sign in sign up
<<
Home , Hornfels

American Mineralogist, Volume 74, pages 53U548, 1989

Isochemical metamorphismof mafic , Laramie Anorthosite Complex' Wyoming: Amphibole compositionsand reactions Clnor, Russ-N.lnnr-pr* Department of Earth and SpaceSciences, State University of New York at Stony Brook, Stony Brook, New York I 1794, U.S.A.

Ansrnrcr Contact by anorthosite and syenite of the Laramie Anorthosite Com- plex, Wyoming, resultedin a thermal overprint of the regionally metamorphosedamphib- olite-facies country rocks at approximately 3.0-kbar pressure.The resulting assemblages in the mafic schist are amp + plg + ilm + qtz t cpx at the lowest-gradehornblende facies,amp + plg + cpx + opx * ilm + t11 + qtz at the orthopyroxene hornfels facies boundary, and amp + plg + cpx + opx * olv * ilm * mt at the highest grade, the olivine hornfels subfacies.Clinopyroxene first appearedat temperaturesof >670 "C, form- ing concomitantly with from reaction of with .Orthopyrox- ene first resulted from reaction of quartz with hornblende between >680 and 820 "C and above 730 'C from dehydration of hornblende where quartz was not available. Quartz appearsto have persistedup to 890 "C in some bulk compositions. Olivine first appeared at907 { from the reaction ofhornblende with plagioclase.The highest grade ofcontact metamorphism preserved in the mafic schist at Morton Pass was the olivine hornfels subfacies,which recorded temperaturesup to 933 "C. The dominant effect of isochemical metamorphism of these schistswas the changein abundanceand composition of hornblende with concomitant modal increaseor decrease in the other phases.Discontinuous reactions break amphibole down to and plagioclase(and eventually olivine) at facies and subfaciesboundaries, while continuous reactions allow for the progressiveenrichment of K, N4 A| Ti, F, and Cl as the quadri- lateral amphibole component, (Ca,Mg,Fe)rSirOrr(OH)r,becomes increasingly unavailable. There appearsto be a linear relationship between AIt" and temperature, strggestingthat systematictrends of other elementswith AlI" also reflect changesin temperature.Fdenite (NaAtSi-,tr-,) and tschermakite (AlrMg-,Si-,) components best describe the enrichment of Na and Al,u in amphibole with increasingtemperature. There are no significantchanges in the compositions of the other phaseswith grade. It is shown that progressivemeta- morphism as exemplified by the various hornfels of this study, is indicated by progressive loss of water with increasingtemperature without lossof Cl or F. Becausean isochemical suite of rocks is available that spansa wide temperature range, reaction spaceis used to provide an effective analysisof the progressof metamorphism in the various hornfels. One effect of reaction-spacemodeling is the derivation of reactions specificto the formation of the faciesand subfaciesat Morton Pass.Although the following reactions correspond to particular samples,they exemplify the incremental changesfrom one grade to the next. The reaction 0.203amp,+ 0.193p1g:0.l55ampr + l.l35qtz + 0.063cpx+ 0.048H,O, describesthe initial development of clinoplroxene and was followed, at higher tempera- tures, by the reaction 0.646amp,+ 0.l0lqtz + 0.030p1g: 0.606ampo+ 0.227cpx+ 0.040H,O, which consumed quartz rather than produced it. This reaction operatedover a wide tem- perature range in the hornblende hornfels facies. The initial orthopyroxene-forming re- action l.480ampo"+ 4qtz : 0.72}amgs"* l.465cpx+ 0.51lopx * l.703plg+ 0.760HrO,

* Presentaddress: Department of Geology, University of Missouri, Columbia, Columbia, Missouri 65203' U.S.A. 0003-004x/89/05064530$02.00 530 RUSS-NABELEK:AMPHIBOLE REACTIONS IN MAFIC SCHIST 531

was the major quartz-consuming reaction. After quartz reacted out, or in hornfels that were not quartz-saturated,orthopyroxene formed by a similar reaction l.480ampo:0.720amp,+ l.465cpx+ 0.51lopx* 0.703p1g+ 0.760H,O. At the highest grade, olivine formed by the reaction 0.677amp,+ 0.249p19: 0.413amp,+ 0.985cpx+ 0.27lopx + 0.l32fo + 0.264H,O, which continued within the olivine hornfels subfaciesuntil the ultimate demise of am- phibole above 940 "C.

INrnooucrroN areal extent of more than 1000 km2 in the Laramie Previousstudies (e.g., Engel and Engel, 1962a,1962b1, Mountains (Fowler, 1930;Newhouse and Hagner, 1957; Harte and Graham, 1975;Iaird, 1980;Laird and Albee, Klugman, 1966). 1981a,l98lb; Abbott, 1982;Robinson, 1982a, 1982b; A particularly good exposure of the high-grade rocks (Fig. Stoddard, 1985)that have dealt with metamorphismof of the cohtact aureoleis in the Morton Passarea l). amphiboles in mafic systemshave provided valuable in- Archean felsic gneissescrop out in the western and cen- formation on the history and role of intensive variables tral regions of this study area. The eastern edge of the during the metamorphism of specific terranes. For ex- sequenceis in concordant contact with a supra- ample, Laird (1980) described systematic amphibole crustal unit (Fig. l) whose relative ageis unclear but that (Graff compositional changesand variations in proportions of is inferred to be younger than the et al., group coexisting phasesduring low-temperature, high-pressure 1982).A metasedimentaryunit of the supracrustal metamorphism of mafic schistsin Vermont. is composed in the north of thin layers of impure - The Morton Pass area of the Laramie Anorthosite stone and calc-silicate that rapidly broaden southward q\artz- Complex, Wyoming, provides an excellent example of into intercalated layers of , calc-silicate, the effects of low-pressure,high-temperature metamor- ite, and pelite. A volcanoclasticunit forms a continuous phism of mafic and therefore complements the 100-m-wide belt and is interlayered with the metasedi- study of Laird (1980).The Morton Passarea is especially mentary unit. This volcanic sequenceis composed of suitable for evaluating mafic schist phase relations be- massive and banded hornfels of mafic and ultramafic causethe last metamorphism occurred in responseto an composrtlons. isobaric thermal overprint by the anorthosite intrusion; The Archean units have been discordantly cut on the therefore, variations in phases and their compositions east by the Laramie Anorthosite Complex, an intrusive primarily expresschanges in temperature. Becauseiso- sequencebeginning with anorthosite including gabbroic chemical suites of the mafic rocks are available, the re- and noritic varieties and concluding the monzonite and actions that occurredat low pressureand medium to very monzosyenite(Fuhrman et al., 1983; Fuhrman, 1986; high temperaturescan be identified. Of particular impor- Fuhrman et al., 1988).The monzonite-monzosyenitese- tance with regard to these types is the presenceof ries appearsto have intruded along the contact between one or two pyroxenes,which, through the application of the anorthosite and the Archean rocks, and anorthosite pyroxenethermometry, indicatethe temperaturesat which is suspectedto lie beneath the younger syenites.Grani- the continuous and discontinuous reactions occur. toid bodies are ubiquitous in the Morton Passarea, oc- The approach to charcterization of the Morton Pass curring perhapsas late-stagedifferentiates ofpossibly lo- hornfels is (l) to identify an isochemical suite of rocks cal partial melts. spanningthe temperature range of metamorphism, (2) to It is evident from relict texturesand assemblagesin the identify the changesin mode and composition lowest-gradecountry rocks that before the anorthosite- from low to high grade, and (3) to derive reactions de- syenite intrusions, the rocks of this area have been de- scribing the formation of facies and subfacies.Reaction formed at least once and regionally metamorphosed to space,following the procedureof Thompsonet al. (1982), greenschistor facies.The mafic schistsin the is used to characterizethe reactions. The result of this volcanoclastic unit at the contact were then metamor- approach is a better understanding of the amphibolite- phosed by the intrusives to the hornblende hornfels and to- facies transition at low pressuresin mafic hornfels facies.Those ofthe hornblende horn- rocks. fels facies are interlayered with massive to finely lami- nated leucocraticbands. Gradations between the two types Gnor,ocy oF THE coNTAcr AUREoLE AT occur locally. Laminations in the massive rocks are in- MonroN Pass terpretedto have formed by local shearingin the anaphib- The Laramie Anorthosite Complex cores the eastern- olites during deformation accompanying regional meta- most of the three prongs of the Rocky Mountain Front morphism and are thereforerelict features.These texturel Range that extend northward into southern Wyoming. were partially to completely overprinted with increasing The complex and its contact aureole crop out over an grade. Layering of the nonhomogeneousrocks is inter- 532 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST

Fig. l. Geologic map of the Morton Passarea, Albany County, Wyoming. Sample numbers referring to the letter locations are thefollowing:A:MPll-1andMPll-2A,B:83-17A,C:MPl2,D:MPl5-4,E:MPl8-1,F:83-10,G:83-11,H:83- l9A,I:83-128,J:83-13,K:MPl-3AandMP1-3B,L:83-14,M:83-18,N:MP6-4,O:83-15,P:83-16,Q:MP3-4, R : MP3-5 and MP3-6. S : MP57-9, and T : MP57-2. preted as a primary sedimentary feature, suggstingthat were counted separately and their weighted averages these were originally volcanoclasticsediments. combined. The modes are given in Table I along with The finely laminated mafic schists,the subject of this the uncertainties estimated by the technique of van der paper, occur as bands 4 cm to 5 m wide. Within these Plas and Tobi (1965). Figure 2 illustratesthe relative units, the degreeof contact metamorphism increasesto abundancesof the phases in terms of volume percent. the east toward the monzosyenite. Five samples-MPl I - l, 83-I 28 (hornblendehornfels), 83-15 (orthopyroxenehornfels), and MP3-6 and MP3-5 InnNrrrrcl.rroN oF AN rsocHEMrcAL (olivine hornfels)-are essentially isochemical (Table 2) ASSEMBLAGE SERIES and are called WR I. Similarly, two samples-83-l9A The mafic schists of Morton Pass comprise a variety (hornblende hornfels) and 83-16 (pyroxene hornfels), of bulk compositions. Two particular units were identi- constitutingWR Il-are also isochemical.Olivine horn- fied on stratigraphicgrounds; eachis effectivelyisochem- fels sample MP3-5 was also analyzedby inductively-cou- ical from low to high grade. These are referred to as pled-plasmaatomic-emission spectroscopy(rcr-ars). The "whole-rock I" and "whole-rock II" (WR I and WR II). measuredcomposition is nearly identical to the calculat- Whole-rock compositions were estimated by means of ed one (Table 2). The compositionsof theseseven sam- modal recombination. This method requires accurate ples are comparable with oceanic tholeiites. The most modes as well as analysesof each phase. On each ho- compelling differencesbetween WR I and WR II are that mogeneousthin section, 1200 point counts were made. WR I contains quartz in at least two of the lower-grade On nonhomogeneousthin sections, the diferent zones samples and has clinopyroxene in all samples. In con- RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST 533 trast, WR II has no quartz or clinopyroxene in the horn- s 70 blende hornfels facies. In accordancewith field evidence lEl 60 for a sedimentary origin, it is suggestedthat prior to metamorphism, these rocks were tuffs. 50 rl PHaSn ASSEMBLAGES 40 The major efect of contact metamorphism on Morton 30 Passamphibolites was the changein the abundanceand El composition of hornblende with concomitant changesin z 20 the proportions of other phases(Fig. 2). In general,am- l0 phibole abundancedecreases along with quartz,while the z proportions of plagioclase, clinopyroxene, orthopyrox- 0 t------o.'-""""""8o-""'--"r"'r ene. oxides. and olivine increase.The hornfels of WR II -10 differ from those of WR I by their greater abundanceof Fl 4 amphibole and lack of quartz and clinopyroxene at low ]------' -' -' - l;' grades.Mineral facies in mafic schist of Morton Passare -l representedby the following assemblages:hbl + plg + > 650 750 Es0 950 ilm + cpx + qtz (hornblende hornfels facies)and opx + TEMPERATURE("C) cpx + hbl + plg + ilm + mag + qtz + olv (orthopyrox- ene hornfels facies). Interlayered pelitic rocks are just Fig. 2. Diagramshowing volume percentof eachphase in pyroxene above the second isograd and contain WR I andWR II vs. temperatureestimates from com- positions.Error bars are<2o/o for volumeand +50 'C for tem- * cordierite + sillimanite + K-feldspar * plg + qtz + peratureestimates. with rare and biotite + + + K-feldspar + plg + opx + qtz + spinel (Bo- chensky, 1982; J. A. Grant and B. R. Frost, unpublished manuscript, 1988).For the most part, thereis very little ANAr,vrrcll TECHNIeUES chemical variation in the phases of the mafic hornfels The describedin this study were analyzedus- except for amphibole, although increasing MgCa-, ex- ing the former ,lnr- rux-su electron microprobe at SUNY- change component in clinopyroxene reflects increasing Stony Brook. Several analysesof each phase were col- temperature. In fact, the "chemical evolution" of am- lected per thin section,and grains were routinely checked phibole can characterizethe prograde metamorphism of for zoning.Approximately 400 amphibole,150 pyroxene, the mafic schist. To understand the effects of contact 100plagioclase, 300 Fe-Ti oxide,and l0 olivine analyses metamorphism on the mafic schist, it is helpful to know were collected. An acceleratingpotential of 15 kV and a accurately the temperatures at which the changestook specimencurrent of I 5 nA (referencedto brass)were used place. for all analysesexcept F and Cl, for which a specimen

TABLE1. Modes in volumepercent of phasesin Morton Pass hornfels

Loc in Fig. 1 Sampleno. Amp cpx opx Plg Apatite Hornblendehornfels MP11.2A 7.18 Jt -z 35.4 0.15 tr A MP11-1 3.84 49.8 6.98 38.9 0.50 tr 83-13 47.2 13.1 38.9 0.11 0.80 tr (tr) 83-12B 0.13 46.7 15.00 37.7 0.15 0.39 tr E MP18-1 IT 54.2 7.14 37.5 1.16 Ir- H 83-19A 61.6 38.2 0.08 IT E 83-17A 0.15 31.9 26.2 41.'l 0.61 tr Orthopyroxene hornfels N MP6.4 n 1( 2.49 34.2 8.76 53.0 1.36 T MP57-2 49.3 12.0 1.12 36.0 0.48 1.04 o MP3-4 0.16 32.6 25.5 3.69 36.9 0.96 D MP57-9 31.7 20.4 4.93 40.2 0.79 1.90 Sulfide (tr) P 83-16 48.3 9.90 1.95 39.6 0.30 o 83-15 22.8 19.8 6.90 49.2 090 0.41 IT Olivinehorntels R MP3.6 13.9 29.9 6.46 47.1 0.49 1.88 0.25 tr Sulfide (tr) R MP3-5 9.23 35.9 7.21 43.6 1.06 148 1.48 IT Sulfide (tr) Note; Estimatederror is between 0 and 2.5o/"(van der Plas and Tobi, 1965).Tra@ amount indicatedby tr. 534 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST

TABLE2. Calculatedbulk compositionsof WR I and WR ll

MP3-5 WR II Sample MP11-1 83-128 83-15 Meas.' Average s.D 83-19A 83-16 sio, 51.8 50.9 50.5 49.7 (nl 50.19 50.6 0.79 51.0 50.3 Al,o3 14.5 13.5 15.1 14.2 12.5 14.52 14.0 1.02 14.0 14.7 Tio, 1.0 0.86 141 1.7 1.80 1.01 1.36 0.41 1.08 1.10 't2.7 FeO 11.5 11.49 11.6 11.6 12.71 11.8 0.51 11.9 11.1 MnO 0.21 0.19 0.18 016 0.19 n 10 0.19 0.02 0.18 0.15 Mgo 6.58 7.49 6.15 6.94 729 7.29 6.89 0.54 7.94 7.79 CaO 12.1 13.1 119 13.0 12.9 12.9 12.6 u.co 10.9 12.4 Naro 2.21 z.Jo 3.03 2.55 2.42 2.42 2.51 0.31 2.74 2.43 KrO 0.05 0.12 0.10 0.17 0.11 0.11 0.11 0.04 0.18 0.13 H.o*o 2.19 nd 1.03 0.18 no no no no 2.3 15 Hro.*"'- 2.54 no no no 0.31 no nd no nd 0.49 Total 99.95 100.01 99.97 r 00.02 100.01 99.61 99.99 99.92 100.1 Notei nd : not determined. . Measured by rcp-AEs,Environmental Trace Substances ResearchCenter, Universityof Missouri, Columbia. ** H,O measuredatter ignitionof volatiles.

current of 12.5 pA was used. Unknowns were arlalyzed to the exchangeof Ca for Fe and Mg between orthopy- with a spot sizeof < l0 pm or 17 pm when reintegration roxene and other mafic phasesupon cooling. was necessaryon exsolved phases.F and cl were mea- The ranges in analytically calculated (Davidson and sured separately,on the same grains, with a spot size of Lindsley, 1985) temperaturesfor all analyzedare approximately l5 pm and countingtimes of about 60 s. indicated in Table 4. The wide range of temperatures Data were reduced by the method of Bence and Albee observedfor individual samplesresults mostly from vari- (1968)with the correctionfactors ofAlbee and Ray (1970). ation betweengrains rather than from zoning in individ- The unit-cell formulae and site distributions for amphi- ual grains, as the grain size is usually too small for sig- bole were obtained following the proceduresof Papike et nificant zoning. The maximum temperaturesof the ranges al. (1974).In this method, all Fe is initially assumedto appearto reflectcontact-metamorphic temperatures. Any be Fe2t, and then Fe3*is calculated(Table 3a). The result temperature values falling lower than the maximum val- is the minimum estimateof Fe3+that is consistentwith ue indicated are from grains that (l) did not equilibrate the chemistry and stoichiometry of the amphiboles. Fe3* at the peak metamorphic temperature becausethey were obtained by this method is believed to be most appro- not saturatedwith orthopyroxene component or (2) have priate for Morton Pass amphiboles since the various re-equilibrated to lower-temperaturecompositions upon hornfels are strongly reduced and any oxy component cooling. (and consequently Fe3*) in the amphiboles is therefore The Davidson-Lindsley thermometer yields small (Russ, 1984).Representative analyses or averages minimum temperaturesof >670 to >770 "C for the of all phasesare tabulated in Table 3. hornblende hornfels facies,which has clinopyroxene but no orthopyroxene. Rocks from the orthopyroxene horn- GnorHnnilroMETRy AND GEoBARoMETRy fels subfaciesyield temperaturesof about 760 to 900 'C, Contact-metamorphic temperatures have been esti- the lower of which appear to be reset. Temperatures of mated using the graphical two-pyroxene exchangether- 907 and 933 "C were estimated for the olivine hornfels mometer of Lindsley (1983) and the analytical augite- subfacies. solution-model thermometer of Davidson and Lindsley Attempts to apply the two-oxide geothermometer of (1985).Quadrilateral components for both thermometers Spencerand Lindsley (1981) to coexistingmagnetite-il- were projected from nonquadrilateral components using menite pairs were unsuccessful.The samplesare strongly the schemeof Lindsley and Andersen(1983). The tem- reduced with primary ilmenite grains having only I to 3 perature estimates are listed in Table 4; error bars on molo/oFerOr. The reintegrated magnetitescontain 30 to both thermometersare + 50 "C. Clinopyroxenesin ortho- 80 molo/oulvdspinel. Such oxide-pair compositions fall pyroxene-absentrocks are not saturated with orthopy- in poorly calibrated regions of the thermometer near the roxene component and therefore give minimum temper- WM buffer where the spinel and ilmenite isopleths are atures.Graphical orthopyroxene temperatures(Lindsley, nearly parallel. Temperaturesobtained from this method 1983)-based upon averagesof all orthopyroxenedata were generallylow (400 to 500'C), suggestingthat oxi- for each sample-are low, 605 to 760.C, The graphical dation-exsolutionprocesses reset the oxide upon cooling. clinopyroxene temperaturesare basedon averagesofcli- A few temperature estimates were obtained from the nopyroxene grains near but not touching orthopyroxene adjacentpelites by Bochensky(1982) using two-feldspar where orthopyroxene is present. The low temperatures geothermometry(Stormer, 1975) and Fe-Mg partitioning calculated for the orthopyroxene grains are probably due betweengarnet and cordierite(Thompson, 1976).Boch- RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST 535

TABLE3a. Electron-microprobeanalyses and structuralformulae of amphibolesfrom Morton Pass hornfels wRl WR II Sample MP11-1 83-12B 83-15 MP3.6 MP3-5 83-19A 83-16 sio, 47.5 47.5 44.6 43.9 41.4 47.1 47.0 Al,o3 /.oc 7.75 9.42 11.7 11.9 7.38 7.98 FeO. 184 17.3 187 17.0 17.O 17.3 170 M9o 10.6 11.2 10.3 10.3 103 11.6 11.8 MnO 0.26 0.27 0.18 0.14 o.17 0.26 0.19 Tio, 1.14 1.41 2.19 z-to 2.71 1.48 1.65 CaO 111 11.3 11.1 11.1 10.9 111 11.3 Naro 118 120 107 2.53 2.44 105 1.37 KrO 0.09 021 U.JO 079 o.92 0.24 o.24 Total 97.9 98.1 qA7 100.3 97.8 07R 98.5 F 0.06 nd 021 nd 0.35 0.08 0.14 cl 0.03 nd 0.06 nd 0.37 0.12 0.21 Cations per 23 oxygens Si /.uo 7.02 o.oJ 6.42 6.25 7.00 6.92 Al(v 0.94 0.98 1.35 1.58 1.75 1.00 1.08 Alvl 040 iaa 030 0.45 0.38 0.30 0.31 Fe2* 2.29 214 2.33 2.O8 2.15 2.14 2.10 Fe3* 0.0 0.0 0.0 0.0 0.0 0.02 0.0 Mg 2.35 2.47 2.29 2.24 2.31 2.5t 2.59 Mn 0.03 0.03 0.02 0.02 0.02 0.03 0.02 Ti 0.13 0.16 0.25 0.30 0.31 o.17 0.18 1.77 179 1.77 1.74 1.76 1.77 1.78 Na(M4) 0.03 0.04 0.04 0.17 0.07 0.0 0.02 Na(A) 031 0.30 0.53 0.55 0.65 0.30 0.38 K 002 0.04 0.07 0.15 018 0.05 0.04 A site 0.34 034 U.OU 0.70 o.82 0.35 0.42 F 0.03 no 0.10 no 0.17 0.04 0.05 0.01 nd 001 no 0.10 0.03 0.05 oH. 1.96 1.89 nd 1aa 193 190 Mg/Fe 1.03 1 15 nan 1.08 1.07 1.20 1.23

Note; nd : not determined. - OH calculatedas 2 - (F + CD

ensky'stwo-feldspar temperature range (655 to 708 'C) thermometer also yields a value lower than temperature obtainedfor his sampleMP- I is about 100'C lower than estimatesfrom orthopyroxene hornfels of that area. This the 812 'C temperatureestimated for the nearby ortho- lower temperature may be the result of poor calibration pyroxenehornfels MPl2. The estimateof 715 + 60'C of this geothermometerarLd/or re-equilibration of Fe and for his sample MP-705 using the garnet-cordieritegeo- Mg between garnet and cordierite upon cooling, an in-

TABLE3b. Averageelectron-microprobe analyses and structuralformulae of pyroxenesfrom Morton Pass hornfels

WRI wR ll 83-15 MP3-6 MP3-5 83-16 MP11-1 83-128 Sample cpx cpx cpx cpx opx cpx opx cpx opx

sio, ^1 7 52.9 azo 51.2 51.9 50.8 3Z-U 51.5 53.3 51.5 Alro3 1.03 0.68 1.29 0.68 1.37 1.30 1.66 1.O2 1 11 0.98 FeO 134 12.3 14.5 32.8 12.4 29.3 13.2 29.1 12.1 29.7 Mgo 107 11.6 11.3 14.9 12.7 17.5 12.O 17.2 11 .9 16.3 MnO 043 0.33 0.3 0.70 0.24 0.2s u,cc 0.25 0.68 Tio, 0.14 0.03 0.05 0.06 0.25 0.25 0.08 0.07 0.01 0.03 21.9 22.8 20.7 0.82 20.9 0.88 208 0.73 21.5 062 Naro 0.19 009 o.23 0.00 0.38 0.00 014 0.00 Total 99.4 100.7 101.0 101.1 100.1 100.5 100.1 100.1 1003 ooo Cations per 6 oxygens Si 1.91 1.99 1.98 1.98 1.95 1.95 1.97 1.98 2.00 1.99 AI 0.05 0.03 0.06 0.03 0.06 0.06 0.07 0.06 0.05 0.05 te'* 043 0.39 0.46 1.06 0.34 0.94 0.42 0.94 0.38 0.96 Fe3* 000 0.00 0.00 0.00 0.05 0.04 000 0.00 0.00 0.00 Mg 0.61 0.65 0.63 0.86 0.71 1.00 0.67 0.99 0.67 0.94 Mn 0.01 0.01 0.01 0.02 0.01 0.02 0.01 0.02 0.01 0.02 Ti 0.00 000 0.00 0.00 0.01 0.00 0.00 0.00 000 0.00 Ca 0.90 0.92 0.84 0.03 0.84 0.04 0.84 0.03 0.86 0.03 NA 0.01 0.01 0.02 0.00 0.03 0.00 0.01 0.00 0.01 0.00 536 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST

TABLE3c' Averageelectron-microprobe analyses and structural formulae of plagioclase,ilmenite, , and olivine in Morton

Plagioclase WRI 83-128 83-15 83-16

sio, 53.4 55.4 55.7 54.5 55.8 57.4 54.0 Al2o3 29.5 28.6 28.7 28.5 28.1 27.40 29.6 FeO 0.23 0.20 0.15 0.29 0.21 0.19 0.15 Mgo MnO Tio, CaO 12.6 11.4 10.9 11.1 10.4 10.5 11.9 Naro 4.47 5.25 5.75 4.94 5.61 6.18 4.85 K.o 0.01 0.05 0.03 0.12 0.06 0.06 0.01 Total 99.4 100.94 101.2 99.40 100.1 101.3 100.6 Cations pef 8 oxygens Si 2.41 2.48 2.48 2.47 2.51 2.55 2.43 AI 1.57 I Fl 1.51 1.52 1.49 1.43 1.57 Fe 0.01 0.01 0.01 0.01 0.01 0.01 0.01 Fe3* Mg Mn Ti Ca 0.61 0.55 0.52 0.54 0.50 0.48 0.58 NA 0.39 0.46 0.50 0.43 0.49 0.53 0.42 K 0.00 0.00 0.00 0.01 0.00 0.00 0.00

An 61 54 F1 47 58 llm Hem usp Mt

FA

' F€F+in oxides was estimated by charge balance. Mineralend-member compositions given in mole percent.

trinsic problemwith this geothermometer(Essene,1982). phism. Although there is variation in the proportions Pressurewas estimatedat approximately 3 kbar by Fuhr- among the phaseswith grade (Fig. 2), there is little com- man (1986),Fuhrman et al. (1988),and Grant and Frost positional variation in phasesother than amphibole. This (unpublishedmanuscript, I 988). suggeststhat there is a strong relationship between the The temperatureestimates at Morton Passare >676 modal abundance of the other phases and amphibole to >776'C for the hornblendehornfels and 760 to 933 composition. The chemographic relations of the mafic "C for the pyroxene hornfels. In SiOr-saturatedrocks, cli- phases plotted on the quadrilateral (QUAD) plane are nopyroxene formed at temperaturesof >676 .C, and or- shown in Figures 4A and 48. Numbers I through 5 (Fig. thopyroxene formed between 680 and 827.C. No quartz 4A) and I and2 (Fig. aB) indicate the progradesequence. was observedin rocks recording temperaturesabove 890 There is little changein the relative proportions of 'C. QUAD In SiOr-undersaturatedrocks, the assemblagehbl + componentsin the amphiboles with increasingmetamor- plg + ilm persisted to >730 .C, clinopyroxene and or- phism although a slight increasein Mg/Fe is indicated. thopyroxeneappeared between 760 and 830 9C,and oliv- However, there is progressive enrichment of the non- ine occurs in rocks metamorphosedat about 905 "C and QUAD components(Fig. 5). The increaseof non-QUAD above. The isotherms inferred from contact metamor- components can be explained by the consumption of phism are shown in Figure 3. The fluctuations in the iso- QUAD components during reactions that form other therms at the highest grade may reflect the uncertainties phases.The composition of the remaining amphiboles in temperatureestimates; notwithstanding, they also sug- evolves by continuous reactions,resulting in amphiboles gestthat igneousrock is presentdirectly under the supra- enrichedin non-QUAD components. crustals. The increaseof total Al with increasinggrade is partic- ularly intriguing. Al"I, an empirical indicator of pressure THn crrnnnrcAl EvoLUTIoN oF AMpHTBoLE (Leake,1965; Hynes, 1982;Hammerstrom and7,en,1986) Samples from WR I and WR II are representativeof has no systematicvariation (Fig. 5A) but AlI" clearly in- the mafic hornfels in Morton Pass,and their phasecom- creaseswith grade. The correlation coefficientsin Table positions relative to temperature are used to develop a 5 show a measureof the linear relationship betweenany quantitative model to characterizg prograde metamor- two of the variables listed. Both Table 5 and Figure 6 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST 537

Passhornfels

WRI WR II MP11-1 83-16 t|m Mt llm 0.22 0.14 0.48 o.17 0.48 0.38 noA 0.11 32.2 34.2 0.25 0.19 153 0.62 1.99 0.18 1.12 50.5 0.0 0.01 44 1 45.8 87.59 46.5 74.O 46.1 67.9 47.1 51.2 49.6 0.0 0.11 0.0 o.27 0.03 0.03 0.11 0.08 16.0 15.5 4.0 1.9 0.17 0.78 0.44 0.54 0.45 1.06 0.63 0.56 48.7 53.3 2.90 51.5 16.1 51.5 24.7 0.49 0.0 0.11

97.3 101.4 92.66 99.9 aql 98.8 95.2 98.9 100.0 100.0 Cationsper 3 or 4 oxygens Cations pel 4 oxygens 0.01 0.00 o.02 0.00 0.02 0.01 0.04 0.00 0.99 1.02 0.01 001 0.07 0.02 0.09 0.01 0.05 0.00 0.00 0.00 087 1.00 1.10 0.95 1.47 1.00 1.73 0.94 1.30 1.24 0.09 0.00 1.73 0.03 0.93 000 0.45 0.06 0.00 000 0.00 0.01 0.00 0.00 0.01 0.00 0.69 0.09 0.04 0.01 0.02 001 0.01 0.01 0.02 0.02 0.01 0.95 1.00 0.08 0.97 0.47 1.00 0.71 0.97 0.00 0.00

95.2 100.3 98.7 100.0 97.00 4.8 -0.28 1.31 -0.02 300 8.30 47.7 74.4 91.7 52.3 zJ.o

suggestthat, at least for WR I and WR II samples,the concentration of AIIV is dependent on temperature. In fact, it appears that for assemblagesthat include hbl + plg as the only aluminousphases, AlIv in amphibolemay be usedas a thermometer(Russ, 1984; Russ and Linds- ley, 1984;Nabelek and Lindsley,1985; Russ-Nabelek and Lindsley, work in progress).Thus, variation in amphibole components with respect to AIIV are relevant to changes in temperature. In Figures 5,A,through 5D, the enclosed areasindicate values for 282 analysesof amphibole from Morton Passand HbH (hornblende hornfels), OpH (or- thopyroxene hornfels), and OIH (olivine hornfels) mark the different facies or subfacies.The overlap of the HbH, OpH, and OIH fields is an indirect result of bulk-com- position differencesbetween amphiboles in all the mafic schists plotted and reflects amphibole composition in rocks with quartz where orthopyroxenemay have formed relatively early. The line of slope 1.0 in Figure 5A, cor- responds to a Tschermak substitution vector (AlvIAlIVMg-,Si-r). A pure Ti-Tschermak substitution from (Fig. 58) requires a l:2 ratio of Ti to AlI". The nearly parallel path with slope 7z of the Ti-Tscher- mak vector to the Ti abundancein amphibole showsthat Fig. 3. Map of isotherms in mafic schists at Morton Pass. the total amount of Ti may be described by this vector Solid circles indicate sample locations. Fluctuations in the iso- but not AlI". Al"I, Fe3*, and Tia* (Fig. 5C) are preferen- therms suggestthat igneousrocks may be presentbeneath parts tially ordered into the amphibole M2 site, causing site of the supracrustalunit. 538 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST

Co Co

1.2 HornblsndoHornf6l3 3 I Ofthopyror€neHornlsls WRI 4.5O Olivin€Hornt€lg

Formulo Proportion

Formulo Proportion

M9 Fe

Fig. 4. (A) Chemographic relations of mafic phasesin WR I plotted on QUAD. Numbers I through 5 indicate the prograde sequence.(B) Chemographicrelations of mafic phasesin WR II, plotted on QUAD. Numbers I and 2 indicate the progradesequence.

chargeimbalances of 1l or -12. The excesscharge is tions dominate over the Tschermak substitutionsin order counterbalancedby Al3* substitution for Si in the tetra- to maintain the required chargebalance. The perfect cor- hedral site as well as Na for Ca in the M4 site (Cameron relation (Table 5) of ED (edenite exchangecomponent) and Papike, 1979).Thus, a positive trend in Fe3** 2Ti with Al'" indicates that edenite is the dominant compo- * Alvr suggeststhat M2-site occupancyof thesecompo- nent for AIIV enrichment and that plagioclaseexchange nents is indirectly related to temperature increases.For (NaSiCa-,Al-,) is the least prevalent. M2-site occupancyof pure Fe3*,Ti4*, Al3t, or any com- K, F, and Cl appear to be retained in the residual am- bination thereof,a 1:l or l:2 correlationwith AlIv is re- phibole.For example,K contentis inverselyproportional quired, giving a ferri-Tschermak, Ti-Tschermak, or to the volume percent of amphibole in WR I rocks (Fig. Tschermakcomposition. From Figure 5C it is obvious 7A). Regressionof the curve (Fig. 7A) gives the relation that a combination of these substitutions is responsible K (in formula proportion units): 1.75 + 0. ll + volo/o for M2-site occupancy. The trend in A-site occupancy amphibole. F and Cl are also progressivelyenriched in with increasinggrade lies along a line with slope of about amphiboles with decreasingamphibole abundance (Fig. 0.5 and AIIY intercept of 0.5, intersectingthe pargasitic 7B), giving the relations F (in formula proportion units) trend only at high grade. Similar to the filling of the M2 :1.66 t 0.17 + volo/oamphibole and Cl (in formula site, A-site occupancymust also counterbalancecharges proportion units) : 0.82 + 0. 17 + volo/oamphibole. Am- in the tetrahedral site. Becausethe pargasite composi- phiboles from WR II have less F but the same amounts tional trend is the sum of I Tschermak and I edenite of Cl as equivalent amphiboles from WR I. The inverse substitution (NaAlSi-rtr-,), the unique A-site composi- relationship with volume percent amphibole (not shown) tional trend for Morton Passamphiboles is due to a ratio is the same, however, suggestingthat amphibole break- other than 1:l for thesesubstitutions. Indeed, it is sug- down is almost entirely a dehydration process. gestedthat A-site occupancywith AlI" edenite substitu- With rising temperature and amphibole breakdown, RUSS.NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST 539

Trsle 4, Temperaturesof metamorphismof MortonPass horn- Trele 5. Linearcorrelation of amphibolecomponents and tem- fels oeraturefor WR I and WR ll

Temperature('C) AIIV Loc. At," 1.00 Graphical Analytical T 0.92 1.00 tn thermometer. thermometer.. Fig. PL 0.88 0.77 100 1 Sample cpx opx Aug( r range)t ED 0.98 u.vc 0 80 1.00 TK 097 noa 0.94 0.94 1.00 Hornblende ho]ntels : A MP11-1 >b5/ >676 (>645 to >676) PL NaSiCa-,Al-, B 83-17A >690 >759 (>640 to >759) ED : NaAlSi, D MP15.4 >690 >734 (>661 to >734) TK: Alv'A1vMgrSi-r F 83-10 >630 >741 (>621 to >7411 G 83-11 >605 >726 | 83-128 >620 >679 (>623 to >679) H 83-19Att >670 >729 (>592 to >729) progressofsuch reactions along a P, T, path can be J 83-13 >500+ >776 fnro E MP18-1 >650 >740 (>663 to >740) calculated and is particularly useful in quantifying pro- K MP1-3A >620 >676 grade metamorphism when an isochemical seriesof sam- Orthopyroxene horntels ples is available for which temperaturesare known. M 83-18 830 to 950 610 903 (695to 903) To visualize the progressof the net-transfer reactions o 83-15 805 605 839 (793 to 839) L 83-14 760 610 760f (711 to 760) that transform an initial assemblageinto the final assem- N MP6-4 715 640 889 (649 to 889) blage at Morton Pass, reaction space, as developed by o MP3-4 720 710 827 (743to827) (1982), The P 83-16 780 650 831 (769to 831) Thompson et al. is utilized. subassemblage T MP57-2 715 610 7731 1687to 773) amp + plg + pyx + (olv or qtz) is compositionally de- S MP57-9 755 640 872 (714to872) fined by the oxidesSiOr, AlrO3, FeO, FerOr,MgO, MnO, Olivine hornfels TiOr, CaO, NarO, KrO, and HrO, which may be con- R MP3-6 850 710 907 (816 to 907) NarO, MgO, AlrOr, SiOr, R MP3-5 870 760 933 (813to 933) densedinto the subspace CaO, and H,O (NCMASH) (Thompsonet al., 1982).Because From Bochensky the systemis open and HrO is continuously lost, water is (1s82): tral.l++ Cord-Gar$ treated as a separatephase denoted "env." Nine possible MP.1 655 to 708 MP-203 707 (+60) exchangecomponents (Table 6) operate within and be- MP-705 715 (+60) tween the additive components amphibole, plagioclase, . Graphicaltemperatures based on the thermometerof Lindsley(1 983). pyroxene,and olivine, but only TK, ED, PL, and MgCa ' The temperaturesare based on average pyroxene compositionsassum- (see above) are considered here to be significant. The ing 3-kbar pressure. Minimum temperaturesare from rocks with no or- phasesamong which continuous exchangereactions may thopyroxene.Uncertainties are +50'C. .. Temperatureestimates based on solutionmodel "A" of Davidsonand Lindsfey (1985) assuming 3-kbar pressure. Minimum temperatures arc from rocks withoutorthopyroxene. Uncertainties are +50'C. t Rangeof temperaturescalculated for all grainsanalyzed. TABLE6. Reactioncomponents, phases, and notationfor the t'l Sample 83-19A does not have clinopyroxene.Temperatures were maficschist subassemblage estimated from clinopyroxenegrains in the reaction rim that divides the mafic layer from a calc-silicatelayer. Component Mineralohase f Probably reset. ft Temperaturesfrom twoJeldspar geothermometer(Stormer, 1975) Additive component $ Cordierite-garnettemperatures after Thompson (1976). NaAlSi30s (ab) plg CarMg5SisO,,(OH), (tr) amp CaMgSi.Ou (dD- pyx sio, (qz) qrz Mg,SiOo (fo) olv continuous reactionsenrich the remaining amphiboles in H.o (env) HrO non-QUAD componentsas QUAD amphibole compo- Exchangecomponent nent becomesunavailable. Thesechanges are best shown AlrMg-1Si-1 (TK) amp,pyx'" through the substitution operators Tschermak, edenite, NaAlSi_1 (ED) amp,pyx NaSiCa (PL) amp,prg, pyx KNa-,, CIOH and FOH-,, respectively(Thompson, 1Al-, ,, MgCa , (MgCa1) amp,pyx, olv prograde 1982;Thompson et al., 1982).The enrichment Other possible exchange components trends of edenite,Tschermak, and KNa-, componentsfor FeAl 1 amp,pyx the hornblende hornfels faciesand the orthopyroxeneand FeMg , amp,pyx, olv (TiFe amp pyroxene FeTiFe , ,) olivine hornfels subfaciesofthe hornfels facies KNa 1 amp,prg are shown in Figure 8. MnFe, amp,pyx, orv Alofei The convention lowercase letters for additive components and Rn.q.crroN sPAcE UPPERCASEtot exchange components (as shown, except for MgCa 1) Simple exchangereactions and net-transfer reactions, is used throughoutthe text. t Additivecomponent di is sufficientto describeboth clinopyroxeneand which have both additive components (phases)and ex- orthopyroxene because di + MgCa-1 : enstatite; hence they are not change components, can describe modal and composi- linearlyindependent. -'E.9., pyroxene. tional changesofphases from one facies to another. The Al,Mg-1Si-1in amphibole: ALMS 1Si-,in 540 RUSS-NABELEK:AMPHIBOLE REACTIONS IN MAFIC SCHIST

1.5 . WRI A tr WRtr 0.3

, 0.9 a oltl o.t ? a.. .a 0.1 )opn

0.0 0,5 1.5 2.0 0.0 0.5 1.0

1.0

0.8

+ 0'6 E F d + 0.4 < 0.4

0.2

0.0 0.5 1.0 t.s 2.o 0.0 0.5 1.0 l.s IV AIIV AI

Fig. 5. Variations of site occupancieswith ALIV. (A) Plot of of Ti vs. AIIV shows a positive enrichment trend for Ti in am- Al'r vs. A|v shows very little systematicvariation. The line of phibole. Ratio of I Ti to 2 AIIVis the Ti-Tschermak substitution. slope I correspondsto pure tschermakitic substitution from tre- (C) Plot of Fe3* + 2(Ti) + Al'r vs. Alr" illustrates M2-site sub- molite at the origin. The enclosedareas represent amphiboles stitution in amphibole with increasing grade. Ferri-Tschermak from all mafic schist analyzedat Monon Pass. HbH : horn- and Tschermak substitutions operate along a line of l:1. Ti- blendehornfels facies,OpH : orthopyroxenehornfels subfacies, Tschermak operatesalong a line of l:2. (D) Plot of A-site oc- OlH: olivine hornfels subfacies.Circles are averagedanalyses cupancyvs. AlIv; A-site substitution is describedby the line A-site from WR I; squaresare averagedanalyses from WR II. (B) Plot : 0.5Alrv - 0.25.

occur are indicated to the right in Table 6. All other ex- rocks.The reactions changecomponents possible in the l2-oxide composi- tional spacemay be accessedfrom the condensedspace. tr: 3.5di+ qz + l.5MgCa t, "ot," (5) Transformation of the NCMASH oxide components ab:4qz + ED, "0," (3) into the phase components (tr, ab, di, qz, fo, and env) and exchangecomponents (TK, ED, PL, and MgCa_,) and yields a set offour independentnet-transfer reactions: di + qz + PL + TK: ab, ",y" (4) fo + qz : di + MgCa-, (l) polygonal t:2di + l.5fo * 2.5q2-t env (2) are the axesofa coordinate systemdetermined for quartz-bearing rocks (Fig. 9). Equation 5 is a simple ab:4qz + ED (3) ReactionsI that eliminatesolivine. di + qz + PL + TK: ab. (4) combinationof and2 By making linear combinations of Reaction 3 with Re- These reactions are sufficient to describe the modal and actions l, 2, and 4 to eliminate quartz, the following re- compositionalvariations in phasesin the defined l2-ox- actions for quartz-absentrocks are determined: ide compositionalspace of the Morton Passhornfels. Note ab + 4fo:4di + ED + 4MgCa "d," (6) that Equation I becomesmore familiar if MgCa , is com- ,, -l bined with di to becomefo qz: enstalite.Particular 4tr + 2.5ED: 8di + 6foI 2.5ab, "r," (7) linear combinations of Reactions I to 4 can eliminate one or more select phasesand result in reactions that have and the subassemblagesobserved at Morton Pass.Because of 4dr + 4PL + 4TK: 3ab + ED, (8) the constraints of working in three- rather than four-di- mensionalspace, two polygonshave beendevised to show Both reaction spaces,the first necessitatingthe presence reaction progressin the quartz-bearingand quartz-absent of quarlz and the second allowing for the formation of RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST 541

18

. WRI 1.6 o WRII

E r.4

l2 I { r-o

o8- 600 700 600 900 1000 TEMPERATURE ('C)

Fig. 6. Plot of AlIv vs. temperature,from WR I and WR II pyroxenetemperature estimates, suggests that AlIv in amphibole is a functionof temperature.

Fig.8. Progradeenrichment trend of KNa-,, Tschermak,and olivine, are therefore related by Reaction 3. Linear com- edenitecomponents in WR I amphiboles.The hornblende horn- binationsofthese reactionsare vectors.From inspection fels faciesis unlabeled;orthopyroxene and olivine hornfels of both setsof reactions,one can identify particular com- subfaciesare indicated. Temperatures derived from clinopyrox- binations of reactions that occurred at Morton Pass.As enethermometry are also shown. a reference point, unpublished experimental results on the olivine hornfelsMP3-5 recrystallizedat 600'C and 3 kbar were used. The experimental amphibole and pla- planes(e.g., di + fo) unlessthey react out of the assem- gioclasecompositions make an excellent referencepoint blage. The modal amount of each phase may be repre- for the models, becausethe rock under thesepressure and sentedby the number of oxygen units in its additive com- temperature conditions must be similar to the precursor ponent (ad) (Thompson et al.). An"o(Table 7) is the change of the contact-metamorphic assemblagesof WR I. The in the number of oxygen units of each of the additive compositionand estimatedmode of the reference-point componentsas a result of reactionsa, B, and 7 or 0, e, amphibolite are listed in Table 7. and f. The total distribution of the exchangecomponents The bounding planes of the reaction-spacepolyhedra and additive componentsbetween the reaction vectors of (Figs.9 and l0) indicate(1) the limits of exchange-com- eachpolygon is listed in Table 7. The polesto the bound- ponent capacityand (2) the limit of availability of a phase ing planes of the polyhedra are also vectors; if shown, for reaction. Exchange-componentlimits are the mini- they would intersect at the origin and have rnagnitude mum and maximum operations of the exchangecom- and direction specifiedby the equations in Table 7. Not ponentsin the phases(indicated in Table 6) in which they all exchange or additive components form bounding occur,for exampletTK-l and [TK+] indicate the min- planes becausethe limits of exchangeof the former can imum and maximum amount of Al"tAlr"Mg-,Si , in ad- never be reached or the latter will never be totally con- ditive components tr and di. The reaction limits of ad- sumed.Because in somecases, the mineral composition ditive componentsform planeswhere phasesdisappear. is known better than the mode. zl". (Table 7) is replaced Phasesnot presentin the original assemblageare formed by the quantity (X,-n^o)/e,where e is the number of oxy- as metamorphism proceeds but do not form bounding gen equivalents per mole of additive component. To ob-

K=c/vol%amphibole F=clvol%amphibole c=1.6610.17 E o.zo c=1.7510.11 E o.rs Cl=ci vol % amphibole c=0.8210.17 g o.rs : 0.r0

= 0.10 c o.os - 0.0s 6 v A 0.00 k o.oo 0 r0 20 30 40 s0 60 0 10 20 30 40 50 60 Volume7o amphibole YolumeVo arnphibole

Fig. 7. (A) Volume percent of amphibole vs. K (in formula proportion units) for WR I. (B) Volume percent of amphibole vs. F and Cl (in formula proportion units) in WR I. 542 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST

-02. 3n

b

C

Quartz- tholeiites

I WRI o Other

Fig. 9. (a) Quartz tholeiite reaction-spacepolygon correspondingto olivine hornfels MP3-5 of olivine tholeiite composition with 5 vol0/oquartz. The coordinate axesa, B, and 7 indicate advancementalong (net-transfer)Reactions 5, 3, and 4. The facesof the polygon are bounding planes, indicating the point ofelimination ofadditive components or the limits ofexchange capacity. MgCa , is abbreviated as M-. (b), (c), and (d) are orthographic views. Letters refer to particular samplesand correspond to the sample numbers and locations in Fig. I and temperaturesin Table 4. tain mass-balanceequations for where i : exchangecomponent l, : additive component each exchangecompo- "r nent, the quantity (X",n"o/e)and the equations of Table j, k: a specificnet-transfer reaction (o,0, ^(,6, e,or f), t 7 were substitutedinto the equationofFerry (1984) : progressof Reaction d, 0, 7, d, e, or f, {, : mole fraction of component I in phasej, u,, : the stoichiomet- X!.,n1+ ,,,ru: x;(,1+ ) ,,,s), 1s'sric coefficient of exchangecomponent i, in Reaction a, B, 4 4 "y,d, e, or (, n, : the number of moles of phaseJ, and the RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST 543

TABLE7. Distributionof theadditive and phase components among Reactions a, B,t, d,e, and f

Plane Reference-pointamphibolite Ouartz tholeiite model applied to locks with tr + ab + qz + di - -aa. 2ng: IT a4, : 4, n,o,: 0.655" : - n3o: ab An"o fi"o n5: Ap + A? 0.340.. - : 3n&: di Lno,: no, n3, 0.875Aa- 0.75A1 o.- - qz Lnq: nq n&:0.093a" - Lg - 0.25L1 n3,: 0.. TK D-* - D9*: -0'125A'Y x+t:0.677 xcj: 0.196 ED r?eo-flPo:-0.125A4 : PL /'lpr- /?B.: -0'12547 XBJ: 0.137XBi* 0.502 - : xs[c"_,: 0.195 MgCa , /'lugc"-, n$sc", 0.06344 sn 0:0.06344 + 0.125A8 Olivine tholeiite model applied to rocks with tr + ab + di + lo Ln,,: n,,- nl, : -Ar ni.: 0.657t tr : A/'r"o: 4o - ngo: 0 3334d + 0.208Ae+ Af n:b 0.343t :ndi -n8i :46+0.5Ae-Al n3,: ol Andi 0t - n3.: fo An" : n" nP": 0.667ad+ 0.25Ae 0t TK h* - n9*: -0.16741 xw:0.677 - : - 0.0264r + 0'042Af xg8': 0.196 ED f,eo f,Po 0.042A0 : fr,' - frqt: -0 16741 XgJ: 0.137XBi* 0'502 PL : MgCa-, flvsc",- n$n""-,: 0.16746 xssc", 0 195 sn 0 : 0.04240 - 0 026At + 0.04241 Note. For definitionof variables,see text. . E g , (final mode of amphibole)- (initialmode of amphibole)is equivalentto 1 negative unit of advancementalong 4. -- The mode is estimated trom iig. b for quartz-bearingor quartz-LOsentassembliges of WR I at 600'C. There is no pyroxene or olivine in the reterenceassemblage. cold-seal-bombapparatus' t Unpublishedeiperimental results from recrystallizationof olivine hornfelsMP3-5 at 600'C and 3 kbar in hydrothermal'

superscripts0 and findicate initial and final conditions. space,see Thompson et al. (1982),Thompson (1982)' The mass-balanceequations that show the distribution and Laird (1982a,1982b). of the exchangecomponents between phasesin terms of The coordinates of the Morton Passhornfels (Figs. 9 the reaction vectors a, p, and 'y and D,€, and f are in Table and 10 and Table l0) are found by multiple regressionof 8. Evaluation of the equations in Table 8 using the max- the solutions of the four mass-balanceequations derived imum and minimum limits of eachexchange component from solving Equation 9 for TK, ED, PL, and MgCa r. yields equations for the exchange-componentfaces. The The solution equations are in the form Aa, + 80" + Cl" bounding-polygon planes are listed in Table 9. As one : y, andA6, * Be, * C(,: y,.The coefficientsA, B, and would expect, the planes of minimum exchange-compo- C of the independent variables are estimated by the nent capacity are located around the origin (Figs. 9 and method of constrained least-squaresin which the depen- l0), and the planesof maximum exchange-component dent variable/ is regressedon a, B, and'y and D,e, and f. capacity are farthest from the origin. The plane [sh] limits To account for the differing magnitudes of the values of amphibole composition to the range of all possible com- A, B, C, and /, the regressionfor each individual sample binations of Tschermak, edenite, and plagioclase ex- was weighted; however, weighting appears to have had changeand is the sameas the shadedplane of Thompson only a small effect on the final result. No analysis errors et al. (1982)and Thompson(1982). For a more complete were incorporated in this procedure, and therefore the discussionof the derivation and construction of reaction errors are simply those of the regression.The correlation

T^"t. S. M"ss-b"|. Quartztholeiite equations - TK: (Xr't -,yot)n|, * (Xr'oi- Xoo)n!, :1Xt'* -0.875Xr'd)a + 0.125(6xt'di 1)'l ED: (tr,r - yo,r)n?, * (tr.oi - Xo'di)n!,:1Xt" - 0.875X'o,1s - 0125A + 0.75X1d',y PL:(21t't-r1o't)nP.*(Xt.'o-Xo'"b)n8o:(Xt'dr-Xo'd)n8:Xtj'a-Xt'eb?-(Xr'ab+0'125)1 MgCa_,:(X'k-Xo't)nior*(Xr'at-Xo'dt)n3:(0.063+Xtv-0.875x1'o'1a+0.75X'o''y Olivine tholeiite equationg - TK. Xt,r - r1o,t)nto,* ()1t,or- Xo'd)ng: -Xr'di6 + (Xt'v -0.sxti,ile + (Xt'dt 0.167)f -,yo,t)nP. - - - - ED. Xr,r * ()6r,oi Xo.di)n3: P$42- Xr'dt)d (0.026 Xlt)r * 1,21tor 0.167)f PL: (;1r't_ yot)n?,*(xr'"0-,1'oao)n3o*(Xr'oi-Xodr)n3.:0.333xf"o6*(fft-o.rot;tao)e-(0.042+xt'ab)f - - - - - + Xt'dif MgCa_,: (Xr,r Xo,r)n tor+ (Xr,d' Xo.d)n li+ (Xt,to Xo,to)nF (- X''0, + O.667Xrr9+ 0.167)d * (Xt't o.sxr'di 0.25xt'to)e 544 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST

Oliuine- tholeiltes H>729 O WRI r WRI] O Other

Fig. 10. (a) Oivine tholeiite reaction-spacepolygon correspondingto olivine hornfels MP3-5 of olivine tholeiite composition. The coordinate axes 6, e, and findicate to advancementalong (net-transfer)Reactions 6, 7, and 8. The facesofthe polygon are listed in Table 6 and are described in the caption to Fig. 9. MgCa-, is abbreviated as M-. (b), (c), (d) are onhographicviews. Letters refer to particular samplesand correspondto the sample numbers and locations in Fig. I and temperaturesin Table 4. coefficientsto the right in Table l0 indicate the percent phibole, plagioclase,and clinopyroxene. Hornfels in the to which the model describeseach sample.In general,the quartz tholeiite space(Fig. 9) fall near the PL, ED, or TK model fits are excellent and suggestthat other exchange faces. Many of these samples are layered and reflect a components not utilized in the model are insignificant. wide variety of compositions. It is interestingto note that The location of an individual point in the reaction WR I samplesin the olivine tholeiite space(Fig. l0) fall polyhedra is sensitive to the sample's bulk composition on the same internal plane, indicating how close they and very sensitive to the composition of the phasesam- truly are in composition. There also appearsto be a sys- RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST 545

Treue9. Equationsfor the boundingplanes of polyhedrain face indicates that temperature-dependentreactions may Figs.9 and10 also be the most important mechanism by which amphi- Equation Plane bole decreasesmodally in the assemblage. ouartz tholeiite Drscussrou -U.b55 : -d ttrl -0.005 : 0.083a + B - 0.25t lqzl It has been shown thus far that changesin amphibole - 0:0.677a 01257 [TK_] composition in Morton Passcannot be ascribed only to 0.867:1.125q + 0.625? lrK+l 0 : 0.196a+ 0.125B+ 0.751 tED_] differencesin bulk chemistry but are a function of the 0.527:0.125a + 0.1258+ 0.75'y IED+l continuous and discontinuousreactions that occur during -0.169 : - 0.137a B + 0.875-l IPL_] prograde metamorphism of the isochemical hornfels. 1.221:2a - 0.5020+ O.377t IPL+l 0 : 0.258a lMgca 1l Discontinuousreactions are chiefly responsiblefor changes 0:0063a+0.1258 lshl in the modal abundanceof phasesand the development Olivinetholeiite of faciesand subfacies.Incremental net-transferreactions 0.658: e ttrl where lowest-graderocks are progressivelytransformed 0:0.667e + 0.1671 [TK_] 0: 0.0426+ 0.17e+ 0.O42( [ED_] into their high-grade equivalents are a combination of 0.529: -0.9586 + 0.474e+ 1.042f IED+l continuousand discontinuousreactions. -0.171 : -0.3336 + + 0.345e 0.m3f tPL_] The reaction-progresscoordinates in terms of the net- 1.226: -0.1676+ 2.104e+ 0.333f IPL+l 0: -1 167d+ 0 6951 lMgca 1l transfer reactions ot,0, ^yand d, e, and f indicate the pro- - 0 : 0.0426 0.026e+ 0 042t lshl gressof each individual sample relative to the reference point. However, to estimate the progressof reactions of each incremental step between samples,the coordinates tematic increasein temperature as the trend of samples of the previous sample are subtracted out. Using WR I approaches the amphibole-out (tr) face in the olivine samples as examples allows specific progressive net- tholeiite model. This suggeststhat the progress of the transfer reactions to be identified for each increase in whole-rock reactionsfor these rocks is dependentpri- metamorphic grade for a particular bulk composition. marily upon temperatureand not on other intensive vari- The assemblageMPll-2A is the lowest-grademafic ables. The systematic trend toward the amphibole-out schist at Morton Pass.The absenceof chlorite and sodic

TABLE10. Reactionprogress values and modelerrors for Morton Pass quartz tholeiiteand olivinetholeiite hornfels

Loc. in Fig 1 Sampleno Corr. coeff.'

Ouartz tholeiite

E t" 5d f, A2 MP11-2A 0.01 + 9.91 0 26 + 0.01 0.07+ 0.01 1.00 F 83-10 0.03 + 0.05 0.37+ 0.03 0.16+ 0.03 0.99 A1 MPl1-1 0.05+ 0.07 0.30+ 0 03 011r 004 099 83-11 0.10+ 0.00 0.29+ 0.03 0.07+ 0.00 1.00 I 83-128 0.09 + 0.03 0.21+ 0.03 0.06+ 0.03 0.96 E MP18-1 0.13+ 0 05 0.37a 0.05 0.14+ 0.05 0.97 B 83-17A 0.07 + 9.93 0.27+ 0.03 0.08+ 0.02 0.99 D MP15-4 0.15 + 0.01 0.31+ 0.01 0.09+ 0.01 1.00 N MP6-4 0.18 + 0.03 0.15+ 0.04 -0.03 + 0.04 0.96 MP3-4 0.05 + 0.02 0 34 + 9'91 0.17I 0.01 1.00

Olvinetholeiite tu t, fr H 83-19A 0.16+ 0.40 0.07+ 0.09 -0.24 + 0.20 0.65 T MP57-2 1.11+ 0.14 0.37+ 0.02 0.12+ 0.07 0.99 S MP57-9 0.56 + 0.03 0.14+ 0.00 0.09+ 0.01 1.00 I 83-14 0.60 + 0.04 0.05+ 0.01 0.09+ 0.02 1.00 P 83-16 1.48+ 0.25 0.22+ 0.03 0.25+ 0.10 0.97 o 83-15 1.53+ 0.22 0.30+ 0.02 0.40+ 0.10 0.98 M 83-18 2.03 + 0.43 0.21+ 0.04 0.25+ 0.10 0.96 R2 MP3-6 2.34+ 1.58 0.37+ 0.10 0.68+ 0.54 0.84 R1 MP3-5 2.63 t 0.39 0.35+ 0.03 0.70I 0.16 0.98 I 83-128 1.19+ 0.42 0.11r 0 07 023 + 0.22 0.72 Model results for data ot Spear ( 1981 ) 12-808(s99 "C) 1.31+ 0.73 0.20+ O.12 0.56+ 0.28 0.65 r2-57A(651 .C) 2.21+ 1.96 0.31+ 0.18 1.08+ 0.62 0.74 12-81A(707 rc) 0.86+ 0.60 0 23 + 0.56 0.43! 0.24 0.95 r2-116A (802.C) 0.86+ 0.23 0.31+ 0.00 0.15+ 0.01 1.00 12-102A (850 .C) 0.18+ 0.81 0.29+ 0.06 0.06l 0.35 0.95 T2-s5A(655.C) -0.03 + 0.04 0.13+ 0.13 0.09+ 0.03 0.99 . The planar correlationcoefficients indicate the goodness-of-fitthat solutions to Equations 14-17 and 18-21 (Table 9) have to a plane. 546 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST

TABLE1 1. Incrementalnet-transfer reactions for selectedhornfels, indicating the reactionoccurring from one hornfelsto the next

f range Reaction fc) 600to ? 0.01gtr+0.196ab+0.066P1[amp,pls]+0.066rKlam":lT.ll,rt#.02sMsca-,[amp]+0262EDlampl (10) 600to 676 0.048tr+0193ab+01osPllamp,plg,pyxl+0.105TK[amp,pyx]:1.135qt2+0063di+0.072M9ca-,[amp,pyx] + 0.298EDlampl+ 0048H,O (11) >676to 729 0.040tr+0.101q2+0030ab+0.057EDlampl:0.227di+0.060M9Ca1[amp,di]+0087Pl[amp,ab,di] + o.o87Klamp,dil+ 0 040H,O (121 >729to 839 0.760tr+ 0.019ED[amp]+ 0.684TK[amp,pyx] + 0.684PL[amp,plg,pyx]: 1.976di + 0.703ab + 114Mgoa-,[amp,pyx]+ 076H,O (13) >839to 907 0.264tr+0.249ab+0.21sEDlampl:0.456di+0.132fo+0.264M9oa-'[amp,pyx]+0.464Pllamp,plg,pyxl + 0.464TKlamp,pyxl+ 0.264H,O (14)

plagioclaseindicates that MPI l-2A is at a higher grade and > 679 "C. Both Reactions I I a and I 0a produce rather than the -hornblendehornfels transition. Ep- than consume q\artz. The formation of quartz continues idote is not observed,possibly because the 3-kbar pres- until amphibole becomes enriched enough in edenite sure is below its stability range (Liou et al., 1914; Apted component to react with quartz, as Reaction 12 suggests. and Liou, 1983).The modal abundanceand composition Reaction l2a persists in the quartz-bearing hornblende of the phases in hornblende hornfels between the hornfels from >679 "C until orthopyroxene begins to greenschist-hornblendehornfels transition and the first form. Although quaftz generallyis a common constituent appearanceof clinopyroxeneare therefore afected by the ofgranulites, it continuesto react out ofthe assemblage reaction that transforms the reference-pointhornfels into when orthopyroxene forms in the mafic schist. For ex- the lowest-graderock MPI l-2A (A2) bV Reaction l0 (in ample,in the orthopyroxenehornfels MP6-4 and MP3-4 'C), Table I l). A certain amount of the continuous reaction (temperaturesestimated at 827 and 889 quartz is all but gone.Reaction 13 (in Table I l) is the resultantnet- ampr + TK + ED + PL + MgCa_,: ?mpz (15) transfer reaction been hornblendehornfels 83-l28 and occurs at every stageof metamorphism. When combined orthopyroxenehornfels 83-15 (WR I) in olivine tholeiite with a fractional amount of Reaction 15 to representthe space.When (l) the continuous amphibole reaction is modal changeof amphibole between the referencepoint added, (2) the pyroxeneis expressedas clinopyroxeneand amphiboliteand sampleMPI l-2A, Reactionl0 becomes orthopyroxene,and (3) Reaction 2 is combined to include quafiz a representativereaction is derived: 0.159amp,+ 0.l96plg: 0.l40amp,+ l.001qtz + 0.019H,o. (l0a) 1.48ampo"-l 4qtz: 0.720amp'"* l.465cpx+ 0.51lopx + l.703plg+ 0.76H,O. (l3a) Clinopyroxeneappears in the assemblageabove 67 6'C in quartz-bearingrocks by Reaction I I (in Table l l). By This reaction expressesthe mechanism by which quartz adding in the continuous amphibole reaction and com- is removed from the assemblage.In quartz-poor hornfels, bining the exchangecomponents with the additrve com- this reaction occurs over a small temperature interval. ponents, the major reaction between the referencepoint Stoddard(1985) observed similar reactionsin the horn- and hornblendehornfels MPl l-l (WR D becomes blende granulite facies terrane in the northwest Adiron- dacks,New York, but unfortunately did not indicate tem- 0.203amp'+ 0.193p1g: 0.l55amp.+ l.l35qtz perature rangesfor his reactions. disappearsfrom + 0.063di+ 0.048H,O. (1la) Quartz the assemblagesof WR I before840'C. At this point, the Although somewhat higher grade,hornblende hornfels major net-transfer reaction that operatesin the orthopy- 83-l2B (WR D doesnot differ greatlyfrom MPll-l in roxene hornfels prior to the development of olivine (as mineral chemistry or modal abundance; however, the exemplifiedby orthopyroxenehornfels 83-15) is modal decreasein quartz between the two (Fig. 2) sug- l.480ampo:0.720amp' + l.465cpx+ 0.5llopx geststhat quartz begins to react out. Reaction progress + 0.703p1g+ 0.760H,O. (l3b) from MPl1-l to 83-128gives the resultantReaction l2 (in Table I l). Condensedinto simplet form, the reaction is This reaction describesalso the initial formation of or- thopyroxene in SiOr-undersaturatedhornfels or the con- 0.646amp,+ 0.101qt2+ 0.030p1g: 0.606ampo tinued changein modal abundance. + 0.227cpx+ 0.040H,O. (l2a) The occurrenceof olivine in high-grademaflc rocks is In the hornblende hornfels facies, Reaction I la is the rare; however, the mechanism for its appearanceis rela- initial clinopyroxene-forming reaction operating in horn- tively straightforward.Olivine joins the assemblageat ap- blendehornfels and appearsto be initiated between>676 proximately 907 "C by Reaction 14 (in Table I l), which RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST 547

transforms orthopyroxenehornfels 83-15 into olivine concomitant changesin the proportion of other phases. hornfelsMP3-6 (WR D. The combinedreaction is Amphibole was progressively enriched in non-QUAD componentsas its modal abundancedecreased and as 0.677amp,+ 0.249p19: 0.4l3ampu+ 0.985cpx component became unavailable, consequent to + 0.27lopx+ 0.132fo QUAD the formation of cpx, opx, and olivine. BecauseAIIV has + 0.264H,O. (l4a) a systematic relationship with temperature, such non- This reaction would continue until the ultimate demise QUAD enrichmentswith AlI" are relevant to increasesin of amphibole, predicted from Figure 2 to occur at about temperature.Thus, it is suggestedthat the "evolution" of 1000 'C, when pargsite is the final reacting amphibole the non-QUAD componentsin amphibole can potential- and no more amphibole can be formed. The continuous ly give quantitative information on the intensive param- progressivedecrease in modal amphibole shown in Fig- eters of metamorphism.Progressive and systematicen- ure 2 is consistentwith Reactionslla, l2a, l3a, and 14a richment of F and Cl with decreasingamphibole above.Reactions lla, l2a, and 13a show plagioclaseon abundanceindicates that amphibole breakdown is purely the reactants'side and Reaction l3a on the products' a dehydration processand that both F and Cl may sta- side, which is also consistent with the variation in the bilize amphibole at high temperatures. mode of plagioclaseshown in Figure 2. Reaction spaceprovides an effectivemeans for analysis Ilmenite and magnetite are minor phasesthat play an of the progressof metamorphism in the mafic hornfels. important role in amphibole breakdown reactions. Gen- Quartz-bearingmafic schist sampleslie within the quartz erally they are found as products of amphibole dehydra- tholeiite model, whose bounding planes are the limiting tion in which ilmenite bufers the Ti content of the re- components of Reactions a, B, and 7. The sample coor- maining amphibole. However, they are compositionally dinates lie somewhat scatteredwithin the polygon with variable owing to low-temperaturere-equilibation and do no apparent systematic relationship between them. The not show significantmodal changes;hence, they have been scattering indicates the effect that bulk composition has ignored in the modeling. on the progressof any individual reaction. Comparative- It is not possible to predict an adequate reaction for ly, the nonlayered and therefore less compositionally the modal and compositional changesthat occurred be- variable samplesof the olivine tholeiite compositions tween the reference-pointhornfels and the quartz-absent show that specific linear combinations of Reactions 0, e, hornblendehornfels assemblage 83-l9,A' (WR II). The ol- and f give resultant reactions that are strongly tempera- ivine tholeiite model, which is strongly dependent upon ture dependent. the presenceof pyroxene, gives a very poor model fit Analysis of reaction progressin quartz-bearing rocks (Table l0). Coordinates for experimental runs by Spear indicates that qlartz and clinopyroxene are products of (1981)under conditions analogousto metamorphismat the reaction between amphibole and plagioclase until Morton Pass,but where Prro : Pro., are listed in Table >616 "C where quartz began to be consumed.Quartz 10. Even thoseexperiments with excellentfits to the ol- completely disappearedby 900 oC; however, it was only ivine tholeiite model do not lie in consistentplacement a minor reaction constituent when orthopyroxene ap- with similar rocksat Morton Pass.Spear (1981) reported pearedbetween >679 and 825 'C. In mafic schist that the first occurrence of clinopyroxene from reaction of had no quarrz, reaction space involving olivine shows hornblendeand plagioclaseat 768 + 8'C (l kbar) and at that clinopyroxene and orthopyroxene appeared in the 788 + 8'C (2 kbar) in olivine tholeiiteexperimental runs. assemblagebetween >729 and about 800 'C (e.g.,WR The transition to granulite faciesdenoted by the first oc- II). The olivine hornfels subfaciesdenoted by the first currence of orthopyroxene was experimentally deter- occurrenceof olivine formed at about 907 'C by the con- minedby Spear(1981) at794 + 8'C (1 kbar)and >805 tinued breakdown of amphibole. The culmination of 'C (2 kbar) by reaction ofhornblende with plagioclasefor metamorphism at Morton Passwas at 933 "C when am- SiOr-undersaturatedrocks. In quartz-bearing rocks, or- phibole was still a major constituent of the rock. thopyroxenewas determinedto have formed at <145'C (1 kbar)(Spear, 1981) and at760"C (l kbar)and 780'C AcrNowr-BocMENTs (2.7 kbar) (Binns, 1968)by reaction of hornblendeand This study was a follow-up on a master's thesis prepared under the quartz.Spear (1981) reported olivine to occur at 820 t guidanceand support ofD. H. Lindsley in the Earth and SpaceSciences 15 (l kbar) also by reaction of hornblendewith pla- Department of the StateUniversity of New York at Stony Brook Thanks "C go gioclase. also to B R. Frost and J. A. Grant for their advice and discussionin Theseexperimentally determined stability ranges the field. J. Laird provided helpful discussionin the construction of re- of the subfacies,except for the olivine hornfels, are in action space.Reviewers F. S. Spear,J. M. Rice, and J Selversbne gave agreement (within the large uncertainties for the calcu- constructive suggestionsfor the improvement of this manuscript. I am lated temperatures)with those at Morton Pass. indepted to P. I. Nabelek for his patienceand guidancein the consum- mation ofthis project. This project was funded in part by NSF grant EAR- CoNclusrous 8618480to D. H. Lindsley. The primary effect of isochemical contact metamor- RnrnnnNcns crtro phism of mafic schist in Morton Pass, Wyoming, was the Abbofi, R.N., Jr (1982) A petrogeneticgrid for medium and high grade changein abundanceand composition of hornblendewith metabasites.American Mineralogrst,67, 865-876. 548 RUSS-NABELEK: AMPHIBOLE REACTIONS IN MAFIC SCHIST

Albee, A.L, and Ray, L. (1970) Correction factors for electron robe mi- geenschist-eclogiterelations. Mineralogical Society of America Re- croanalysisof silicates, oxides, carbonates,phosphates, and sulfates. viewsin Mineralogy,98, 138-159 Analytical Chemistry, 42, 1408-t4l 4. Laird, J., and Albee, A.L. (1981a)High pressuremetamorphism of mafic Apted, M.J., and Liou, J.G. (1983) Phase relations among greenschist, schist from northern Yermont. American Joumal of Science,281, 97- -amphibolite,and amphibolrte in a basaltic system.American 126. Journal of Science.283-A. 328-354. - (l 98 I b) Pressure,temperature, and time indicators in mafic schist: Bence,A.E., and Albee, A.L. (1968) Empirical conection factors for the Their application to reconstructing the polymetamorphic history of electron microprobe analysisof silicatesand oxides. Journal of Geol- Vermont. American Journal of Science,28 l, 127-17 5. ogy,76,382-403. Leake, B.E. (1965) The relationship betweencomposition ofcalciferous Binns, R.A. (1968) Hydrothermal investigationsof the amphibolite-glan- amphiboleand gradeof metamorphismIn W.S. Pitcherand G.W. Flynn, ulite-facies boundary. Geological Society of Australia Special Publi- Eds., Controls of metamorphism, p. 299-318. Wiley, New York. cation 2, 341-344 Lindsley, D.H. (1983) Pyroxene thermometry. American Mineralogist, Bochensky,P. (1982) Contact metamorphism effecrsofthe Laramie An- 68,477493. orthosite Complex at Morton Pass,Wyoming. M.S. thesis, 55 p. Uni- Lindsley,D.H., and Andersen,D.J. (1983)A two-pyroxenethermometer. versity of Wyoming, Laramie, Wyoming. Proceedingsofthe I 3th Lunar and PlanetaryScience Conference, Jour- Cameron, M., and Papike,J.J. (1979)Amphibole crystalchernistry. Fort- nal of GeophysicalResearch, 88, Supplement,A887-A906 schritte der Mineralogie, 57, 28-67. Liou, J. G., Kuniyoshi, S., and Ito, K. (1974) Experimentalstudies of the Davidson, P.M., and Lindsley, D.H. (1985) Thermodynamic analysisof phaserelations betweengreenschist and amphibolite in a basaltic sys- quadrilateral pyroxenes.Part II. Model calibration from experiments tem. American Journal of Science,27 4, 613-632. and applicationsto geothermometry.Contributions to Mineralogy and Nabelek, C. R., and Lindsley, D.H. (1985) TetrahedralAl in amphibole: Petrology,91, 39G404. A potential thermometer for some mafic rocks. Geological Society of Engel,A.E.J., and Engel,C.G. (1962a)Progressive metamorphism of am- America Abstracts with Programs, 17, 673. phibolite interlayers,northwest Adirondacks, p. 37-82 GeologicalSo- Newhouse,W.H., and Hagner,A.F. (1957) Geologicalmap of anorthosite ciety of America, Buddington Volume, New York (Boulder,Colorado). areas,southeastern part of laramie Range,Wyoming. U S. Geological - (1962b) formed during proglessivemetamorphism Survey Mineral InvestigationsField Studies,Map MF I19. of , northwest Adirondack Mountains, New York. Geo- Papike, J.J., Cameron, K.L., and Baldwin, K. (1974) Amphiboles and logical Society of America Bulletin, 73, 1499-1514. pyroxenes: Characterization of other than quadrilateral components Essene,E J. (1982) Geologic thermometry and barometry. Mineralogical and estimatesof ferric Fe from microprobe data. GeologicalSociety of Societyof America Reviewsin Mineralogy,10, 153-196 America Abstracts with Programs,6, 1053 Ferry, J.M. (1984) Phasecomposition as a measureof reaction progress Robinson, Peter. (1982a) Metarnorphosedigneous rocks at high temper- and an experimental model for the high-temperaturemetamorphism ature: The breakdown of amphiboles.Mineralogical Society ofAmerica of mafic igneousrocks. American Mineralogist, 69, 677-69 l. Reviewsin Mineralogy,98, 182-206. Fowler, K.S. (1930) The anorthositearea of the Laramie Mountains, Wy- -(1982b) Amphiboles in metamorphosedcalcareous sedimentary oming. American Journal ofScience, 5th ser., 19, Part I, 305-315 and rocks Mineralogical Society of America Reviews in Mineralogy, 98, Part II, 373-403. 206-207. Fuhrman, M.L. (1986)The petrologyand mineral chemistry of the Sybille Russ, C.A. (1984) Contact metamorphism of mafic schist, I-aramie An- Monzosyenite and the role of temary feldspars Ph.D. dissertation, State orthosite Complex, Morton Pass,Wyoming. M.S. thesis, 111 p State University of New York at Stony Brook, Stony Brook, New York. University of New York at Stony Brook, Stony Brook, New York. Fuhrman,M.L., Lindsley,D.H., Russ,C.A., and Frost,B.R. (1983)Con- Russ, C.A, and Lindsley, D.H. (1984) Progradecontact metamorphism tact relations betweenferrosyenite and anorthositeof the l,aramie An- of mafic schist,Morton Pass,Laramie Mtns., Wyoming. EOS, 65, 291 orthositeComplex (LAC), Wyominig (abs.).EOS, 64,327. Spear,F.S. (1981) An experimentalstudy ofhornblende stability and Fuhrman, M. L., Frost, B.R., and Lindsley, D.H. (1988) Crystallization compositional variability in amphibolite. American Joumal of Science, conditions of the Sybille monzosyenite,Laramie Anorthosite Complex, 28r,697-734. Wyoming. Journal of Petrology, 29, 699-729. Spencer,K.J., and Lindsley, D.H. ( 198I ) A solution model for coexisting Grafl P.J.,Sears, J.W., Holden,G.S., and Hausel,W.D (1982)Geology iron-titanium oxides. American Mineralogist, 66, I 189-1201. of the Elmers Rock greenstonebelt, Laramie Range, Wyoming. Geo- Stoddard, E.F. (1985) Zoned plagioclaseand the breakdown of horn- logrcal Survey of Wyoming, Report No. 14, 23 p. blendein pyroxeneamphibolites CanadianMineralogist, 23, 195-204. Hammerstrom, J M , and Zen, E. (1986\ Aluminum in homblende: An Stormer, J C. (1975) A practical two-feldspargeothermometer. American empirical igneousgeobarometer. American Mineralogist,7 1, 1297-1313. Mineralogist, 60, 667-674. Harte, B., and Graham, C M (1975) The graphicalanalysis of greenschist Thompson, A.B. (1976) Mineral reactionsin pelitic rocks: I. Prediction to amphibolite facies mineral assemblagesin metabasites.Journal of of P-T-X (Fe-Mg) phaserelations. American Journal of Science,276, Petrology,16, 347-37O. 401424. Hynes, A. (1982) A comparison of amphiboles from medium- and low- Thompson, J.8., Jr. (1982) Reaction space:an algebraicand geometric pressuremetabasites. Contributions to Mineralogy and Petrology, 81, approach. Mineralogical Society of America Reviews in Mineralogy, 1r9-r25. 10,33-52. Klugman, M A. (1966)Resume of the geologyof the Laramie Anorthosite Thompson,J.B., Jr., Laird, J., and Thompson,A.B (1982)Reactions in Mass. Mountain Geologist, 3, 7 5-84. amphibolite, greenschistand .Joumal of Petrology, 23, l- Laird, J. (1980) Phaseequilibria in mafic schist from Vermont. Joumal 27. of Petrology, 21, 1-37. van der Plas, L., and Tobi, A.C. (1965) A chart forjudging the reliability -(1982a) Amphiboles in metamorphosed basaltic rocks: Green- of point counting results.American Journal of Science,263, 87-90. schist facies to amphibolite facies. Mineralogical Society of America Reviewsin Mineralogy,98, 113-137. MnNuscnrpr RECEIVEDMencH 18, 1988 - (1982b)Amphiboles in metamorphosedbasaltic rocks: Blueschist- MlNuscnrrr AccEnED Jaruanv 20, 1989