosite massif. SamplesAL-4, SC- morphic pressuresof approximately 8 kbar (Jaffe et al., 6, PO-17,and SR-29 (Bohlenand Essene,1978) are also 1978;Bohlen and Boettcher,1981). As describedabove, more Fe-rich than inverted pigeonitesdescribed by Smith however, re-examination of some of the specimenscol- (1974) from a portion of the Nain anorthosite massif that lectedby Jaffeet al. (1978)has shownthat Fe-richortho- crystallized at higher pressure(Betg, 1977) than the area pyroxene in these gneissesis, in fact, inverted pigeonite describedby Ranson(1986). Ranson (1986) reintegrated (Figs. 7 and 8). The composition of inverted pigeonite pyroxenesin a pyroxene-bearingquartz monzonite from from sample PO-17, when compared to the "forbidden a part of the Nain complex that crystallized at approxi- zone" boundariesin the solvi oflindsley (1983),suggests mately 3 to 4 kbar. This rock contains the assemblage that igneouscrystallization of this rock must have taken augite + invertedpigeonite + olivine + quartz. Pyroxene place at a minimum of approximately 9 kbar (Fig. 10). compositions from this rock (Fig. 10) are consistentwith Lindsley's(1983) "forbidden zone" boundariesand rep- DrscussroN resent the compositions at which low-Ca pyroxene is re- Valley(1985) has pointed out that thereis a high degree placed by olivine * quarrzat the pressure(3-4 kbar) and of uncertainty associatedwith the position of "forbidden temperatureat which this pyroxene-bearingquartz mon- zone" boundariesin Lindsley's (1983) phasediagrams zonite crystallized.The relatively common occurrenceof when applied to natural pyroxenesthat contain nonquad- inverted pigeonite more Fe rich than that found in sam- rilateralcomponents and has suggestedthat PO-17 could ple PL-187 (Fie. l0) in Adirondack syenitic rocks indi- have crystallized at pressuresas low as 3 kbar. It should catescrystallizatior' al pressureswell in excessof 3 kbar. be pointed out in this regard,however, that the argument If crystallization pressuresfor syenitic rocks are appli- in favor of deep crystallization for Adirondack pyroxene- cableto anorthosite,then the conclusionspresented above bearing quartz syenitesis not based solely on the com- conflict with the shallow-intrusion hypothesis of Valley positionof samplePO-I7. Although samplePO-I7 is the and O'Neil (1982)and with the modelsof Whitnev (1983) most Fe-rich sample yet studied, numerous other sam- and Mclelland (1986). Although it is possible,given the ples such as AL-4 and SC-6 show textures indicative of uncertainties of metamorphic geothermometry and of igneous pigeonite and are sufrciently Fe rich to require phase relationships for natural pyroxenes, that Fe-rich high-pressurecrystallization. Samples P-10, IN-11, and pigeonite could have been produced during regional SR-29 from the study of Bohlen and Essene(1978) like- metamorphism, this is not thought likely. In order to wise require crystallization pressuresin excessof 3 kbar. evaluate this possibility, phaserelations in the pyroxene These pyroxenescontain low percentagesof nonquadri- quadrilateral have to be examined. lateral componentsand are significantlymore Fe rich than Figure 1l is a schematicrepresentation-based on nat- inverted pigeonitesdescribed by Ranson (1986) from the ural assemblagesdescribed by Smith (1974) and the ex- OLLILA ET AL.: PYROXENE EXSOLUTION 267

Fig. 6. Invertedpigeonite from sampleAL-4. Coarse"001" Fig. 7. Photomicrographof intergrownaugite (top) and or- augitelamellae are oriented diagonallyfrom lowerleft to upper thopyroxeneOottom) from sampleSC-6. The coarseaugite la- right and c of host orthopyroxeneis vertical.This grain illus- mellain orthopyroxeneis interpretedas an epitaxialintergrowth tratesthe pigeoniteequivalent of stage4 in Fig. 2. ofaugiteand pigeonite (now inverted) and suggeststhat pigeonite wasthe initial low-Capyroxene to crystallizein this rock. perimental work of Huebner and Turnock (1980) and be metamorphosed fayalite-rich olivine granite. If one Turnock and Lindsley(198l)-of the phaserelations be- uses the l0-km maximum depth of intrusion suggested tween olivine (ol), quartz (q), orthopyroxene (opx), and by Valley and O'Neil (1982)for theserocks, then olivine clinopyroxene(cpx) at three different temperatures(Zl < should have replacedorthopyroxene in rocks with l00Fe/ T2 < T3). As can be seen in this diagram, Ca clearly (Fe + Mg) ratios of >75 (Bohlen and Boettcher,l98l) increasesthe stability of orthopyroxene over olivine for and should have replaced pigeonite with l00Fe/(Fe + the assemblageopx + ol + q. Ca, however,has a different Mg) ratios of > 78 at 825 "C (Lindsley and Grover, I 980). effect on the assemblageopx + cpx + ol + q. In this At higher temperatures,olivine should replacepigeonite assemblage,orthopyroxene is saturatedwith Ca, and any at more magnesiancompositions. increasein Ca content oforthopyroxene or pigeonite re- According to the shallow-intrusion model, low-Ca py- flects higher temperatures.These higher temperaturesin roxeneswith 100Fe/(Fe+ Mg) ratios of >78 should re- turn favor the stability of olivine + quafiz over ortho- cord metamorphic rather than igneous temperatures. pyroxene or pigeonite. These effects are evident in the Bohlen and Essene(1978) reintegratedcompositions of solvi of Lindsley (1983) in which the "forbidden zone" exsolved, Fe-rich, low-Ca pyroxenes in Adirondack sy- for pyroxenesexpands to more magnesiancompositions enitic gneissesand concluded that these rocks had ig- as the Ca content of either orthopyroxene or pigeonite neous precursorsthat crystallized between 900 and 1000 increases.This can be seen in Figure 10, which shows "C (Fig. l0). Any hypothesis of shallow intrusion, how- pyroxene compositions plotted on the l0-kbar solvus of ever, requires that these rocks were originally olivine- Lindsley(1983). bearing granitesand would require that the temperatures The shallow-intrusion model as stated by Whitney determined by Bohlen and Esseneare metamorphic. (l 983)requires that pyroxene-bearingquartz syenitegneiss The requirement that metamorphic temperatures be 268 OLLILA ET AL.:PYROXENE EXSOLUTION

Tner-e1. Electron-microprobeanalyses of pyroxenesand formulae based on six oxygens

9-'t7 AL-4 Sample: Cpx H Cpx L Opx H Opx L AugI Pig I Cpx H Cpx L Opx H OPx L Aug I Pig I Cpx H 48.28 sio, 50.38 s0.39 CU.UC 49.26 50.01 50.15 50.80 50.82 49.10 48.57 50.03 49.44 Alros 2 43 2.86 1.90 1 65 2.18 2.20 191 187 o.73 0.53 143 0.95 123 Tio, 015 0.23 0.07 0.03 0.11 0j2 o.21 o.24 0 09 0.12 0.18 ojz 0.23 CrrO. 0.00 l\4gO 1163 11.21 16.23 16.46 13.20 14.70 0.21 0.24 0.09 0.12 0.18 0.12 3.21 ZnO 0.09 FeO. 12.47 13.05 31.56 31.40 18.64 25.91 17.09 16.75 38 01 3776 2424 33.84 28.01 MnO 0.27 0.22 0.54 0.57 o37 0 44 0.20 0.18 0.45 0.45 0.29 0.40 0.58 21.65 21.15 0.39 0.60 14.79 6.72 20.40 20.79 0.52 0.62 1A RA 4.49 18.79 BaO 't01 Naro 0 58 0.53 0.04 0.03 0.40 0.19 105 0.01 0.06 0.68 0.21 0.68 Total 99.56 99.66 10078 100.00 99.70 100.43 100.44 100.61 100.40 99.97 100.28 100.43 101 .09 Si 1.925 1.923 1.937 1 926 1.925 1.932 1.957 1.954 1.971 1.961 1.958 1.968 1.949 AI 0 075 0.077 0 063 0074 0.075 0.068 0.043 0 046 0.029 0.025 0.042 0.032 0.051 Fe3* Total 2.000 2.000 2.000 2.000 2 000 2 000 2.000 2.000 2.000 2.000 2.000 2.000 2.000 AI 0 034 0.053 0 023 0 002 0.024 0.032 0.044 0.039 0.005 0.000 0.024 0.012 0.008 Ti 0 004 0 007 0.002 0.001 0.003 0.003 0.006 0 007 0.003 0.004 0.005 0 004 0.008 Cr 0.000 Pgs+** 0.075 0.050 0 039 0.072 0 074 0.042 0.062 0.072 0.019 0.023 0.058 0.029 0.083 Mg 0.662 0.638 0.936 0 960 0.757 0.844 0.507 0.511 0.687 0.714 0.576 0.651 0.193 Zn 0.003 Fe2* 0.323 0.367 0.982 0.955 0.526 0.793 0.488 0.467 1 256 1 252 0.735 1.097 0.862 Mn 0.009 0.007 0.018 0.019 0012 0014 0.007 0.006 0 015 0.015 0.010 0.013 0 020 0.886 0 865 0 016 0.025 0.610 0.277 0.842 0.857 o.022 0.027 0.569 0.191 0.813 Ba Na 0 043 0.039 0.003 0.002 0.030 0 014 0.07s 0.078 0 001 0 005 0.052 0.016 0.053 Total 2.036 2.026 2.019 2.036 2.036 2019 2.031 2.037 2 008 2.040 2.029 2.013 2.042 12.1 41.4 Wo.. 43.1 42 0 0 9 1.3 28.6 15.8 43.7 44.8 1.1 1.4 zu.5 '10.7 En 382 36.8 48.4 49.4 42.2 43.4 287 28.8 35.0 35.8 310 32.7 Fs 18.7 21.2 50.8 49.2 29.3 40.8 27.6 26.4 63.9 62.8 39.6 55.1 47.9 Notei H : host, L : lamellae,| : retntegratedbulk composition . Total Fe reoorted as FeO. "" Fe3*and Wo, En, and Fs calculatedaccording to the methodof Lindsley(1983). published et al. (1978). Lamellae t pigeonite compositions are based on opticai-reintegrationof exsolved grains and host compositions by Jatfe cornpoiitions *ere determinedby means of energy-dispeirsiveanalysis on a sianning electron microscope.These analysesare not as accurate as the wavelength-dispersiveanalyses [ublished by Jaffe-etal. out do establishthat there is no significantdifference between host and lamellaecompositions. greater than 900 oC for large areas of the Adirondacks Another argument against a metamorphic origin for leadsto a number of inconsistencies.One problem is with the Fe-rich pigeonite in these rocks is the nature of the the pyroxenesthemselves. If metamorphic temperatures exsolution textures. Pyroxenesfrom similar rocks in the were greater than 900 oC, pyroxenes with l00Fe/(Fe + Nain anorthosite massif (Ranson, 1981; Huntington, Mg) ratios of > 92 should be hypersolvus(Lindsley, I 983). 1980)and the Laramiemassif (Frost and Lindsley, 1981; This is clearly not the case in Fe-rich rocks from the Livi, 1987)and pyroxenesfrom many other plutonic suites Mount Marcy quadrangle(Fig. l0). A secondproblem is (Robinson et al., 1977) show exsolution textures similar that numerous mineral assemblagesand geothermome- to the relict textures observed in the Adirondacks. The ters that limit metamorphic temperaturesto lessthan ap- difference between Adirondack pyroxenes and unmeta- proximately 800 "C (Bohlenet al., 1985)are common in morphosed plutonic pyroxenes is in the breakdown of the areas where a 900 'C metamorphic temperature is high-temperature pigeonite or augite lamellae (stage 3, required by the shallow intrusion model. A third problem Fig. 2) to intergrowths of augite and orthopl'roxene (stages is that integated pyroxene compositions in these rocks 5 and 6, Fig. 2). Although it is diftcult to assessthe rel- indicate a range of temperatures.Although this is rela- ative importance of thermal-induced versus stress-in- tively easily explained by igneous processes,it is much duced changes,the breakdown is most easily explained more difficult to explain if the pyroxenesare the product as resulting from deformation associatedwith regional of regionalmetamorphism. A fourth problem is that there metamorphism becauseof the correlation between rock is no evidence of olivine pseudomorphs in any of the fabric and degreeof preservationof relict lamellae.Rocks rocks studied so far. Similar orthoferrosilite-bearingrocks with igneous textures or weak metamorphic fabrics lo- from the Lofoten Islands of Norway retain clear-cut evi- cally contain pyroxenes with relict igneous exsolution dencethat the orthoferrosilite replacedfayalite-rich oliv- textures,whereas pyroxenes from granoblasticrocks only ine (Ormaase4 1977).There is no textural evidencethat contain low-temperature, fine exsolution lamellae. Exso- this has taken place in the Adirondacks. lution textures in relatively undeformed rocks from the OLLILA ET AL.: PYROXENE EXSOLUTION 269

TABLE1-Continued

PO-171

Opx H Cpx H Cpx L Opx H Opx L Aug I Pig I Cpx H Cpx L Opx H Opx L Aug I Pig I Pig I 46.74 5036 50.51 48.68 49.18 49.92 49.31 48.69 49.72 45.50 44.67 47.86 45.97 46 18 0.25 2.04 2.06 0.92 0.86 1.91 1.31 0.65 0.91 0.36 0.36 0.83 0.45 0.50 0.09 0 27 0.26 0.09 0.06 0.18 0.15 0.06 0.015 0 11 012 0.13 0.11 0.12 000 0 06 0.06 0.06 0.02 0.06 0.06 0.06 395 9.2s I36 12.06 1249 10.15 1113 1.17 1.2'l 1.36 1.03 1.22 1.33 1.32 0.24 0.19 0 19 0.37 0.41 o.24 0.34 0.33 47.30 1587 16.03 JO.dC JO.OJ 22.05 29.71 29.18 2963 49.81 49.92 34 52 46.78 45 44 u-oz 17 0.18 0.50 0.47 0.32 0.39 0.43 0.30 1.10 1.38 0 60 1.00 0.96 o '15 0.69 20.64 20.11 0.87 0.44 18 7.47 19.07 19.s3 0.84 0.66 14.35 3.52 4.70 0.00 0.06 000 110 1 13 o.20 0 09 0.68 0 51 0.73 0.7'l 0.03 0.03 0.54 0 13 0.18 99.88 99.70 9964 100.17 100.22 100.39 99.98 100.65 10241 99.60 98.60 10037 9974 99.84 1.990 1.948 1.953 1.9s4 1.965 1.942 1.954 1 984 1.990 1.983 1 975 1.984 1.983 1 983 0.010 0.052 0.047 0.044 0.035 0 058 0.046 0.016 0.010 0.017 0.019 0.016 0.017 0.017 0 002 0.006 2 000 2.000 2 000 2.000 2 000 2.000 2.000 2.000 2.000 2 000 2.000 2.000 2.000 2.000 0 002 0.041 0.046 0.000 0.005 0.029 0.015 0 032 0 033 0.001 0.000 0 025 0.006 0.008 0.002 0.008 0.008 0.003 0.002 0.005 0.004 0.004 0.005 0.004 0.004 0.004 0.004 0.004 0.000 0 002 0.002 0.002 0 001 0.002 0.002 0.002 0.002 0.078 0.071 0 052 0.033 0 070 0.061 0.029 0.022 0.009 0.007 0.025 0.012 0.014 0 251 0.533 0.539 0.722 0 744 0.589 0 657 0.071 0.072 0.088 0.068 0.075 0 086 0.084 0 008 0.006 0.006 0.012 0.013 0.007 0.011 0.010 1.682 0.435 0.448 1.183 1.191 0.648 0.923 0.965 0.970 1.806 1.833 1j72 1.676 1.618 0.022 0.006 0.006 0.017 0.016 0.011 0.013 0 015 0.010 0.041 0.0s2 0.021 0.037 0.035 0.031 0.855 0.833 0.016 0.019 0.633 0.317 0 833 0.830 0.039 0.045 0.637 0.163 0.216 0.001 0.000 0.001 0.001 0 000 0.083 0 085 0.016 0.007 0.051 0 039 0.057 0.055 0 003 0.003 0.043 0.011 0.015 2.001 2 039 2.036 2.032 2.017 2.036 2.029 2.014 2.012 2.006 2.013 2.011 2.009 2.007 1.6 44.9 43.6 1.9 1.0 31.8 21.1 42.8 42I 2.O 1.7 32.7 12.0 15.1 12I 30.4 30.8 371 38.1 325 32.8 3.9 4.0 4.6 3.5 4.1 4.3 42 85.6 24I 25.6 609 60.9 46.1 53.3 53.2 93.4 94.8 63.2 83.7 80.7

Santanoni quadrangle most closely resemble exsolution imum pressureof 4.6 kbar is a conservativeestimate since textures in pyroxenes from unmetamorphosed igneous it assumesthat nonquadrilateral components are com- rocks. pletely partitioned into orthopyroxene,that inversion took 'C Regardlessof their 100Fe/(Fe + MC) ratios, all pyrox- placeat a temperature25 below the minimum stability enes that contain relict high-temperature lamellae have ofpigeonite (Lindsley, I 983), and that the orthopyroxene 'C. partially broken down to intergrowths of augite and or- contained 30/oCaSiO, at 800 Bohlen and Boettcher thopyroxeneoriented along (100) ofthe host grain. This have suggestedthat this is the maximum CaSiO, com- breakdown suggeststhat regional metamorphism took ponent for Fe-rich orthopyroxene at temperaturesbelow placeat temperaturesbelow the stability of pigeonite.This approximately 850 "C. A minimum crystallization pres- conclusion is entirely consistentwith metamorphic sureof approximately6.3 kbar for sampleAL-4 is a much geothermometry(Bohlen et al., 1985)and with pyroxene more reasonableestimate according to Lindsley's (1983) phaserelations (Lindsley, 1983). forbidden-zoneboundaries, but even ifone usesthe con- If initial exsolution and inversion took place during servative pressureestimate basedon orthopyroxene, the coolingofthese rocksfrom igneousconditions, as is con- minimum pressurerequired for the crystallizationof AL-4 sistentwith the histories of pyroxenesfrom both the Nain is well in excessof the 3-kbar maximum suggestedby and Laramie anorthositemassifs (Huntington, 1980;Frost Valley and O'Neil (1982)for anorthositicrocks at Wills- and Lindsley,1981; Ranson, 1986; Livi, 1987),then or- boro, New York. Likewise, the metamorphic pressurede- thopyroxene as well as pigeonite equilibria apply to the termined for sample PO-17 (approximately 8 kbar) by igneouspart ofthese rocks'histories.The orthopyroxene Jafe et al. (1978)and Bohlen and Boettcher(1981) also host in sample AL-4 (Fig. 6 and Table l) requires a min- representsa minimum igneous pressure if inverted pi- imum pressure in excessof 4.6 kbar if inversion took geonite in this rock inverted as it cooled from igneous place at approximately 800 'C, according to the geoba- conditions. rometer of Bohlen and Boettcher(1981). An ideal ionic The association of anorthosites with granitic and sy- model as describedby Bohlen et al. (198l) was used to enitic rocks is a worldwide phenomenon, and although correct for comoonentsother than Ms and Mn. The min- geochemicalevidence has ruled out a comagmatic origin 2',10 OLLILA ET AL.: PYROXENE EXSOLUTION

o lamellae . host, from Jaffe et al. ( 1978) r opticEllgreintegrEted bulkcomposition " SEMreintegrEted bulkcomposition

t Fs Fig.9. Energy-dispersiveanalyses ofexsolution lamellae and invertedpigeonite in samplePO-17 plotted on the Fe-richpart of the pyroxenequadrilateral. Analyses were obtained by useof a Kevexquantitative match routine using host compositions from Jaffeet al. (1978)as standards. Pyroxenes were analyzed for Ca, Mg, Fe, and Si and are plottedas molo/oWo [Cal(Ca+ Mg + Fe)1,Fs [Fe/(Ca+ Ms + Fe)],and En [Mgl(Ca+ Mg + Fe)]. Fig. 8. Photomicrographof invertedpigeonite from sample PO-17.Orthopyroxene host is at partialextinction, and (100) augitelamellae are vertical. This grain is thepigeonite equivalent of Valley and O'Neil (1982) and Valley (1985), the as- of stage6 in Fig. 2. No traceof originalhigh-temperature la- sumption that pyroxene-bearingquartz syenitesintruded mellaeare visible, and augite lamellae have coalesced and coars- at the same time as anorthosite may have to be re-ex- ened.This grainwas chosen for opticalreintegration because it amined in more detail. is not in contactwith an augitegrain. Bulk compositlonswere There is some isotopic evidenceto support the conten- determinedboth includingand not includingthe largerareas of tion that pyroxene-bearingquartz syeniteswere intruded augite.Both compositionsare listedin Table I and arewithin at a differenttime than anorthosite.Silver (1969),on the the pigeoniterange. Optical reintegration values were checked basis of a U-Pb study of zircons from Adirondack sy- by meansof energy-dispersiveanalysis on a scanningelectron rocks had an igneous microscope.Four areasthat were as largeas possiblewhile enitic rocks, concluded that these avoidingcracks were scanned. Results are plotted in Fig. 9. crystallizationage of approximately 1130 Ma. In contrast,Ashwal and Wooden(1983) concluded, on the basis of a Nd-Sm study, that anorthosite in the Mount Marcy for these rocks, it is commonly thought that they are co- massif in the Adirondacks has a crystallization age of geneticor at leastroughly coeval (Emslie, 1978;Morse, approximately1288 Ma. Critical evaluationof theseages, l98l; Philpotts, 198l; Mclelland, 1986).Furthermore, basedupon differentdecay systems, must await more data, it has commonly been assumedin the Adirondacks that but it at least appearspossible that anorthositic and sy- the gradational contactsbetween anorthositic and syenit- enitic rocks were intruded at different times and under ic rocks (de Waard and Romey, 1969;Davis, 1971;Jaffe different conditions. A better understanding of the rela- et al., 1983)imply that theserocks were magmasat the tionship between anorthositic and syenitic rocks awaits same time and that mixing occurred between the two more detailed isotopic studies, a more complete under- rock types.This sort of relationship has beendocumented standing of the wollastonite deposit at Willsboro, New for the Nain anorthositemassif by Wiebe (1980),and if York, and an understandingof the relationship between such is the casefor the Adirondacks, then crystallization the anorthosite found at Willsboro and that found in the pressuresfor pyroxene-bearingquartz syenitesare appli- Marcy massif. If, however, conclusionsregarding shallow cableto anorthosite.Given the discrepancy,however, be- emplacementof anorthosite and the deep emplacement tween the conclusionsDresented here and the conclusions of syenitic rocks are both correct, then a model such as OLLILA ET AL.: PYROXENE EXSOLUTION 271

I O kbar mole 5 2es-rroo:10:o:g

En 25 50 75 tat Fig. 10. Compositions ofpyroxenes from anorthositic and syenitic rocks from the Adirondacks and Labrador (PL-187) plotted on the 10-kbarsolvus ofLindsley (1983).P-10 is from Bohlenand Essene(1978) and PL-187 is from Ranson(1986). The dashed line is the l0-kbar boundary ofthe "forbidden zone" (Lindsley, 1983).Pyroxene compositions to the right ofthis line are metastable with respectto the assemblageaugite + olivine + q:uartz.The 5-kbar "forbidden-zone" boundary (dotted line) is also shown. Reintegratedcompositions of exsolvedgrains are plotted exceptfor sampleAL-4. Alteration made reintegrationof inverted pigeonite impossible in sampleAL-4; however, analysesof host augite and orthopyroxenewere obtained. Exsolution textures (Fig. 8) clearly indicate that this samplecontains inverted pigeonite.Sample PO- I 7 suggeststhat crystallization of theserocks took placeat pressures greaterthan or equal to approximately 9 kbar. presentedby Martignole(1986) in which anorthositein- trusion is followed almost immediately by crustal thick- ening and in which syenitesare intruded into thickened crust provides at least one solution to this problem.

CoNcr,usroNs The complex exsolution textures described above are En interpreted to result from the combined effectsof cooling from igneous conditions and regional metamorphism. Adirondack pyroxenesdiffer from typical igneouspyrox- enesin that relict coarseexsolution lamellae have broken down to intergrowths of augite and orthopyroxene oriented along (100) of the host grain. Reintegrated compositions of magnesianto moderately Fe-rich pyroxenes (samples9-17, GB-2, and BA-5) showingrelict igneous exsolution textures record temperatures well above regional metamorphic temperaturesand confirm that these are igneous rather than metamorphic pyroxenes.We interpret the mobilization of exsolution lamellae to be causedby regional metamorphism at temperaturesbelow

-J

Fig. I l. Schematicphase relations among pyroxenes,and o1- ivine * quartz (ol) at different temperatures(Tl < T2 < T3) and constant pressure.The three-phasetriangle opx-aug-ol (Zl) moves to the left, and the Ca content of opx increasesas temperature increases.At 72, pigeonite becomesstable and forms via the reaction augite + orthopyroxene * olivine + qtf,artz: tent stabilizesboth orthopyroxeneand pigeoniterelative to oliv- pigeonite.As temperaturefurther increases,the three-phasetri- ine + quartz when the assemblageis not saturatedwith respect angle pig-aug-ol moves to the left, and the gap in Ca content to augite. If the assemblageis saturated with augite, then in- between pigeonite and augite narrows (73). These trends are creasedCa content of either orthopyroxeneor pigeonite reflects evident in the boundaries ofthe "forbidden zones" in the solvi an increasein temperature that in turn stabilizes olivine + quartz ofLindsley (1983)and Fig. l0 ofthis paper.Increased Ca con- relative to low-Ca pyroxene. 272 OLLILA ET AL.: PYROXENE EXSOLUTION those necessaryto stabilize pigeonite during the Grenvil- application of olivine-quartz-orthopyroxene barometry. Earth and lian metamorphic event. This conclusion is basedon the PlanetaryScience Letters, 47, l-10. Bohlen, Valley, W and Essene,E.J. (1985) Metamorphism in the differencesbetween Adirondack pyroxenes S.R , J , and typical ig- Adirondacks I. Petrology, pressureand temperature.Joumal of Pe- neouspyroxenes, the ubiquitous occurrenceof (100) in- trology,26, 971-992 tergrowths ofaugite and orthopyroxene regardlessofthe Buddington,A.F. (1939)Adirondack igneous rocks and their metamor- Fe/Mg ratio of the pyroxene,and the correlation between phism GeologicalSociely of America Memoir 7, 354 p -(1969) New York StateMuseum the degreeofpreservation ofrelict igneousexsolution tex- Adirondackanorthositic series and ScienceService Memoir 18,215-231. tures and metamorphic fabric in the host rock. Davis, B.T.C. (1971)Bedrock geology ofthe St Regisquadrangle, New Fe-rich inverted pigeonite from sample AL-4 contains York. New York State Museum and ScienceService Map and Chan coarse"001" augite lamellae that are most reasonably Series16, 34 p. interpreted as resulting from cooling from igneous con- de Waard,D , and Romey,W D. (1969)Pelrogenetic relationships in the ditions, and these lamellae have partially anorthosite-chamockiteseries of Snowy Mountain dome, south-central broken down Adirondacks New York State Museum and ScienceService Memoir to intergrowths of augite and orthopyroxene.Low-Ca py- l 8, 307-3l 5 roxenefrom samplePO-I7 has a bulk compositionthat Emslie,R.F. (1978)Anorthosite massifs, rapakivi granites,and late Pro- clearly indicates original crystallization as pigeonite even terozoic rifting of North America. PrecambrianResearch, 7 , 6l-98. though low-Ca pyroxenesin this rock now consist ofin- Frost,R, and Lindsley,D H (1981)Crystallization conditions offerro- syeniteassociated with the l,aramie anorthosite,Wyoming. Geological tergrowths of orthopyroxene and augite with augite la- Societyof AmericaAbstracts with Programs,13, 455. mellaeoriented along (100) ofthe host.The compositions Hess,H H (1941)Pyroxenes of common mafic magmas Par-t2 Ameri- of these pyroxenes, as well as pyroxenes described by can Mineralogist,26, 573-594. Bohlen and Essene(1978), require crystallizationpres- -(1960) Stillwaterigneous complex, Montana, a quantitativemin- eralogicalstudy GeologicalSociety of America Memoir 80, 230 p. sures well in excessof 3 kbar, and the composition of Huebner,J S, and Tumock, A.C (1980)The meltingrelations at 1 bar low-Ca pyroxenein samplePO-17 (Fig. l0 and Table l) of pyroxenescomposed largely of Ca-, Mg-, and Fe-bearingcompo- suggestsa minimum igneous crystallization pressure of nents.American Mineralogist, 65, 225-2'71. greater than approximately 9 kbar. Fe-rich igneous py- Huntington, H.D. (1980) Anorthositic and related rocks from Nukasor- roxenesindicate that models requiring shallow emplace- suktokh Island, Labrador. Ph.D. thesis, University of Massachusetts, Amherst, Massachusetts ment of Adirondack anorthosite cannot be applied to the Jaffe,H.W., Robinson,Peter, Tracy, RJ., and Ross,Malcolm. (1975) syenitic rocks that surround the Marcy anorthosite mas- Orientation of pigeonite exsolution lamellae in metamorphic augite; sif. correlation with composition and calculatedoptimal phaseboundaries. AmericanMineralogist, 60, 9-28. JaFe,H W , Robinson, Peter, and Tracy, R.J. (1978) Orthoferrosilite and AcxNowr,BlcMENTS other iron-rich pyroxenesin microperthite gneissof the Mount Marcy area,Adirondack Mountains AmericanMineralogist, 63, I I l6-1136. This paper, in part, is derived from Paul Ollila's Ph.D researchat the Jaffe,H W., Jafe, E.B.,Ollila, P.W., and Hall, L.M. (1983)Bedrock ge- University of MassachusettsThe authors would like to thank David ology of the High Peaksregion, Marcy massii Adirondacks,New York. Leonard of the University of Massachuseltsand Jerry Calvin of Vassar Contribution 46, Department of Geology and Geography,University Collegefor their assistancewith electron microprobe and sprra-roaanal- of Massachusetts,Amherst, Massachusetts, 78 p ysis An early version of this manuscript was reviewed by William Ran- Lindsley, D H (1983) Pyroxene thermometry. American Mineralogist, son, and his comments and suggestionswere most helpful. A critical re- 68,477-493 view by JacquesMartignole also significantlyimproved the quality ofthis Lindsley,D.H., and Grover, (1980)Fe-rich pigeonite: A geobarom- paper. J.E. eter.Geological Society ofAmerica Abstractswith Programs,12,472. Livi, K.J.T. (1987)Geothermometry of exsolvedaugites from the Lara- RsrrnpNcns crrno mie anorthositecomplex, Wyoming. Contnbutions to Mineralogy and Petrology,96, 37 l-380. Albee, A.L., and Ray, L. (1970) Correction factors for electron micro- Martignole, J. (1979) Charnockite genesisand the Proterozoiccrust. Pre- probe microanalysis of silicates,oxides, carbonates,phosphates, and cambrianResearch. 9. 303-310 sulfatesAnalytical Chemistry, 42, 1408-1414. -(1986) Someproblems on crustalthickening in the centralpart of Ashwal, L.D. (1982) Mineralogy of mafic and Fe-Ti oxide rich differen- the Grenville Province GeologicalAssociation ofCanada SpecialPa- tiatesof the Marcy anorthositemassif, Adirondacks, New York Amer- per31. icanMineralogist, 67, | 4-21. Martignole,J, and Schrijver,K. (1971)Association of(hornblende)-gar- Ashwal,L.D., and Wooden,J.L (1983)Sr and Nd isoropegeochronology, net-clinopyroxene"subfacies" of metamorphismand anorthositemasses. geologichistory, and origin ofthe Adirondack anorthosite.Geochimica CanadianJournal ofEarth Sciences,8. 698-704. et CosmochimicaActa, 47, I 875-1887. - (1973) Effect of rock composition on appearanceof gamet in an- Bence,A E., and Albee, A.L. (1968) Empirical correction factors for the orthosite-charnockitesuites. Canadian Journal of Earth Sciences.10, electron microprobe analysisof silicatesand oxides Journal of Geol- tt32-1139 ogy,16,382403. Mclelland, J.M. (1986)Pre-Grenvillian history of the Adirondacksas an Berg,J.H. (1977)Regional geobarometry in the contactaureoles of the anorogenic,bimodal calderacomplex of mid-Proterozoicage. Geology, anorthositic Nain complex, Labrador Journal of Petrology, 18, 399- t4,229-233 430 Morse, S A. (l 98 l) A partisan review of Proterozoicanorthosites. Amer- Bohlen, S.R., and Boettcher,A L (l 98 1) Experimentalinvestigations and ican Mineralogist,67, 1087-l 100 geological applications of orthopyroxene geobarometry. American Ollila, Pw. (1984)Bedrock geology ofthe Santanoniquadrangle, New Mineralogist,66, 951-964. York. Ph D. thesis,University of Massachusetts,Amherst, Massachu- Bohlen,S.R., and Essene,E.J. (1978) Igneous pyroxenes from metamor- setts phosed anorthosite massifs Contributions to Mineralogy and Petrol- Ollila, P.W., Jafle,E.B., and Jaffe,H.W. (1983)Pyroxene exsolution his- ogy,65,433442. tory in the Adirondack anorthositemassif. Geological Society of Amer- Bohlen,S R, Essene,E J, and Boettcher,A.L. ( I 98I ) Reinvestigationand ica Abstractswith Programs,14,69-70. OLLILA ET AL.: PYROXENE EXSOLUTION 273

Ollila, P.W, Jaffe,H.W., and Jaffe,E.B. (1984) Iron-rich inverted pi- Valley,J W, andO'Neil, J.R. (1982) Oxygen isotopic evidence for shal- geonite:Evidence for the deep emplacementof the Adirondack anor- low emplacementofAdirondack anorthosite Nature, 300, 497-500. thosite massif.Geological Society of America Abstractswith Programs, Whitney,P.R. (1983) A three-stagemodel for the tectonichistory of the 15.54 Adirondackregion, New York. Northeastern Geology, 5,6l-72. Ormaasen, D E (1977) Petrology of the Hopen mangerite-charnockite Wiebe,R.A. (1980)Comingling of contrastedmagmas in the plutonic intrusion,Lofoten, north Nonvay Lithos, 10, 291-310. environment.Journal of Geology,88, 197-209. Philpotts,A.R. (1981)A model for the generationof massif-typeanor- thosites Canadian Mineralogist, 19, 233-254. M.a.NuscnrprRECETvED Ocrosen 13, 1986 Ranson,WA. (1981)Anorthosites of diversemagma types in the Put- M.nNuscprmAccEmED Noveunen 20. 1987 tusaluk Lake area,Nain complex, Labrador CanadianJournal ofEarth Sciences,18,2641. -(1986) Complex exsolution in inverted pigeonite: Exsolution AppnNorx 1. S,tnnpm DESCRIPTIoNS mechanismsand temperaturesof crystallizationand exsolution.Amer- 9-17. Coarse-grainedsubophitic gabbroic anorthosite consist- ican MineralogisL,T l, 1322-1336 ing of plagioclase(An46) 800/0, augite I 6010,inverted pigeonite I 0/0, Rietmeijer,F J.M. (1979)Pyroxenes from iron-rich igneous rocksin Ro- and ilmenite 30/0.Sample was collectednear the summit of Don- galand,S.W. Norway Ph D. thesis,University of Utrecht, Netherlands. Mountain, Santanoniquadrangle. Robinson,Peter. (1980) The compositionspace ofterrestrial pyroxenes- aldson pyroxene-plagioclase-quartz- Intemal and extemal limits Mineralogical Societyof America Reviews AI-4. Medium- to coarse-grained, in Mineralogy,7, 419-476 microperthitegneiss consisting of qttartz 160/0,microperthite 610/0, Robinson,Peter, Jaffe, HW., Ross,M., and Klein, Cornelis,Jr (1971) oligoclase 150/0,augite 2.50/0,inverted pigeonite l.4o/o, hom- Orientation of exsolution lamellae in clinopyroxene and clinoamphi- blende 3.00/o,gamet0.3o/o, opaques 0.30/0, apatite O.4o/o,and zir- boles: Consideration of optimal phase boundaries.American Miner- con 0.10/0.Sample was collected in the northeastern corner of alogist,56, 909-939. the Santanoni quadrangle on the top of a hill I km south of Robinson,Peter, Ross, M., Nord, G L, Jr., Smyth,J R, and Jaffe,H W CamerasPond. (1977)Exsolution lamellae in augite pigeonite: and Fossilindicators of BA-5. Fine-grained pyroxene monzonite granulite consisting lattice parametersat high temperature and pressure American Min- microperthite 190/o,plagioclase 25010, inverted pigeonite 130/0, eralogist,62, 857-873. of garnet Silver, LT (1969) A geochronologicinvestigation of the Adirondack avgite l4o/0, 2jo/o,apalite 4o/0,and opaques50/0. Medium complex, Adirondack Mountains, New York. New York StateMuseum to coarsegrains ofplagioclase and pyroxene,which retain relict and ScienceService Memoir 18,233-251 igneoustextures, are enclosedin a fine-grainedgranoblastic ma- Smith, D. (1974)Pyroxene-olivine-quartz assemblages in rocks associated trix consisting of microperthite, plagioclase,and garnet. The with the Nain anorthositemassif, Labrador JournalofPetrology, 15, sample was collected on Basin Mountain in the Mount Marcy 58-78 quadrangle. Turnock,A.C., and Lindsley,D.H. (1981)Experimenral determination of GB-2. Medium- to fine-grainedgabbro ganulite consistingof pyroxenesolvi for P < I kb, 900 and 1000'C. CanadianMineralogist, plagioclase(Anru to Anrr) 33ol0,orthopyroxene (includes inverted t9, 255-261. pigeonite) 150/0,augite l9o/0,gamet 160/0,ilmenite 110/0,magne- Valley,J.W (1985)Polymetamorphism in the Adirondacks:Wollastonite granoblastic, at contactsofshallowly intrudedanorthosite. In A.C. Tobi and J L R tite 2o/o,and apatite 4010.Pyroxenes occur both as Touret, Eds, The deep Proterozoic crust in the North Atlantic prov- fine-grainedaggregates and as coarser(up to 2 mm) grains that inces.D Reidel Publishing Company, Dordrecht, Netherlands NATO retain relict igneous exsolution textures. The sample was col- AdvancedScience Institute Series. ser c. 158.217-236 lected in Guideboard Brook in the Mount Marcy quadrangle.