bastite pseudomorphs after orthopyroxene, clinopyroxene ...rruff.info/uploads/cm17_729.pdf ·...

12
Canadian Minzralogist Vol. 17, pp.729-74A 0979) BASTITE PSEUDOMORPHS AFTER ORTHOPYROXENE, CLINOPYROXENE AND TREMOLITE MICHAEL A. DLINGAN Department of Geological Sciences, Southern Methodist University, Dallas,Texas 75275, U.S.A. ABsTRAc"r Intergrowths consisting almost exclusively of lizardite platelets with bastite texture are pseudo- morphio after orthopyroxene, clinopyroxene and tremolite, from which they are derived by topo- tactic reaelion, Some bastites have inherited the hieh Al content of the original pyroxene, which may account for the predominance of lizardite in pyroxene bastites. Serpentine-dehydration reactions during progressive metamorphism of serpentinites explain why antigorite is the stable serpentine form in the presence of tremolite, forsterite and diopside. The low-temperature replacement of these phases by lizardite is in contradiction with some of the observed reactions. Phase relations in aluminous systems remain controversial. Secondary diopside is sporadically associated with bastites after both tremolite and clinopyroxene. Textural relations sug. gest that this diopside nucleated during the forma- tion of associated bastites. The assemblage li?;;r- dite * diopside as a product of serpentinization may represent a metastable analogue of the as- semblage diopside + antigorite. Sourraernp Des intercroissances consistant presque exclu- sivement de plaquettes de lizardite i texture de bastite sont pseudomorphes de l'orthopyroxdne, du clinopyroxBne et ds la tr6molite, dont elles pro- viennent par r6action topotactique, Certaines bas- tites ont h6rit6 leur haute teneur d'aluminium du pyroxdno originel, ce qui expliquerait la pr6domi- nance do la lizardite dans ces bastites. Les r6ac- tions de d6shydratation de la serpentine pendant le m6tamorphisme progressif des serpentinites rendent compte du fait que l'antigorite soit la forme stable en pr6sence de tr6molite, forst6rite et diopside. Le remplacement ?r.basse temp6rature de ces phases par la lizardite serait incompatible avec certaines des r6actions observ6es: les relations de phases dans les systdmes alumineux restent sujettes i controver$e. Le diopside secondaire associ6 aux bastites qui remplacent tr6molite et clinopyroxlne serait, d'aprds les relations texturales, contemporain des bastites. Les assemblages d lizardite et diopside, en tant que produits de serpentinisation, marquent peut-etre l'analogue m6tastable de I'association anti gorite + diopside. (traduit par la R6daction) INrnopucTtoN This paper presents petrographic descriptions of bastites after orthopyroxene, clinopyroxene and tremolite, and new data on the occurrenceof secondary diopside formed in association with the serpentinization of clinopyroxene and tre- molite. The purpose of this paper is to combine these petrographic observations and derive some constraints on the reactions involved in bastite formation. Microprobe analysesof various ser- pentine minerals and pseudomorphs, including bastites, are presented in a companion paper (Dungan 1979). Investigationsof serpentinepetrology reported here and in accompanying papers @ungan 1977, 1979) are an outgrowth of a broader $tructural and petrological project concerning ultramafic and mafic rocks in the Sultan area, North Cascades, Washington (Johnson et al. 1977, Vance & Dungan 1977, Yance et al. l98O). However, the mineralogical portion of the orig- inal project on serpentines was expanded to include samples of ultramafic rock from several additional localities. These localities are the Darrington (Vance 1972) and Weden Creek (Heath 1972) peridotites, which crop out in a belt adjacent to the Sultan bodies and the Feather River peridotite, northern Sierran foothills, California (Hiet- anen 1951, 1973, Ehrenberg 1975), Figure I is a generalized geological map of the Dar- rington-Sultan-Weden Creek area. The meta- morphic history of the Darrington-Sultan peri- dotiies and metaserpentinites is discussed in detail in Vance & Dungan (1977). PnEvlous Wom or''l BASTITBS Haidinger (1845) defined the term bastite as a serpentine pseudomorph after grthgpy' roxene, anA is usage is similarly restricted in the Glossary of Geology (Gary et al. 1972). However. bastite also has been used to describe pseudomorphs after clinopyroxene T4- amphi- 6ole (Winchell & Winchell 1951, Klinkhammer 1962. Hochstetter 1965). I concur with Wicks & Whittaker (1977), who Proposed that the term ba.stite be expanded to include a prefix indicating the identity of the parent mineral, 729

Upload: vandieu

Post on 28-Apr-2018

215 views

Category:

Documents


1 download

TRANSCRIPT

Canadian MinzralogistVol. 17, pp.729-74A 0979)

BASTITE PSEUDOMORPHS AFTER ORTHOPYROXENE,CLINOPYROXENE AND TREMOLITE

MICHAEL A. DLINGANDepartment of Geological Sciences, Southern Methodist University, Dallas, Texas 75275, U.S.A.

ABsTRAc"r

Intergrowths consisting almost exclusively oflizardite platelets with bastite texture are pseudo-morphio after orthopyroxene, clinopyroxene andtremolite, from which they are derived by topo-tactic reaelion, Some bastites have inherited thehieh Al content of the original pyroxene, whichmay account for the predominance of lizardite inpyroxene bastites. Serpentine-dehydration reactionsduring progressive metamorphism of serpentinitesexplain why antigorite is the stable serpentine formin the presence of tremolite, forsterite and diopside.The low-temperature replacement of these phasesby lizardite is in contradiction with some of theobserved reactions. Phase relations in aluminoussystems remain controversial. Secondary diopsideis sporadically associated with bastites after bothtremolite and clinopyroxene. Textural relations sug.gest that this diopside nucleated during the forma-tion of associated bastites. The assemblage li?;;r-dite * diopside as a product of serpentinizationmay represent a metastable analogue of the as-semblage diopside + antigorite.

Sourraernp

Des intercroissances consistant presque exclu-sivement de plaquettes de lizardite i texture debastite sont pseudomorphes de l'orthopyroxdne, duclinopyroxBne et ds la tr6molite, dont elles pro-viennent par r6action topotactique, Certaines bas-tites ont h6rit6 leur haute teneur d'aluminium dupyroxdno originel, ce qui expliquerait la pr6domi-nance do la lizardite dans ces bastites. Les r6ac-tions de d6shydratation de la serpentine pendant lem6tamorphisme progressif des serpentinites rendentcompte du fait que l'antigorite soit la forme stableen pr6sence de tr6molite, forst6rite et diopside. Leremplacement ?r. basse temp6rature de ces phasespar la lizardite serait incompatible avec certainesdes r6actions observ6es: les relations de phasesdans les systdmes alumineux restent sujettes icontrover$e. Le diopside secondaire associ6 auxbastites qui remplacent tr6molite et clinopyroxlneserait, d'aprds les relations texturales, contemporaindes bastites. Les assemblages d lizardite et diopside,en tant que produits de serpentinisation, marquentpeut-etre l'analogue m6tastable de I'association antigorite + diopside.

(traduit par la R6daction)

INrnopucTtoN

This paper presents petrographic descriptions

of bastites after orthopyroxene, clinopyroxeneand tremolite, and new data on the occurrence ofsecondary diopside formed in association withthe serpentinization of clinopyroxene and tre-molite. The purpose of this paper is to combinethese petrographic observations and derive someconstraints on the reactions involved in bastiteformation. Microprobe analyses of various ser-pentine minerals and pseudomorphs, includingbastites, are presented in a companion paper(Dungan 1979).

Investigations of serpentine petrology reportedhere and in accompanying papers @ungan 1977,1979) are an outgrowth of a broader $tructuraland petrological project concerning ultramaficand mafic rocks in the Sultan area, NorthCascades, Washington (Johnson et al. 1977,Vance & Dungan 1977, Yance et al. l98O).However, the mineralogical portion of the orig-inal project on serpentines was expandedto include samples of ultramafic rock fromseveral additional localities. These localitiesare the Darrington (Vance 1972) andWeden Creek (Heath 1972) peridotites, whichcrop out in a belt adjacent to the Sultanbodies and the Feather River peridotite,northern Sierran foothills, California (Hiet-anen 1951, 1973, Ehrenberg 1975), FigureI is a generalized geological map of the Dar-rington-Sultan-Weden Creek area. The meta-morphic history of the Darrington-Sultan peri-dotiies and metaserpentinites is discussed indetail in Vance & Dungan (1977).

PnEvlous Wom or''l BASTITBS

Haidinger (1845) defined the term bastiteas a serpentine pseudomorph after grthgpy'roxene, anA is usage is similarly restricted inthe Glossary of Geology (Gary et al. 1972).However. bastite also has been used to describepseudomorphs after clinopyroxene T4- amphi-6ole (Winchell & Winchell 1951, Klinkhammer1962. Hochstetter 1965). I concur with Wicks& Whittaker (1977), who Proposed that theterm ba.stite be expanded to include a prefixindicating the identity of the parent mineral,

729

DARRIIIICTON

S[,LTAN

730

i.e., opx-bastite, cpx-bastite and amphibole-bas-tite or tremolite-bastite. Lizardite, the predom-inant mineralogical constituent of bastites, ischaracterized by variable orientations betweenits optic and crystallographic axes, resulting inoccurrences of both length-fast and length-slowvarieties (Wicks 1969, Wicks & Zussman 1975).Serpentine fibres with positive and negativeelongation have been referred to as y- and a-serpentine, respectively. Although Wicks &Zussman (1975) have demonstrated that thesedistinctions do not necessarily have mineralogicalsignificance, they have been adopted as petro-graphic terms (Francis 1956, Coats 1968, Wicks1969) and are used in this paper in a strictlydescriptive sense.

Recent studies by Wicks (L969), Wicks &Whittaker (7975, 1977) and Wicks & Zussman(L975) presented the most complete and de-tailed data concerning the mineralogy of ser-pentine textures. This paper draws heavily onthe results of these studies, particularly theevidence that retrograde bastites are compriseddominantly of lizardite and that submicroscopiclizardite platelets in bastites are oriented withrespect to the parent phases Petrographic de-scriptions of bastites in Wicks & Whittaker(L977) are in essential agreement with obser-vations made in this study. Wicks & Whittaker(L977) also included a complete list of refer-ences on bastite petrography and mineralogy.These will not be repeated here, since the more

THE CANADIAN MINERALOGIST

F;-1 TERTIARYGRANODIoRITEU A''IDOUARTZMONZONITE

f7m DIABASE{3PABRO COMPLEXW OFJUMBOMOUNTAINAREA

L SERPENTINFE,PERIMTITEI WITH L@AL GABARGTVEHRLITE

;:3 TERTIARY CONTINENTAL cLASTlcu ANDVOLCANTCROCKS

$ srnrenrrnrrEMELANGE

PRE.TERTIARYPHYLLITE ON THE EAST;

U SLATE.GBAYIIACKE,CHERTI \ O N T H E W E S T

\\ rrulrs

0 5 l 0 l S k m

r--1--------rEt-f-------l o'm'\ s | |

I wegrNold I

V-/rrf 30.

GENERALIZED GEOLOGIC MAP OF THE DARRINGTON€ULTANAREA, NOFTH CASCADE MOUNTAINS. WASOIINGTON

Ftc. 1. Generalized geological map of the Darrington-Sultan area, Wash-ington. Ultramafic rocks occur in two en ichelon belts emplaced alonghigh-angle faults and locally are intruded by Tertiary plutons.

recent studies by Wicks and others either in-corporate or supersede the earlier works.

PernocnerHY oF THE BAsrrrEsOpx-bostites

Those Sultan serpentinites that have not beendeformed or affected by the pre-emplacementor contact metamorphic events exhibit well-preserved cumulus textures, with orthopyroxeneoccurring as an intercumulus and cumulusphase. The serpentine pseudormorphs that re-place orthopyroxene in the Darrington and Sul-tan peridotites are similar to opx-bastites de-scribed from many localities. The dominantfeature of the internal structure of opx-bastitesis the uniform replacement of the parent pyrox-ene by 7-serpentine, which parallels the [110]pyroxene cleavage (Figs. 2b. c). Typically, opx-bastites appear as smooth, featureless "plates"of 7-serpentine of uniform birefringence andextinction. Wicks & Zussman (1975) reportedthat the submicroscopic lizardite platelets thatcomprise opx-bastites exhibit only minor varia'tions in orientation. Most opx-bastites are freeof included magnetite and are colorless in plane-polarized light (Figs. 2b, c). Secondary ser-pentine that has replaced "primary" opx-bastiteis present in some ultramafic rocks in the Sultanand Darrington areas. Bastites in some rocksthat have been thermally upgraded are partlyto completely replaced by antigorite or talc *chlorite.

BASTITE PSEUDOMORPHS 73r

Cpx-bastites

The layered ultramafic cumulates in theSultan area are generally clinopyroxene-rich.The most abundant rock-types are wehrlite andclinopyroxene-rich lherzolites, but all gradationsto olivine-poor cumulates, including websterites,are present. Clinopyroxene typically occurs asa cumulus phase modified by adcumulus growth,resulting in large, branching clinopyroxenecrystals (up to 1 cm) that partly or completelyenclose smaller olivine grains in a subpoikilitictexture.

The majority of the cumulus ultramafic rocksin the northern Sultan area contain clinopyrox-ene that is unaltered except for the incipientdevelopment of diallage and the attendant for-mation of clear diopside overgrowths (Fig. 3).However, extensive sampling has revealed severalexamples of partial and complete serpentinepseudomorphs after clinopyroxene. The partialpseudomorphs (Figs. 2a, d) were extremelyuseful in establishing textural criteria for therecognition of cpx-bastites in completely ser-pentinized rocks.

Excep! in a few localities, ultramafic cu-mulates in the southern Sultan area are com-pletely serpentinized. Based on textures in part\replaced clinopyroxenes, the presence of abun-dant clinopyroxene in the parent rocks is in-ferred.

As noted by Wicks & Whittaker (1977), cpx-bastites (Figs. 2d, e) do not develop into therelatively featureless plates of 7-serpentine thatare characteristic of opx-bastites (Figs. 2b, c).Rather, the cpx-bastites almost invariably con-sist of low-birefringence intergrowths of 7- anda-serpentine (lizardite). These intergrowths maybe isotropic, but more commonly they consistof fine-grained serpentine intergrowths with apreferred orientation parallel to the cleavage ofthe parent clinopyroxene (Figs. 2c, d, e, f).

Additional differences between opx- and cpx-bastites are present in the Sultan serytentinites.Magnetite occurs with many cpx-bastites, bothas rectangular networks intergrown with thelizardite and as rims surrounding partly ser-pentinized clinopyroxenes (Fig. 2a). Althoughmagnetite is present in some opx-bastites asdisseminated blebs (Fig. 2c) or linear segrega-tions, it rarely develops the extensive networkcharacteristic of cpx-bastites. Finally, whenviewed in plane light the cpx-bastites typicallyexhibit a pale-green or yellowish-green color incontrast to the colorless opx-bastites. Takentogether, the textural criteria discussed aboveserve to distinguish cpx- and opx-bastites wherethev occur in the same thin section. However,

both types of pseudomorph encompass sometextural variability; for example, in a significantnumber of samples in which both opx-bastiteland cpx-bastites occur, magnetite is not exten-sively intergrown with the cpx-bastites and mayeven be more abundant in the opx-bastites.Some of the variability may be the result ofthe prograde effects documented by Wicks &Plant (1979).

Tremolite-bastites

Tremolite occurs in two textural habits inolivine-rich metaserpentinites. The typical texturein statically recrystallized rocks, such as theWeden Creek peridotite, consists of radiatingsheaves of tremolite needles piercing granular,subequant olivine or olivine * talc or olivine *antigorite. A second mode of occurrence ischaracterized by euhedral to subhedral tremolitegrains, ranging in shape from stubby to highlyelongate (length:width ) 1O:1) in a matrix ofolivine. These amphiboles define a minerallineation in several of the Feather River sam-ples, suggesting that this second habit is favoredby syntectonic recrystallization. Serpentinizationof tremolite in the Weden Creek and theFeather River peridotites is widespread; it isinterpreted as entirely retrograde in origin.

Contact-metamorphosed ultramafis rocks inthe southern Sultan Complex are generallycharacterized by the assemblage forsterite *talc f tremolite. Serpentinization of this as-semblage is advanced in some areas. The trem-olite in these rocks is typically fine-grained,but two excellent examples of coarse-grained( ) 5 mm), partly and completely serpentinizedtremolite were found (Fig. ab).

Amphibole-bastites are readily distinguishedfrom pyroxene-bastites on the basis of the dis-tinctive habit of the parent amphibole. However,this criterion will not distinguish bastites aftertremolite from those after anthophyllite. Twotextural varieties of serpentine are present intremolite-bastites, and their sequence of forma-tion and relative abundances depend on severalfactors, including: (l) the prominence of the(O01) parting in the amphibole, (2) the habitof the amphibole and (3) the degree of ser-pentinization of the matrix olivine. The serpenti-nization of tremolite is usually initiated by apseudomorphic replacement of the amphiboleparallel to its cleavage and along grain boun-daries (Fig. 4a). The erowth of 7-serpentine(usually lizardite) along the amphibole cleavageplanes results in bastite textures that tend toresemble opx-bastites rather than cpx-bastites(Figs. 4b-h). The stage of initial replacement

Frc. 2. (a) Sample O4-L24, northernSultanComplex. Clinopyroxene partly serpentinized to lizardile ast-serpentine along cleavage planes. Note the abundance of maguetite associated with the cpx-bastites.Partly crossed polars. Scale in mm (as in b to f). (b) Sample 7l-109, southern Sultan Complex. Opx-bastite (on left) is composed of a nearly featureless plate of ?-serpentine. Two cpx-bastites exhibit typicalintergrofih of relatively high-birefringence ?-serp3ntine with isotropic serpentine. Bastites are set in amatrix of mesh-textured lizardite. Opx-bastite is cut by a shear zone of secondary serpentine. Magnetiteassociated \pith either t1rye of pseudomorph is nearly lacking in this area. Crossed polars. (c) Sample7L-279, serpentinized clinopyroxene-rich websterite, southern Sultan Complex; z-serpentine opx-bastite inthe centre of the picture is surrounded by cpx-bastites. Magnetite is more abundant in opx-bastites thancpx-bastites in this sample. Post-serpentinization deformation has caused a set of slip surfaces to devel-op in opx-bastite normal to ?-serpentine fibre-direction. Elongate prisms of diopside (D) jut into tleopx-bastite from margins of several cpx-bastites. Partly crossed polars. (d) Sample 7l--268, I-argesubpoikilitic cpx-bastites include abundant magostite as blades parallel to relict clinopyroxene cleavage

732

BASTITE PSEI'DOMORPHS 733

is usually advanced before the surrounding olivinegrains exhibit more than incipient serpentiniza-tion. The degree of serpentinization of tremolitegrains within a single'thin-section varies, al-though in rocks in which the olivine is totallyserpentinized, tremolite is also generally com-pletely replaced. Where the (001) parting inthe parent amphibole is particularly well devel-oped, replacement along the parting occurs inthe form of cross-veins consisting of 7. orcy-serpentine. Tremolite that occurs as sheavesof elongate needles rarely develops a goodparting and, therefore, is rarely cut by cross-veins.

The growth of cross-veins in tremolite-bas-tites seems to be partly a function of the degreeof serpentinization of the matrix olivine. Theinitial stages in the serpentinization of olivineare almost invariably represented by the growthof cross-fibre veinlets along the orthogonalparting of the olivine. In rocks that containrelatively coarse-grained olivine and large tre-molite euhedra (e,9,, Feather River peridotites),the early vein-forming stage in olivine is re-flected in the amphibole-bastites by the growthof cr-serpentine (or very rarely, 7-serpentine)veinlets normal to the amphibole cleavage.These cross-fibre veinlets have a fibre directionnormal to the long dimension of the vein. Thecross-veins in the tremolite-bastites are com-monly extensions of serpentine veinlets in theadjacent olivines. Where the veinlets are welldeveloped and closely spaced a pseudomeshtexture tends to form. This sonsists of orthogonalareas of y-serpentine bastite, with or withouta central core of unreplaced tremolite, whoseboundaries are defined by the a-serpentine veins(Figs. 4d, e). Despite the superficial resem-blance of the cross-veined tremolite-bastites tothe classic olivine mesh-texture, the timing offormation of the veins relative to the inter-vening "mesh centres" may not be as rigidlyprescribed in tremolite-bastites. Whereas cross-veins may be early and may directly replacetremolite, the mode of vein formation may alsobe secondary in that some veins seem to replaceearlier formed 7-serpentine rather than theparent mineral. The replacement of tremolite-bastites by "secondary" serpentine is furtherillustrated by the occurrence of irregular-shapedpatches of a-serpentine after 7-serpentine. Thesepatches and veins are more abundant in com-

Ftc. 3. Secondary bladed diopside (d) rims onprimary clinopyroxene (cpx). Note that elongatediopside cross-cuts adjacent opx-bastites.

pletely serpentinized rocks.An additional textural variety of serpentine

sporadically associated with tremolite-bastites isa rim of 7-serpentine that occurs as radiatingbundles oriented roughly normal to the grainboundaries of the bastite (Figs. 4e, f). Thetiming of formation of these rims is unclear,as is their textural significance.

' SecoNoanv DtopsrprPeters (1968) recognized that diopside formed

during the metamorphism of serpentinitesis compositionally distinct from relict, primaryclinopyroxenes in alpine peridotites. Evans &Trommsdorff (1970) subseguently demonstratedthat diopside * antigorite is a stable assemblagein greenschist-facies metaserpentinites. Second-ary diopside occurs as rims surrounding dial-lagic clinopyroxene (Fig. 3) and in associationwith pyroxene-bastites as rims and discretegrains. Secondary diopside has been identifiedpositively or tentatively in several of the rockscontaining tremolite-bastite from both theFeather River and the Sultan-Darrington areas.

planes. Matrix mesh-textured lizardite has recrystallized to antigorite. Partly crossed polars. (e) Sample7l-125. A small poikilitically enclosed clinopyrox:ne and the surrounding orthopyroxene are bothpseudomorphed with typically contrasting results. The high-birefringent opx-bastite (magnetite-free)contrasts with the low-birefringcnt intergrowth (magnetite is present) characteristic of cpx-bastites.Crossed polars. (f) Sample 7l--268. Typical cpx-bastites composed of high-birefringence lizardite as?-serpentine in a matrix of low-birefringence to isotropic lizardite. Crossed polan.

Ftc. 4. Photomicrographs of tremolite-bastite: (a) Sample WC #4. lncipient replacement of tremolite(dark) by ?-serpentine (light) along tremolite cleavage planes. Slightly crossed polan. Scale in mm(as in b to b). (b) Sample 7l--265, southern Sultan Complex. Tremolite-bastites after large elongate tre-molite euhedra in a matrix of antigorite. The dominant mode of replacement is t-serpentine parallel toremolite cleavage. The relict (001) tremolite parting is preserved in several bastites, but cross-veinsare not well developed. Crossed polars. (c) Sample 22N-10. Tremolite-bastite in a matrix of me$h-tex-tured lizardite and sparse relict olivine (very bright areas). Gamma-serpentine (including extremely fine-grained opaque material) dominates over small, bipartite cross-veins that are free of magnetite. Figure5e shows a detail of small, secondary diopside grains intergrown with this bastite. Crossed polan. (d)Sample 8-2. Feather River. Completely serpentinized tremolite composed predominantly of r-serpentinewith varying amounts fine-grained disseminated magnetite. Prominently developed are several cross-veins of a-se4rentine containing a thin central parting of magnetite. Bastite is also rimmed by t-ser-

734

BASTITE PSEUDOMORPHS 735

luru cEr cpr cpr cpt tr6o. 8o!p. cpr

",r to; 0.@ 0,@ o.@ n.d. 0.02 ! .d. o.o?-0.28A l , d 2 0 . 0 1 0 . @ 0 . 0 1 0 . m 0 . 1 8 o . t o 1 . 3 { . 6cr;oi 0.08 0.06 0.09 n.d. o.m o.@ o. l9-0.82tuo

- 18.2 lE.3 18.3 r8. l 23.6 {1.0800' 0.71 l . ! t .0 0.81 t ,3 1.6 2.24.2cd 25.4 25.5 25.6 23,2 12.8 !.d.m o. lE 0.15 0.21 ! .d. 0.07 0.14 0.10-0.1!N q o 0 . 0 2 0 . @ 0 . 0 2 0 . @ 0 . @ ! . d . o . l r { . 2 9d l o o . d . o , d . . . d . ! . d . 0 . 0 1 0 . 0 1

T o t a l * , . * l ; ; ; ;

TAStr I. I{CMPROBS AIIALYSES OF SECOilDARY |)IOPSIDEultramafic rock from the southern Sultan Com-plex that contains narrow bands of coarse-grained tremolite up to 0.8 cm in a matrix offine-grained serpentine (antigorite?) and mag-netite. The tremolite grains are partly serpentin-ized and include randomly oriented diopsiderhombs. Somewhat irregular, elongate patches ofan isotropic phase with high relief, presumablyhydrogrossular, are also present. Microprobeanalyses of the tren'rolite, diopside and tremo-lite-bastite are listed in Table 1.

Sample 22N-10 (Figs. 4c, 5e, f) is an olivine-tremolite-chlorite rock from the Feather Riverperidotite. It contains completely serpentinizedtremolite in a matrix of olivine and mesh-textured lizardite (3O:70). The tremolite-bas-tites contain small clusters of radiating diopsiderhombs. The diopside has nucleated on themargins of the bastite and on included mag-netite grains and has grown at high angles tothe 7-serpentine that formed parallel to thetremolite cleavage. The scattered clusters ofsecondary diopside comprise about 5Vo of thetremolite-bastites in which they occur.

Secontlary cliopside associatecl with cpx-bastites

Sample 7I-268 is a serpentinite from thesouthern Sultan Conrplex that consists of ap-proximately 507a cpx-bastites in a matrix ofpartly recrystallized mesh-textured lizardite,carbonate and secondary diopside (Figs. 5b, c).The diopside occurs most abundantly as clustersof rhombs radiating outward from the marginsof the cpx-bastites. Fibrous mattes of muchfiner grained diopside crystals are intimately in-tergrown with the rhombs. Discrete clusters ofdiopside rhombs and associated needles are alsopresent in the nratrix serpentine and as aggre-gates surrounding altered chromian spinel grains.

The secondary diopside in sample 7l-103(also from the southern Sultan Complex) issimilar to that in 7l-268 in that it is relativelycoarse-grained, abundant and forms rims around

5t 2.@t 1.993 t ,963 1.997 8.0m 4.058r t - - 0 . 0 0 7 - 0 . 0 0 2At 0.00I - 0.m3 0.@ 0.029 0.011Cr 0.@2 0.@2 0.m!ug 0.982 0.982 0.9& 0.98J 4.S6 5.135F€ 0.022 0.G0 0.031 0.025 0.151 0.121ca 0,9& 0.S6 0.989 0.986 l .m5e 0.@5 0.005 0.007 - 0.006 0.ol l[ ! o.ml - o.ml 0.0J! t - _ _ : _ : : : 0 . 0 0 1 0 . 0 0 2

toral 3.998 4.030 4.@8 4.m3 14.983 9.931

l s l . l 1 . 4 1 . 5 1 , 2ro 49.5 6.8 49.4 49.48o 49.4 49.E 49.r 49.4

:Total fe as feo"langes of varlous orides in prl@rJ clinopraoren€ tn Su)tan Conpler cmuldies(uungln, l9/4).qlqU - lndicates a concentration bolor the deielion limlt of the olcroprob€1n-<o.ol pe.cent).&!L - lndicates lhat the el@nt 6s tut analJzd.Col@ 1-4 - cat lons @lculated on the basis of 6.0 orygens.Col@ 5 - caiions alculated @ the basis of 23.0 orygens.Col@ 6 - cat ions calculated on th€ lasis of 14,0 orygeni.

Seconclary diopside associated with the lormationol tremolite-bastites

Where the secondary diopside occurs as re-Iatively large grains, it is characterized by adistinctive rhombic habit. It is distinguished fromrhomb-shaped tremolite euhedra by its higherrelief, a tendency to assume a more irregularoutline than tremolite and a lack of cleavage.Extremely fine-grained needles or aggregates ofequant diopside are present in some tremolite-bastites in the absence of the more easily identi-fied large grains. In the latter case, identificationby petrographic methods is tentative, as thegrains are too fine and too intimately intergrownto permit measurement of optical properties(Figs. 49, 4h, 5i).

Sample 71167 is a contact-metamorphosed

pentine oriented normal to bastite margin; this appears very dark in photo. Matrix consists of mesh-textured lizardite and minor relict olivine. Crossed polars. Same scale as a(f). (e) Sample 8-2. Unusualincipient tremolite-bastite adjacenttothe one shown in 5(d). Rectangular islands of tremolite (T) arecut by two types of cross-veins. Dark, offset veins of a-serpentine extend into matrix serpentine. Moreabundant bipartite cross-veins of ?-serpentine cut across tremolite with minimal offset. Early and ex-tensive vein formation is atypical. Juxtaposition of partly and completely serpentinized tremolite (e, d)demonstrates local variations in stages of development of tremolite-bastites. Crossed polars. Same scaleas 4(f), (f) Sample 8-2. Two tremolite-bastites containing variable amounts of relict tremolite. Thebastite on the left has a well-developed rim of radiating ?-serpentine with the fibre direction ori-ented roughly normal to the bastite margin. Partially crossed polars. (g, h) Sample 3-1, Feather Riverperidotite. Two photomicrographs of the same view. Plane-polarized light on the left, partly crossed polarson the right. An aggregate of stubby tremolite grains (T) in a matrix of olivine has been partly serpentinized whereas the olivine has not. The (001) tremolite parting is poorly developed and serpentinecross-veins are absent. Dark, fine-grained aggregates of secondary diopside(?) are associated with the tre-molite-bastites.

Frc. 5. Photomicrographs of secondary diopside associated with tremolite-bastites and cpx-bastites. (a)Sample 7l--268. Rhomboidal diopside and associated fine-grained needles form a rim around a cpx-bastite. Note that the diopside has nucleated on the margin of the bastite and is absent from the volumeof the parent pyroxene. Plane-polarized light. Scale in mm (as in b to i). (b) 7l-265, Detail of arhomb-needle combination intergrowth. This diopside has grown in matrix serpentine and has nucleatedon magnetite rather than the margin of a cpx-bastite. Diameter of the needles 1G-20 pm. (c) 7l--26E.Rhomb-and-needle diopside partly rimming two cpx-bastites. Bastite at top is partly replaied by carbonate.Partly crossed polars. (d) High magnification of the lower left area shown in 5(e). Note similarityto secondary diopside in Figure 3. (e) Sample 22N-10, Feather River peridotite. Small, fine-grained ag-gregates of secondary diopside (D), which have nucleatgd on the margin of a tremolite-bastite. Plane-polarized light. (f) Sample 22N-10. Detail of the interior of another tremolite-bastite in which scat-

736

BASTITE PSEUDOMORPHS 737

cpx-bastites and altered chrome spinel (Figs.59, h). There is a very strong correlation be-tween the distribution of secondary diopside andthe availability of nucleation sites provided byopaque phases. Sample 71179 is a serpentinizedwebsterite (Fig. 2c). A few of the cpx-bastitesare partly rimmed by diopside overgrowths re-sembling those illustrated in Figure 3.

Table 1 lists microprobe analyses of second-ary diopsides associated with cpx-bastites in 71-103 and 71168, They are very similar to theanalysis of diopside ftom7L-267, which formedas a result of the serpentinization of tremolite.Column 7, Table 1 summarizes the ranges ofTiOr, ALOo, CrrOs, FeO, MnO and NarO inprimary clinopyroxenes from the Sultan Com-plex cumulates (analyses taken from Dungan1974). A comparison of these values with anal-yses of secondary diopside indicates a substan-tial reduction in the amount of all the com-ponents except Mn. Microprobe analyses ofdiopside in serpentinites and metaserpentinitesfrom other localities by Peters (1968), Tromms-dorff & Evans (1972), Frost (1973) and Ehren-berg (1975) are virtually identical to those inTable 1. Taken as a whole, these data establishthat secondary diopside in serpentinites has arestricted compositional range that tends toapproach pure CaMgSLOo.

Crystallographic correspondence between thebastite-type pseudomorphs and. parent silicates

The textural relationship between bastiticlizardite and its parent silicates is very con-sistent and far less complex than in me$h ser-pentine; i.e., the bastites are predominantly com-prised of lizardite platelets that generally ex-hibit rigid parallelism to the prismatic cleavageof the parent silicates. The significance of thisorientation was demonstrated by the X-ray microbeam studies of Wicks (L969), who showedthat there was a three-dimensional crystallo-graphic correspondence between intergrownlizardite and orthopyroxene. The underlyingstructural cause for this topotactic relationshipis found in the parallelism of 4op" orrd crro, whichtranslates into parallelism of the oxygen-anion

frameworks of the two phases. That is, the slose-packed oxygens retain some structural integrityduring the conversion to lizardite, whereas thecations are free to migrate (Brindley L963).Pseudomorphism of orthopyroxene by orientedlizardite such au, // ""n*,

brt // b,n, and cu" //dop" olso yields the following correspondence be-tween unit-cell parameters: copx = arz, bop l6rro ond 3ann* = 4cn, (Wicks 1969), Wicks didnot determine crystallographic interrelationshipsbetween lizardite and tremolite or lizardite andclinopyroxene. However, the textural similaritiesamong the three bastite types suggest thatsimilar parent-bastite relations obtain.

A well-documented example of an orientedtransformation between a pseudomorphous sheetsilicate and a chain silicate has been describedby Eggleton (1975)- Here nontronite has re-placed hedenbergite topotactically and with sucha high degree of orientation that single crystalsof nontronite result. Eggleton concluded that theability of nontronite to form such perfectpseudormorphs is due to the oriented transfor-mation of 1.5 unit cells of pyroxene to one unitcell of nontronite with no volume change. Thenontronite exhibits the same crystallographicrelationship to hedenbergite as lizardite does toorthopyroxene: (Dooot // ba.r, aooor, // cr"a aodc*ot // an'd).

Pnesn RBlerroxs

Recent experimental and petrological studiesof progressively metamorphosed serpentinitehave produced new information concerning thephase relations among amphiboles, clinopyrox'ene, olivine and antigorite in these rocks (Evans& Trommsorff L97O, L974, Trommsdorff &Evans 1972, 1974, Springer 1974, Frost 1975).One important result of this work is the recog-nition that there is a lack of correspondence be-tween the serpentine mineralogy of the dehydra-tion reactions recorded in thermally upgradedrocks and retrogressive pseudomorph-formingreactions (Evans er al, L976, Dungan L977).Thecritical difference is that antigorite participatesin the'dehydration reactions and lizardite plusminor chrysotile are the constituent phases of

tered, fine-grained diopside is abundant. Many small grains are oriented parallel to ?-seryentine. TVoprominent clusters of radiating diopside (D) needles have nucleated on a magnetite segregation. Dia-meter of the needles is 5-10 p,m. Plane-polarized light. (g) Sample 7,1-103, southern Sultan Complex.Coarse-grained clusters of rhomb-shaped diopside nucleated o4 magnetite grains. (h) Sample 7l-L03.Cpx-bastites (Bast.) surrounded by secondary diopride (D). Note the concentration of diopside aroundthe bastites. Diopside is absent from sheared serplntine in upper left of photomicrograph except at itsmargins. Slightly crossed polars. (i) Sample 3-1, Feather River peridotite. Fine-grained diopside (D)associated with the serpentinization of tremolite (Trem.). Micron-diameter needles have nucleated onthe margin of the bastite and grown inward at high angles to the ?-serpentine, which has replacedthe tremolite parallel to its cleavage. This photomicrograph is a detail of Figure 4 (g) and (h). Plane-polarized light.

738 THB cANADIAN

pseudomorphs (Wicks 1969, Wicks & ZussmanL975). Until recently (Johannes L975), chryso-tile was the only serpentine phase to have beenreported in experimental reversals of serpentine'olivine equilibria (Johannes 1968, Chernosky1973). This dichotomy led Trommsdorff &Evans (1972) to postulate that natural andsynthetic chrysotile-dehydration reactions weremetastable relative to analogous antigoritedehy-dration reactions. Evans ef al, (1976) presentedan expanded treatment of this hypothesis basedin part on the recognition that chrysotile €antigorite -f brucite is a reversible, naturallyoccurnng reaction that defines the upper stabilitylimit of chrysotile. Although lizardite-antigoriteand lizardite-chrysotile phase relations are notas well documented as those between chrysotileand antigorite, many reactions in which lizarditeparticipates may be metastable, by analogy withreastions in which chrysotile is the serpentinephase (Dungan 1977). Caruso & Chernosky(1979) argue that aluminous lizardite may havea real field of stability overlapping that of low-Alantigorite. This problem is dealt with in thecompanion paper (Dungan 1979).

Retrograde reactions associated with the for-mation of the three types of bastite pseudo-morphs discussed in this paper are characterizedby contradictions with equilibrium considerationsbased on the following observations: (1) diop-side and antigorite (+ brucite) form the stableassemblage in greenschist-facies metaserp€nti-nites; (2) the reaction diop * antig ? fo *tlenx + HrO defines the upper stability-limit ofdiopside f antigorite; (3) orthopyroxene andserpentine have mutually exclusive stability fieldsand cannot be related by reversible equilibria;i.e., t}Le first appearance of enstatite in progradereaction-sequences occurs at temperatures abovethe upper stability-limit of antigorite. The oc-surrence of secondary diopside as a reactionproduct of the serpentinization of clinopyroxeneand tremolite and the textural relationships be-tween this diopside and the bastite pseudomorphsprovide additional confirmation of a low'tem-perature stability field in serpentinites.

The presence of dio'pside in association withtremolite-bastite suggests that the replacementoccurs vic the reaction fo * trem + HrO =diop * /iz. This reaction is the metastable anal-ogue of the antigorite-dehydration equilibriumgiven above. Dungan (L977) argued that lo *Hzo + Iiz * bruc is similarly metastable relativeto fo * HaO * antig * Drac. Serpentinizationof clinopyroxene in the Sultan Complex is in-ferred to occur in part because the high-tem-perature clinopyroxene of the cumulus wehrlitesis unstable at low temperatures. The equilibration

MINEMLOGIST

to a stable composition seems to lead to com'plete pseudomorphous replacement by lizarditewith or without the concomitant gowth of rimsof secondary diopside. The data of Caruso &Chernosky (1979) on aluminous $ystems demon-strate that lizardite compositions must be con-sidered in any interpretation of serpentine para-geneses. Dungan (1979) demonstrates that manypyroxene bastites are Al-rich. Tremolites inmetaserpentinites are low in Al, suggesting thattremolite-bastites are Al-poor lizardites.

TUB .RrecrtoN MoDEL

Brindley & Zussman (L957) and Brindley( 1963) demonstrated that lizardite-olivine inter-growths exhibit a tendency to be oriented mu-iually by a three-dimensional crystallographicrelationship. One interpretation of the conse'quences of this topotactic relationship is thatihe replacement of one mineral by the othertakes place by a mechanism of heterogeneousreaction in which the oxygen-anion frameworkof the parent phase is preserved and cationsare free to migrate within the structure (Brind-ley 1963). Although serpentine pseudomorphsafter olivine are complex intergrowths, datapresented by Dungan (L977) $uggest that theielative ease of nucleation of one phase on theother is an important factor in determining thetextures and mineralogy of experimental andnatural cases of serpentinrolivine reaction.

Textural relations in bastites and the X-raydata of Wicks (1969) indicate that the orienta-tion of lizardite is more strongly controlledwith respect to the parent phase in bastite thanin meshlextures after olivine. Eggleton's (1975)characterization of a strongly oriented trans-formation of hedenbergite to nontronite em-phasizes the facility with which a topotacticallybriented sheet silicate can replace a chain silicateif the unit-cell dimensions of the two phasesare closely comparable.

I propose that bastites form in much the samewayls the nontronite-hedenbergite intergrowthsdiscussed by Eggleton (1975). I also suggestthat it is tle enhanced ease of nucleation ofthe lizardite that follows from its inheritanceof the structure of the pyroxene or amphibolethat is responsible for the excursions from othermodes of replacement.

Implicit in ttre application of the proposedmechanism of reaction to bastites is the preser-vation (roughly) of constant volume during thesubstantial gains and losses of various com-ponents during the pseudomorphic serpentiniza-tion of pyroxenes and amphiboles. The reactionmodel involves two related processes' nuclea-tion and growth. Compared to the role of

BASTITB PSBUDOMORPIIS 739

nucleation, the mass balance of the growth pro-cess may be of secondary importance in deter-mining the nature of mutual intergrowths be-tween bastites and parent silicates. Constant-volume replacement requires a loss or gain ofcertain components on the scale of the pseudo-morph during serpentinization. The Si excessthat results from the replacement of orthopy-roxene by serpentine may react with brucite(previously or simultaneously produced by theserpentinization of associated olivine grains) toproduce a net gain of serpentine within theadjacent mesh pseudomorphs. Overall volumeincrease is reduced by the conversion of bruciteto serpentine. However, the effectiveness of localchemical compensation depends on parent-sili-cate volume relations that rarely will be optimal.If the hydration of enstatite were otherwise iso-chemical, serpentine alone could not be the re-placement product. The higher Si/Mg in ensta-tite indicate$ that it would be replaced by talcf serpentine at low .metamorphic grade. Theabsence of microscopic crystals of talc inter-grown with lizardite in opx-bastites indicatesthat Si in excess of the amount needed to formserpentine is expelled from the volume of theparent enstatite. The serpentinization of clinopy-roxene involves a large loss of Ca from thevolume of the parent silicate and generallydoes not occur until associated olivine andorthopyroxene are completely replaced. \ilhereabundant secondary di:opside is associated withcpx-bastites, it has replaced pre-existing matrixserpentine (usually mesh). This demonstratesthat Ca and Si expelled from the parent clino-pyroxene are balanced locally by a reversemigration of Mg into the volume of tle cpx-bastite, with little volume change. The morecommon case, in which Ca is removed entirelyfrom the area in which the clinopyroxene isreplaced, requires corresponding migrations ofSi and Mg. The occurrence of secondary diop-side entirely outside the volume of the parentclinopyroxene is a significant point in supportof the topotactic model. If nucleation effects'were not predominant, a structureless aggte-gate of diopside and lizardite might result. Thefavored interpretation of this phenomenon isthat in the initial stages of reordering of theunstable primary clinopyroxene, lizardite nuclea-tion and growth predominate to the extent thatexpulsion of Ca from the structure is complete.This effect is similar to the loss of Si duringformation of opx-bastites and also applies intremolite-bastites where secondary diopside isabsent or far less abundant than would be ex-pected from the amount of Ca that must havebeen released.

CoNcLusloNsIn summary, the three bastite-forming reac-

tions occur for three different reasons. Cpx-bastites result from re-equilibration of clino-pyroxene to a stable low-temperature composi-tion. Enstatite is replaced at temperatures wellbelow its stability field, so that the reaction isnonreversible. Tremolite-bastites are formedmetastably as part of a multiphase retrogradeassemblage by a reaction for which there is aprograde analogue. Despite these differencesand the mass transport required to producenearly monomineralic pseudomorphs by replace-ment of these three compositionally distinctparent phases, the products of low-temperaturealteration of these minerals are lizardite-bastitesthat exhibit a remarkable degree of texturalsimilarity. The petrographically observed paral-lelism and the sonsistency of orientation of thelizardite platelets with respect to structural ele-ments of the parent phases, as shown by theX-ray data of Wicks (L969), imply that thereactions are topotactic. Where reactions proceedas oriented transformations, nucleation andgrowth of newly formed phases are greatly en-hanced because of the less drastic reorganizationof the parent structure inherent in this type ofintergrowth. An important factor in the forrna-tion of bastites seems to be the ability of lizarditeto form oriented intergrowths with orthopyrox-ene, clinopyroxene and tremolite, Where theparent minerals are aluminous, lizardite stabi'lity may be real and compositional control maycontribute or even predominate (Caruso &Chernosky 1979).

AcrNowEocEMENTsThis research was initiated at the University

of Washington, Seattle, and was partly supportedby funds from NSF grant 11-2589. An earlyversion of the manuscript was reviewed byBernard Evans, Ron Frost, Bob Coleman andJudith Moody. The revised manuscript has alsobenefited by comments from Fred Wicks andJoe Chernosky. Linda Dungan, Elinor Stockon,Etta Dorman and Nazlee Coburn participated inpreparation of the manuscript at various stagesof its painful evolution.

RErnneNcrs

BnrNolsv, G.W. (1963): Crystallographic aspects ofsome decomposition and recrystallization reac-tions, 1z Prggress in Ceramic fuience (I.8. Burke,ed.), Macmillan, New York.

& ZussueN; J. (1957): A structural studyof the thermal transformation of serpentine miu-erals to forsterite. Amer. Mineral. 42, 461474.

CAnuso, L.J. & CnnnNosKY, J.V., Jn. (1979): Thestability of lizardite. Can. Mineral. L7,757-769,

740 THE CANADIAN MINERALOGIST

Cnrnxosrv, J.V., Jr. (1973): The stability of chryso-tile, MggSigO6(OH)r, and the free energy offormation of talc, Mg.SiOr(OH)2, Geol. Soc,Amer. Program Abstr. 5, 575,

Coers, C.J.A. (1968): Seupntine minerals fromManitoba. Can. Mineral. 9, 322-347.

Cor-ruaN, R.G. (1971): Petrologic and geophysicalnature of serpentinites. GeoI. Soc. Amer. Bull.82, 897-918.

DuNceN, M.A. (1974)r The Origin, Empl.acement,and Metamorphism of the Sultan Mafic-Ultra-mafic Complex, North Cascades, SnohomishCountry, llashingtron. Ph.D. thesis, Univ. Wash-ington, Seattle, WA.

(1977) : Metastability in serpentine-olivineequilibria. Amer, Mineral. 62, l0L8-1029.

(1979): A microprobe study of antigoriteand some serpentine pseudomorphs. Can. Mineral,L7, 771-7E4,

EccrntoN, R.A. (1975): Nontronite topotaxial afterhedenbergite. Amer. Minerar. 60, 1063-1068.

EHnENrnnc, S.N. (1975): Feather River ultramaficbody, northern Sierra Nevada, California. Geol.Soc, Amer. Bull. 86, 1235-1243.

EveNs, B.W., JoHeNNns, W., OrERDoorra, H. &TnouusoonrF, V. (1976): Stability of chrysotileand antigorite in the serpentinite multisystem.Schweiz, Mineral, Petrog. Mitt. 56, 79-93.

& Tnouusoonrr, V. (1970): Regionalmetamorphism of ultramafic rocks in the centralAlps: paragenesis in the system CaO-MgO-SiOgHrO. Schweiz. Mineral. Petrog. Mitt.50, 481-492,

& - (1974): Stability of enstatite *talc, and CO2-metasomatism of metaperidotite,Val d'Efra, Lepontine Alps. Amer. L Sci. 274'274-296.

FneNcrs, G.H. (1956)l The serpentinite mass inGlen Urquhart, Inverness-Shire, ScoIland, Amer,,1. Sci. 254, 201-226.

Fnosr, B.R. (1973): Conlact Metamoryhisrn of theIngalls Ultramalic Complex at Paddy-GoEasyPass, Central Cascad,es, Washington. Ph.D. thesis,Univ, Washington, Seattle, WA.

(1915): Contact metamorphism of serpenti-nite, chlorite blackwall and rodingite at Paddy-Go-Easy Pase, Central Cascades, Washington.l. Petrology L6, 272-313.

Genv, M., McAru, R., Jn. & Wor.r, C.L., eds.(1972): Glossary of Geology, Amer. Geol. Inst.,Washington, D.C.

HATDINGER, W. ( 1845) : NaturwissenschaltlicheAbhandlungen Vols. 14, IYien.

HEAT{, M. (1971): Bedrock Geology of the MonteCristo Area, Northern Cascades, lilashington,Ph.D. thesis, Univ. Washington, Seattle, WA.

IITETANEN, A. (1951): Metamorphic and igneousrocks of the Merrimac area, Plumas NationalForest, California, Geol, Soc. Amer. Bull. 62,565-608.

(1973): Geology of the Pulga and BucksLake quadrangles, Butie and Plumas Oounties,California. U. S. Geol. Surv. Prof. Pap. 73L.

HocHsrErrER, R. (1965): Zur Kenntnis der Ser-pentinmineralien. Ph.D, thesis, Univ. Saarlandes,

Saarbriicken, Germany.JoueNNrs, W. (1968): Experimental investigation

of the reaction forsterite + HrO * serpentine+ brucite. Cont. Mineral. Petology 19, 309-315.

(1975): Zur Synthese und thermischenStabilitiit von Antigorit. Fortschr. Mineral. 58,t-36.

JoHNsoN, M., DuNcer, M.A. & Vescr, JA. (1977):Stable isotope compositions of olivine and dolo-mite in peridotites formed by deserpentinization,Darrington area, North Cascades, Washington.Geochim. Cosmochim, Acta 4L, 43t435.

KLTNKHAMMER, B.F. (1962): Ultrabasite des Ost-bayrischen Grenzgebirges. Ph.D. thesis, Univ.Saarlandes, Saarbriicken, Germany.

PETERs, T. (1968): Distribution of Mg, Fe, Al, Caand Na in coexisting olivine, orthopyroxene andclinopyroxene in the Totalp serp€ntintite (Davos,Switzerland) and in the alpine metamorphosedMalenco serpentinite (N. Italy). Contr. Mineral.Petrology 18, 65-75.

SrnrNcnn, R.K. (1974): Contact metamorphosedultramafic rocks in the western Sierra Nevadafoothills, California. I. Petrology 15, 160-195.

TRoMMsDoRFF, V. & EveNs, B.W. (L972): Pro-gressive metamorphism of antigorite schist in theBergell tonalite aureole (Italy). Amer. J. Sci.272,423-437.

& - (1974): Alpine metamorphismof peridotitic rocks. ^ScAueiz. Mineral, Petrog.Mitt. 54,333-352.

VANcE, LA. (1972): Mid-Tertiary tectonic empla-cement of metamorphic peridotites on the Dar-rington area, Cascade Mountains, Washington.Geol. Soc. Amer. Program Abstr. 4, 253,

& DuNceN, M.A. (t977): Formation ofperidotites by deserpentinization in the Darring'ton area, Cascade Mountains, Washingfon. Geol.Soc. Amer. Bull. 88, 1497-L508,

DuNceN, M.A., Br-eNcneno, D.P. & RHoDEs,J.M. (1980). Tectonic setting and trace elementgeochemistry of Mesozoic ophiolitic rocks inwestern Washington. Amer. J. Sci. 280A, 359-388.

Wrcrs, F.J. (969): X-ray and Optical Srudies on' Serpentine Minerals. D. Phil. thesis, Oxford Univ.,

Oxford, England,& Pr-eNr, A.G. (1979): Electror-micro-

probe and X-ray-microbeam studies of serpentinetextures, Can. Mineral. 1?, 785-830.

& WHnTAKER, E.J.W. (1975): A reappraisalof the structures of serpentine minerals. Car.Mineral. L3,227-243,

& - (1977): Serpentine textures andserpentinization, Can, Mineral, L5' 459-48E.

& ZussIvreN, J. (1975): Microbeam x-raydiffraction patterns of the serpentine minerals.Can. Mineral. L3, 244-258.

WrNcnr,r.r., A.N. & WrNcrnu., M. (1951): Elementsof Optical Mineralogy II. John Wiley & Sons,New York.

Received October 1978, revised manuscript acceptedOctober 1979,