Orientation of Pigeonite Exsolution Lamellae in Metamorphic Augite ...

American Mineralogist, Volume 60, pages 9-28, 1975

Orientation of Pigeonite Exsolution Lamellae in MetamorphicAugite: Correlation with Composition and Calculated

Optimal Phase Boundaries

Howano W. Jerre. Perrn RosrNsoN. Ronrnr J. Tnacv,

Department ol Geology, Uniaersity of Massachusetts,Amherst, Massachusetts 0 1002

eNn Mnrcor,u Ross

U.S. Geological Suraey, Reston, Virginia 22092

Abstract

Optical examination of metamorphic augites coexisting with orthopyroxenes ranging incomposition from Fs,u to Fs"s from the Adirondacks, the Hudson Highlands, and the CortlandtComplex, New York, and the Belchertown Complex, Massachusetts, shows three sets ofexsolution lamellae. X-ray single crystal photographs show these are orthopyroxene lamellaeon (1@), and pigeonite lamellae, termed "0Ot" and "l00", oriented on irrational planes near(001) and (100). Optically observed angles of the phase boundaries of "001" and "100"lamellae with respect to the c axis of host augite vary with iron-magnesium ratio determinedfrom electron probe analyses of 29 pyroxenes, supplemented by 76 measurements of gammaindex of refraction. The angles are largest in magnesian specimens and are in near agreementwith angles of optimal phase boundaries calculated from measured lattice parameters of hostand one or both sets of pigeonite lamellae. Compositional control of lamellae orientation isrelated to compositional control of lattice parameters, which appear not to have changedsignificantly since exsolution in the range 800-500'C.

Introduction

Many igneous and metamorphic rocks containpyroxenes which have unmixed during cooling. Theexsolved pyroxenes form as lamellae which havebeen generally described in the past :rs beingoriented parallel to the (100) or (001) latticeplanes of the host (Poldervaart and Hess, 1951).Metamorphic augites containing sets of exsolutionlamellae in three different orientations were firstdescribed by Jaffe and Jaffe (1973)1 from theiroccurrence in augite-orthopyroxene granulites andamphibolites associated with Precambrian crystallinegneisses of the Monroe quadrangle in the HudsonHighlands of New York (Fig. 1). These authorsnoted that only one of the three sets of lamellaewas parallel to a crystallographic axis, the c axis,of the host augite, whereas the two more prominentsets of pigeonite were not parallel to the (0Ol)

'The paper by Jaffe and Jaffe (1973) was submitted forpublication in 1969, but not published until 1973 becauseof a freeze on funds at the N.Y. State Museum and ScienceService. It thus preceded the study and publication of thepaper by Robinson, Jafte, Ross, and Klein (1971).

and (100) directions of the host, as had been gen-erally accepted (Poldervaart and fless, 1951).These significant relations were established solely bypetrographic microscopic study of grain mounts bynoting that the obtuse angle between the two promi-nent sets of lamellae was 122" rather than the105-106" required by two sets of lamellae parallelto (001) and (100) of augite.

Subsequently, a study of these exsolution phe-nomena in more detail by Robinson, Jaffe, Ross,and Klein (197l)L established by single crystal X-rayphotographs that the lamellae parallel to (100) ofthe augite host were orthopyroxene. They foundthat the two more prominent sets of lamellae, desig-nated as "001" and "100" lamellae, were indeedpigeonite' with phase boundaries, separating lamel-

'The definitions of pigeonite and augite used here arebased partly on the work of Ross, Huebner, and Dowty(1973). Pigeonite is defined as a monoclinic pyroxene con-taining less than 20 mole percent CaSiOs component (Wo (20). The FeSiO" component can vary from zero to 100mole percent, although for the end-members only, we woulduse the names clinoenstatite (En'.) and clinoferrosilite(Fs"-). Pigeonite can be unambiguously identified by single

10 TAFFE, ROBINSON,

EXPLANATION

lF.n Mesozoic ond Cenozoic Cpver

T;-l Belchertown, Corllondl Complexes

D'Kl Anorthosite

ffi.,:iln Precombrion

Ftc. l. Generalized geologic map slrowing locations fromwhich coexisting pyroxene$ \ilere obtained: (1) Monroequadrangle, Hudson Highlands; (2) Mpunt Marcy quad-rangle, Adirondacks; (3) Belphertown Compkx; (4) East-ern part of the Cortlandt Complex.

lae from host, that are not oriented parallel to aspecific rational lattice plane of the host. Theseauthors also demonstrated that the orientation ofexsolution lamellae of pigeonite in augiten and alsothat of monoclinic amphiboles in each other, was inaccordance with the optimal phase boundary theory

crystal X-ray diffraction by ( I ) its space group symmetryHh/c at room temperature, and (2) by having a B angleof greater than 107.5' (usually 108-109") at room tem-perature. The names clinoenstatite, glinobrEnzite, clino-hypersthene, and clinoferrosilite are not used for naturallyoccurring pyroxenes because there is no consensus on therange of Fs or Wo content to be attached to those names.Some would prefer to call a clinopyroxene of cornpositionWo$ (En, Fs)* 'bigeonite" but would call a clinopyroxene

TRACY,,{ND ROSS

of Bollmann, as applied to exsolution in feldsparsby Bollmann and Nissen (1968).

Soon thereafter, a study of glain immersionnrounts revealed that the exsolgtion phenomena dis-covered in the two-pyroxene granulites of the Hud-son Highlands, described above, are even morewidespread in two-pyroxene-bearing anorthosites,granulites, and charnockitic gneisses in the MountMarcy quadrangle of the Adirondacks. Detailedareasurements of the gamma indices of refractionof orthopyroxeRes coexisting with these clinopy-roxenes indicated a wide range of iron-magnesium1sfi6s-n36ely, 100 (Fe2* * Fe3. * Mn)/(Mg +Fe2* * Fe8* * Mn)-fronr 40 to 95 (Tabtre 1,Fig. 2). Pyroxene assemblages richer in magnesiumwere obtained from the BElchertown fnfiusive Com-plex of central Massachusetts (Fig. 1; Emerson,1898, 1917; Guthrie and Robinson, 1967; Guthrie,1,972; Hall, L973) and the Cortlandt ultramaficcomplex of southeastern New York (Fig. 1; Shand,1942;Tracy, l97O). Mea$ure(nents of the gammaindex of retraction of orthopyroxenes from theserocks indicated iron-magnesium ratios of 15 to 35(Table 1). Thus pyroxene pairs were available forstudy in which the orthopyroxenes range in com-position from Fsrs to Fse5 (Fig. 2). In all casesthese orthopyroxenes coexisting with augite con-tain augite exsolution lamellae parallel to (100).

Although of different primary origins, all of thepyroxene pairs under discussion are believed tohave equilibrated under generally similar meta-morphic conditions of high tqmperature, inter-mediate pressure, and low humidity equivalent tothe granulite facies. Hudson Highlands and Adiron-dack specimens appear to have been involved inPrecambrian granulite facies metamorphism of re-gional extent, although the Adirondack specimensalso contain some relict features of earlier plutonicevents. The specimens from the Belchertown Com-plex came from a Devonian batholith. Subsequentto syntectonic emplacernent, the batholith under-went hydration during continued regional metamor-

of composition Wog (En, Fs)* clinohypersthene. Yetpigeonito on cooling may unmix enough augite to give itthis low calcium content.

Augite is defined as a elinopyroxene having greater than20 mole percent CaSiO' (Wo ) 2O) aad Fs content fromzero to E0 mole perc€nt. Augite is unambiguously identifiedby having space group symmetry C2/c ar:.d a p angle ofless than 107.5' (usually 105.5-106.5"). The names diopside and hedenbergite are reserved for the augite end-members CaMgSLOo and CaFeSLOo respectively.

PIGEONITE EXSOLUTION LAMELLAE IN METAMORPHIC AUGITE

Frc. 2. Compositions of co-existing pyroxenes from the Hudson Highlands (open circles),Adirondacks (closed circles), Belchertown Complex (open squares), and Cortlandt Complex(closed squares). All compositions were determined by electron probe, except for thosemarkedby the smaller closed circles, which were determined optically. Solid tie lines'indicate specimensfor which X-ray single crystal data have been obtained; long-dashed tie lines, only opticaldata on exsolution lamellae. Short-dashed lines indicate trend of zoning in angites 447,T65,and A21. Stippled pattern indicates limits of mutual solid solution of synthetic Ca-Mg-Feaugite and orthopyroxene at 810"C as determined by Lindsley, King, and Turnock (1974).

1 1

H E

E N t 5

phism that left the central core with the relictgranulite facies mineralogy of an orthopyroxene-augite rronzodiorite. The interior of the eastern endof the Ordovician Cortlandt Intrusive Complexseems also to have preserved essentially granulitefacies assemblages although undergoing recrystalliza-tion during later regional metamorphism (Tracy,1970). Further detail concerning the specimens andtheir geologic setting is given in the Appendix.

Optical Properties and Composition of Pyroxenes

The compositions of coexisting pyroxenes in 38specimens were estimated by measurement of indicesof refraction in immersion oils (Table 1) andverified by thirteen sets of electron probe analyses(Table 2). Optical curves based on the new probeanalyses (Figs. 3, 4) proved more reliable thancurve.s based on literature data and were used forall optically determined compositions reported inthis paper.

Orthopyroxene

Measurement of the gamma index of refractionof orthopyroxenes in fragment mounts is simplebecause both the excellent {210} cleavage and thegod {100} parting provide numerous plates ori-ented with the Z-vibration direction parallel to the

stage of the microscope. According to Deer, Howie,and Zussman (1963), each atomic percent of(Fe'z. * Fe3* * Mn) leads to an increase of O.@125in the gamma index of refraction of orthopyroxenesfrom both igneous and metamorphic parageneses.This correlation incorporates data for orthopy-roxenes from plutonic rocks (Hess, 1952, 1960),from volcanic rocks (Kuno, 1954), and from high-grade metamorphic rocks (Muir and Tilley, 1958;Howie, 1955), all of which are plotted on Figure 3.Electron probe and optical data obtained duringthis study show less scatter and a new curve relatingiron-mapesium ratios with gamma indices of re-fraction was constructed (Fig. 3). This was usedto obtain the iron-magnesium ratios listed in pa-rentheses in Table 1.

Augite

Ordinarily, one measures the beta index of re-fraction of augite in order to obtain an estimate ofthe iron-magnesium ratio as suggested by Hess(1949). However, an evaluation of the optical datafor analyzed augites listed by Hess (1960) and Deeret aI (1963) shows that the gamma index of re-fraction yields estimates of the iron-magnesiumratios comparable to those obtained from the betaindex. It is much more informative and convenient

t 2 ]AFFE, ROBINSON, TRACY, AND ROSS

f'esrE 1. Optical Measurements of Angles of ExsolutionLamellae of Pigeonite in Host Augite, .y Indices of Re-fraction, and Compositions of Coexisting Pyroxenes fromthe Adirondacks, Hudson Highlands, Belchertown Complex,

and Cortlandt Complex'r'

Sanp]e ] -Aug fe -Aue y-opx fe -Opx

e r t t o l o e c r s , n . t . ' '

r 7 8

t 7

C o - 1 7

T65'o

T526

lo 20 30 40 50 60 70 80 90 loo

IOO (Fezt+ Fe3*+ Mn) / (Mg + Fe2t 1ps3+ + Mn )

Frc. 3. Variation of the .y index of refraction and theratio 100 (Fe* + Fe''. + Mn)/(Mg { Fe',* + Fe'. + Mn)for orthopyroxenes. Data from Deer, Howie, and Zussman(1963) and present study (labelled points). Indices ofpure synthetic enstatite (1.658) and orthoferrosilite (1.789)from Stephenson, Sc.lar, and Smith (1966) and Lindsley,MacGregor, and Davis (1964), respectively. Solid curvefitted visually to data points of present study (except T65)and synthetic end members. Dashed line is linear leastsquares best fit of same data points. The equation for thisl ine is:

n7 - 1.6626 + 0.1297 [(Fe* + Fee + Mn)/

( M g * F e P * f F e " * * M n ) 1 .

the extensive substitution of Alrv, {lvr, pss*vr, Tivr,and NaYrrr in augites, and a plot of published data(Fig. a) shows a considerable scatter of points.Electron probe and optical data obtained duringthis study show much less scatter and a new curverelating iron-magnesium ratios with gamma indicesof refraction was constructed (Fig. 4). This wasused to obtain the iron-magnesium ratios listed inparentheses in Table 1

n"l

i{kPir- 3

; {a -5}ID-I

7 r 5 ( 2 6 )1 r t ( 2 9 )1 r 1 ( 2 9 )J L I S ( 3 C )7 1 9 5 ( 3 2 )

J o - 8 1 . 7 2 0 3 3 . 0 ( 3 3 )P r - l r , 7 2 0 ( 3 3 )co-4 1.120 lq. : (33)s b - 2 r . 1 2 4 ( 3 9 )P o - r 1 . 1 2 6 ! l { C r , z l

s l - 5 1 , 7 2 6 ( 4 2 \: { a - 3 r , 1 2 7 ( 4 4 ,v r { - 2 I . 1 2 7 ( 4 4 'c a - 6 r . 7 2 9 4 6 . 9 ( 4 7 \c D 1 , 7 3 3 ( 5 3 )

ca-17 r.736 l l :7 (57>c i a n t i i 1 . 7 3 5 ( s 7 )J B - 2 L . 1 4 I ( 6 5 )G o - 2 I . 1 4 6 ( 7 2 )s b - r r . 7 4 8 5 ( 7 5 )

P o - 1 3 1 . 7 s 0 5 g L f Q e 's c - 6 L 7 5 9 8 7 , 5 ( 9 2 )P o - 1 7 r . 7 6 0 9 1 . 4 ( 9 3 )

l lUDSoN l{ IGHltr \NDS, N, Y.

J 5 1 s 1 . 7 r 3 ( 2 3 )J 2 4 1 S L 7 r 6 ( 2 7 )J 2 2 3 r , 7 2 0 3 5 . 4 ( 3 3 )J 4 3 1 P I 7 2 2 ( 3 6 )

BELCHERTOWN, MASS.

l r 5I l 0t l 3A 2 1

N . D .L . 7 1 41 , 7 r 8L 120I . 7 I 8

( 4 0 )( 4 3 )( 4 4 t( 4 3 )

1.720 !L9 (44)r .72O (44 'r . 7 2 7 4 9 . 2 ( 4 9 \I . 7 2 1 ( 4 9 )r . 7 3 1 5 3 . 9 ( s 4 )

\ . 7 2 9 5 ( 5 1 )r . 7 2 9 ( 5 r )1 . 7 3 0 ( s 1 )1 .738 :Z j l (57)r . 7 3 8 ( s 7 )

,r.752 99:L (68',,L 7 5 9 ( 7 4 ' )r .757 (72 ' , )I . 7 7 3 ( 8 5 )r . 7 7 r ( 8 3 )

1 . 7 7 6 0 8 8 . 9 ( 8 8 )) - ,785 923 (96 \1 ,786 95 .s (97)

r .7 r7 (12)1 . 7 1 s ( 4 0 )r . 7 2 9 5 0 1 ( s r )r .731 (52 ' )

!.682 14{ (16)1 .703 (32)r .7o4 (32)r .7o25 (3 r )r ,705 l l f (33)

r .69 t (21 \r ,6935 (25 t1 .701 (32 'r .696 22 I (26 \r 1o2 25 ,7 (3 r )

I t 6 . 5 ' l 0 'N . D . I Z '

r 1 9 " 1 1 5 '1 1 7 ' 7 . s '1 1 7 ' 1 1 "

126 5"

1 3 0 . 5 'r 2 4 . 5 "1 2 8 "

t29'129.123'

r24'

1 2 l 'L 2 2 "125.1 2 0 'I 2 t '

r r 5 . 5 't t5 '113 't r2 '

11 t . 5 .1 I I '

t720

t T t o

| 7 0 0

I O Y U

| 680

| 670

1 . 1 0 4 1 2 . 9 ( 9 )1 . 7 1 4 ( 2 4 \r , 7 r 4 ( 2 4 \r . 1 1 6 ( 2 7 'I . 7 1 6 2 6 J \ 2 7 ,

1 1 i "1 1 7 "I 1 5 '1 1 6 '1 1 7 '

1 r 5 . 5 o

I I 5 "1 1 6 . 5 "r l 5 '1 1 5 '

1 1 2 . 5 .1 1 3 "I I 2 .1 1 1 . 5 01 1 1 . s '

1 u , 5 " i / d1 1 r . # #rrr" I I

115 '115 ' l0 '116 '114 ' 6 '

722" 22"1r9 ' 13 .r r 9 , 5 o 1 2 , s 'r20" 13"r20" 12 ,5"

120 ' 16"r2z ' r7 '119 ' 16 '1 2 2 . 5 " 1 9 . 5 'r23" 17"

I 2 '1 , 2 '

7 "

8 . 5 '

8 , 5 '5 '

3 '2 '1 "0 , 5 "0 . 5 "

0 "

124.125'r22 '120'

C0RTLANDt CoyPLDX, N. Y.

1 6 2 r t o t 5 ( r 4 )r s 8 I 7 0 8 ( 1 5 )T 2 5 I 1 r 4 ( 2 4 )\ 5 2 1 . 7 1 8 2 3 . 1 ( 3 0 )1 6 5 t . 1 2 4 2 L . O ( 3 9 )

7 4 t '1 3 2 '732"

1 3 6 'r 3 9 '1 3 5 'I L 2 "1 4 0 "

* A l l n i c r o s c o p i c m e a s u r e m e n E s b y I { . i l . J a f f e .

n* re - 100( r .2++Fs3+1y6;71yg+pg2++pe3++un; derermt ,ed by erec t ron probea n a l y s i s ( u n d e r l i n e d ) o r o p r i c a l l y u s i n g F i g u r e 3 o r F t g u r e 4 , r h i s p a p e !

r i A d i r o n d a c k s p e c i m e n s e x c e p c C i a n t a r e f r o n ) l o u n t ; 4 a r c y 1 5 r q u a d r a n g l e .C i d n t i s f r o m a d j o i n i n g E l i z a b e r h r o D n q u a d r a n g t € .

+ r ; l l c c e n E o p t i c a l r e - e x r m i n a t i o n o f P o - 1 3 , S C - 6 a n d P o - ] 7 h a s r e v e a l e d av e r y v e r y f r n e s e c o n d s e r o f " 0 0 I ' r p i g e o n i E e I a n e l l a e o r l e n r e d a r I t 2 o ll L 2 5 ' , a n d l I 2 5 ' r e s p e c t i v e l y T h e s e a p p e a r t o r e p r e s e n r a s e c o n de p i s o d e o t e x s o l u t i o n a t l o a e r r e n p e r a ! u r e , s l m i l a r E o s e c o n d a n d t h i r ds e t s o f " 0 C I " r a m e l l r e o b s e r v e d 1 n s o n e i g n c o u s a u 8 i r e s ( R o s s , R o b i n s o n ,a n d l . f f e , r 9 7 2 ) .

to measure the gamma index because grains thatlie parallel to the optic plane (010) yield, in ad-dition to the gamma index of refraction: (1) maxi-mum relief between exsolution lamellae and host.(2) maximum exsolution angles, "001" n c and"100" n c, and (3) the relative orientation of theZ-vibration direction with respect to exsolutionlamellae and the c-crystallographic axis (see Jaffe,Robinson, and Klein, 1968). Neither B or y, how-ever, yields accurate iron-magnesium ratios due to

i ' r ? ? 3 , G o - 4

PIGEONITE EXSOLUTION LAMELLAE IN METAMORPHIC AUGITE l 3

'""-o ro 20 30 40 50 60 70 80 90 roo

loo (Fe2*+ Fe3* + Mn)/(Mg+ Fez*+Fe3t + Mn)

Ftc. 4. Variation of the 7 index of refraction with theratio 100 (Fe* + Fe"* + Mn)/(Mg f Fe2* + Fes* + Mn)in augites coexisting with orthopyroxene. Data from Hess(1960); Deer, Howie, and Zussman (1963); and presentstudy (labelled points). Line is both visual and least squaresbest fit to data points of present study, excluding T65.Equation for line is:

n7 - 1.6977 + 0.0669 [(Fe* + Fei l + Mn)/( M g + F e ' , * 1 F s " * * M n ) ] .

Composition ol Coexisting Pyroxenes

Electron probe analyses and structural formulaeof coexisting pyroxenes are listed in Table 2 andpresented graphically in Figure 2. In doing theanalyses an effort was made to avoid putting theelectron beam on large concentrations of exsolutionlamellae. When the beam was deliberately aimed atconcentrations of lamellae, the analysis usually gavesome intermediate composition along the tie lineconnecting augite and orthopyroxene compositions.Thus, in general, analyses with maximum and mini-mum cd8 values were considered to represent thecompositions of the augite and orthopyroxene, re-spectively, on which the optical measurements weremade. Exceptions to this occurred in augites 447,T65, and A21 (Fig. 2) in which analyses on aI>parently clear grains yielded a range of ca\ valuesbetween distinct upper and lower limits. Theanalyses, with low ca3 values of 38.7, 39.1, and37.0 respectively, can be interpreted in two ways:(a) they represent distinct primary chemical zoningin these metamorphosed igneous augites, or (b)they represent analyses of places on the polishedsurface of the augite crystal occupied by thin

pigeonite lamellae parallel to the surface. In allthree examples (Fig. 2), the chemical zoning trendof augite lies on the Fe-rich side of the augite-ortho-pyroxene tie line. This suggests that pigeonitelamellae are more Fe-rich than the coexisting ortho-pyroxene, but does not rule out the possibility ofprimary zoning, which is supported in the case of447 by some variation in the index of refraction(1.700 to L.704). In this case the highest indexcorresponding to the highest Ca content was em-ployed for optical determinative work.

Keeping in mind the uncertainties discussedabove, lhe cas values from probe analysis of co-existing augite and orthopyroxene can be used toestimate temperature conditions of exsolution on thebasis of experimental work in the pyroxene quad-rilateral, particularly the 810'C isotherm (Fig. 2)of Lindsley, King, and Turnock (1974). Un-fortunately, unlike the lunar specimens evaluated bythese authors, only four of the specimens reportedhere contain less than 3 wt percent of "nonquadri-lateral" components. Nevertheless, the plotted trendsof the Adirondack specimens in Figure 2 arc gen-erally parallel to the experimental solvi and suggesttemperatures of equilibration lower than 810"C.

As is to be expected, the orthopyroxene generallyhas a higher Fe/Mg ratio than the coexisting augite.Distribution of Fe and Mg between the coexistingpyroxenes may be described in terms of a distribu-tion coefficient KD (Kretz, 1963) wherc

O p x - A u g

KDM g - F e

_ (Mg/Fe + Mg)oo*. (Fe/Fe + Ms)o'"

(Fe/Fe * Mg)o"- 1Mg7Fe + Ms)""*

This simple formulation was employed in calculatingthe values of Ko in Table 3. More elaborate formula-tion involving consideration of full site occupancy(Saxena, 1971) was rejected at this time becauseof uncertainties in site assignment related to analyti-cal uncertainties and inadequate knowledge of theoxidation state of Fe.

The effect of Fe3* on Kn is illustrated by specimenA21, which is probably the most oxidized rock inthe suite, in that the pyroxenes coexist with titano-hematite. Wet chemical analyses of purified augiteand orthopyroxene separates (Ashwal, 1974) yieldratios of Fe1+fFe2+ * Fes* of 0.333 and 0.085 re-spectively, and a much lower Kn value (Table 3).Estimates of Fes* content can be made from electronprobe analyses, by summing the formula to fourcations, but this method is heavily dependent on'ca = lMa/(Ca * Mg f Fe * Mn)

t 4 IAFFE, ROBINSON, TRACY, AND ROSS

Taers 2a. Electron Probe Analyses of Coexisting Pyroxenes

447-1 441-2 .r52 T65-r a65-2 A2l-I A2L-2 J o - 8 G o - 4 J 2 2 3 P o - l C a - 6 C a - I 7 P o - I 3

AUGITE

s i o ^ 5 1 1 8 l t 5 2 . 4 0 i l 4 7 . 6 b *

T i o ^ . 3 1 . J 6 . 3 6

[ - o " 3 . 6 1 4 . ] I 6 . 9 7

c r ^ o " . 9 8 . 6 3 , 2 9

Y s o 1 5 . 2 7 1 7 . 5 1 I 4 9 8

N i o . 2 3

F e o 1 . 8 1 5 . b 8 7 , 7 0

I l n o . 1 8 . 0 9 . 3 2

C a o 2 2 , 5 2 1 a . 2 2 2 L 1 0

B a O . 0 1

N a " 0 . 9 1 . 9 0 . 6 7

l ( " 0 . 0 1

T o t a l 9 9 . t L 9 9 9 0 I O O . l 1

ORTHOPYROXENE

s 1 0 2

T l .o2

I z o :

c t 2 o 3

Mco

N i o

FeO

MnO

C a O

B a 0

N a 2 O

K 2 o

T o t a l

5 3 , 7 2 * 5 r . 3 7 *

. 0 8 . I 4

. 1 0 0 5

l r . 2 8 2 8 . 0 9

9 . 1 6 1 4 . 4 0

. 3 2 . 3 5

1 . 0 6 r 2 1

. 0 0 . 0 0

. r 0 0 0

0 3 0 0

9 8 2 1 1 0 0 . 4 8

4 9 . 8 8 * 4 9 . 6 6 + 5 0 . 2 9 t |

. 1 8 . t 8 . 0 2

2 . 7 7 2 . 4 4 1 . 3 9

. 4 6 , 6 2 0 2

1 2 . 7 6 1 2 . 6 3 1 2 9 6

, t 4 . 0 8

t o . 9 2 1 2 , 5 2 7 2 , 4 9

, o

2 2 , L 5 2 r . 5 6 2 1 , 1 l

, 0 3 . 0 3

5 5 . 6 8 . 1 4

. o 2 . 0 3

1 0 0 . 1 5 r 0 0 . 8 2 9 9 . r 7

4 9 . 9 2 * 4 8 . 9 7 * 5 0 , 7 3 1 j

. t 0 l t . 0 r

1 . 5 9 r . 0 5 1 . 0 6

. 0 0 . 1 6 , 0 0

1 9 . 8 3 1 8 . 0 6 r 1 , 2 0

. 0 4 - 2 8

2 8 . 0 8 3 0 . 3 6 2 8 8 1

. 5 8 . 8 7 1 8 8

- 5 3 . 6 0 6 1

. 0 2 . 0 0

. 0 4 . 0 6 0 L

. o 2 . 0 2

1 0 0 . 7 5 r 0 0 . 5 4 1 0 0 . 3 7

4 7 . 1 7 * 4 8 . 6 I *

1 . 6 1 r . 6 6

6 . 2 8 5 . \ 2

, 0 9 . I 5

1 3 . r 3 1 4 . 8 1

1 . 2 L 1 0 . 1 2

. 1 9 , 2 5

2 2 . 4 5 1 8 . 4 3

. 6 8 . 6 5

. 0 0 . 0 0

9 8 , 8 0 9 9 . 7 9

5 0 . 9 8 *

3 6

4 . 2 5

. 0 7

2 6 4 5

1 6 1 9

2 L

1 . 1 1

I 2

. 0 0

9 9 , 7 5

5 1 . 4 1 i l s Z . r T t l

. 3 1 . 1 0

2 . 2 7 1 9 9

. o 7 . o 2

1 1 . 6 3 1 5 . 0 5

8 . 3 5 r t . 4 7

- 2 3 . 2 4

2 t . 2 9 1 1 . 7 2

. 8 9 . 8 5

9 8 4 2 9 9 . 5 r

5 2 . 5 5 1 1

. 0 3

1 . 2 0

. 0 3

2 4 . 0 0

2 0 . 6 3

. 7 4

0 0

9 9 . 6 4

4 8 . 9 5 * 4 9 . 0 4 t

. 2 t . r 7

2 . 7 7 2 5 0

. 4 4 3 3

1 1 . 1 2 I 0 - 4 2

, 1 8 . I 8

1 4 . 0 1 1 6 . L 2

2 r , 6 2 2 0 . 2 4

. 0 2 . 0 4

. 7 7 r . 0 5

, 0 2 . 0 2

1 0 0 . 3 6 r 0 0 . 4 0

4 8 . 1 7 * 5 0 . 2 7 *

. t 2 . 0 9

r . 6 0 1 . 1 8

. 1 5 . O 4

1 5 . 9 7 1 4 . 5 0

. 0 5

3 2 , 6 9 3 3 . 9 6

5 0 , 4 1

7 4 , 6 6

' 0 4

. 0 6 . 0 0

. 0 4 . 0 0

1 0 0 , 1 3 r 0 r . 1 r

4 8 . 7 2 * 4 7 , 1 7 *

. 1 6 . 1 4

1 . E 6 r . 2 4

, 1 8 . 0 7

8 . 3 1 3 , 1 1

. I 8 z n o , 2 4

1 9 . 5 1 2 6 . 0 4

. 5 8 . 3 1

r 9 , 9 1 1 9 . 6 8

. 0 3

. 8 5 . 7 8

. 0 4 . 0 0

r 0 0 . 3 3 9 8 . 7 7

4 8 . 0 5 # 4 4 , 9 2 *

, 0 6 . 1 0

, 6 9 . 3 8

. 0 0 . 0 5

1 0 . 4 3 3 . 4 4

z n o . 4 5

3 8 . 6 5 4 4 . 3 2

r , 0 3 . a 2

. 6 8 . 6 8

. 0 1 , 0 0

. 0 0

9 9 , 6 0 9 9 . 1 6

4 8 . 1 8 * 4 7 , 5 3 *

. 0 8 , 2 1

. 8 9 1 . 0 8

. 0 0 , t l

2 . 2 r r , 2 3

. 2 9 , 4 6

2 7 . 4 7 3 0 . 3 9

. 3 5 . 4 4

1 9 . 2 4 1 8 . 2 9

. 1 1 . 5 7

. 0 0 . 0 0

9 9 . 4 4 1 0 0 . 3 r

4 5 . 0 5 * 4 5 . r 1 *

, 2 6 . 4 7

. 0 7 . 0 9

2 . 3 4 l . 3 5

. 5 6 . 6 2

4 9 . 4 0 5 0 . 1 4

. 8 3 I 3 0

. 1 2 . 1 5

. 0 0 0 8

, 0 0 0 {

1 0 0 3 5 r o o . 1 4

'R J . T racq, ana lgs t ; on MC uode l 4OA ekc t ron probe, us ing sxandard Bence-A lbee cor rec t jon pracedurcs , a t l rs t i tuae a f Mater ia fs Sc ience,un jvers ixg o f Connect icu t , S tar rs ' except T65, Opx Ca-6 , Po-13, { -5 , Po-17 a t Deqr t ren t o f Ear th and P lanetarg Sc je rces . dassac}usetCsInsx i tu te a f rechno loqu,

#Nelson E ick l ing , ana lgs t ; on ARL-EMX e lec t ron probe, os ing s tandard cor rec t ion pracedures , a t U ,S. Gea log ica l Surveg, Wash ing ton , D C.

Sanple

T,rst-p 3. Distribution Coefficients of Fe and Mg inCoexisting Augite and Orthopyroxene

Fe / F e+MB KD Sanple

the accuracy of the SiO2 analysis and was not usedfor this paper.

It thus appears likely that much of the scatter ofKn values in the range 0.577 to 0.671, (specimens

Jo-8 through Po-17) is a function of analytical diffi-culties. They are, however, all lower than the gen-eralized value of Kr : 0.690 suggested for the810'C isotherm by Lindsley et al (1974), andhence consistent with equilibration at a lower tem-perature. On the other hand, the extremely highvalues of KD exhibited by specimens 447-1, T52,and T65-1 appear to be related to very high Al2O3andfor TiO2 content in augite, possibly combinedwith higher temperature of formation of these ultra-mafic rocks. The still higher values for 447-2 andT65-2 are related to the low ca' values of theseaugite analyses. If the interpretation is correct thatthese low Ca analyses are due to pigeonite exsolu-tion lamellae, then they may be combined with thehigh Ca analyses to yield a Kn which might beinterpreted as that between augite and pigeoniteduring the exsolution (Table 3).

The analyses of Table 2 permit generalizations

Fe/ Fe+Mg KD

447-1

447 -2

1 4 7 - lt t47 -2

't52

T 6 5 - I

'165-2

T65- lT65-2

A2 l -1

A 2 l - 2

A21- 1A2I-2

A 2 Lt,lET

J o - 8

G o - 4

J 2 2 3

P o - 1

C a - 6

C a - I 7

s c - 6

P o - 1 7

AugoPx

Augo P x

Augopx

A u gopx

Au8opx

AugoPx

AugoPX

AugoPx

AugoPX

AugoPx

Augopx

AugAu8

Augo p x

A u go P x

Augo P x

AugAug

A u goPx

Augopx

A"gAug

AugoPX

, 1 2 3. t 4 l ' d ) 4

: i ; i 1 . roe

, 123

. 2 2 41 7 L

?19 .as' /

. 2 7 7

. ' r l i ' 806

. 2 5 6, / r )

. \ ; ; . 8 8 6

. 'r33 .807

, L 9 4 q < o. 1 0 1

. 3 2 4

. 4 4 1

q o n

. 3 5 1

. 1 4 8 4 ' ) / l

. 4 1 4. o l q

, ) J )

. 4 6 5. o o r

. 5 6 8

. a a i ' o u l

. 8 7 4, g 2 2

' ) 6 0

. 9 5 4


Tanr-r 2b. Formulas Per Six Oxygens for Coexisting Pyroxenes

l 5

4 4 1 - 2 't52 J o - 8T65-1 T 6 5 - 2 c a - L l P o - I 3

AUGITE

s i 1 , 9 0 9 1 9 0 9

l . r . 0 9 1 . 0 9 12 . 0 0 0 2 . 0 0 0

A r . 0 6 6 . 0 8 6

r i - 0 0 9 . 0 I 0

c r " . O 2 9 . 0 1 8

M g 8 3 9 . 9 5 1

N i

F e - . 1 I 8 , I 7 3

M n - . 0 0 6 . O 0 3

c a . 8 9 0 . 7 I I

Ba

N a . 0 6 6 . 0 6 4

K2 . 0 2 3 2 . 0 1 6

f e * * I 2 . 9 1 5 . 6c a l l { l 4 8 . 0 3 8 . 1

ORTITOPYROXINE

s i

AI

AI

C r -

N i

Mn-

C a

B a

Na

t . 9 2 4

, 0 7 62 - 0 0 0

, 0 1 8

. 0 0 2

0 0 3

r , 6 7 0

. 004

2 7 4

: 0 1 0

0 4 r

. 0 0 7

. 0 0 12 . 0 3 0

2 . O

7 . 8 2 2 1 . 9 4 8

. r 7 8 . O 5 22 , 0 0 0 2 , 0 0 0

. 0 4 8 . 0 4 9

. 0 4 6 . 0 0 9

. 0 0 4 . 0 0 2

. 8 2 7 . 7 6 9

. 3 r 7 . 2 6 4

, 0 0 7 . 0 0 7

. 7 3 9 . 8 5 9

. 0 4 6 . 0 6 5

2 . O 3 4 2 . 0 2 4

2 8 . 1 2 6 . 13 9 , 1 4 5 . 3

. 1 4 32 . 0 0 0

. 0 3 9

, 0 0 9

. 0 0 1

1 . 4 3 6

. 4 9 3

. 0 0 5

, 0 4 2

. 008

2 . 0 3 3

2 5 . 12 . L

r . 9 5 9 r . 9 6 2 I . 9 5 4

. 0 4 1 , 0 3 3 . 0 1 92 , 0 0 0 1 . 9 9 5 r 9 7 3

. 0 r l

. 0 0 2 , 0 0 2 . 0 0 3

, 0 0 0 . O 0 2

. a 4 2 , 6 3 5 . 2 2 3

Zn 'OI4

1 , 1 0 7 L . 3 2 0 1 , 7 5 8

. 0 r 2 . 0 3 6 . 0 3 0

, o 2 1 . 0 1 0 0 t 2

. 0 0 1

2 . 0 0 3 2 , 0 2 4 2 . 0 6 2

5 7 . r 5 8 . 1 8 8 , 9r 3 1 . 5 r . 6

r . g r 4 L . 9 4 6 r . 9 7 8 1 . 9 6 2

. 0 8 6 , 0 5 4 , O 2 2 . 0 1 82 . 0 0 0 2 . 0 0 0 2 . 0 0 0 2 . 0 0 0

. 0 0 6 . 0 2 1 . 0 1 4

. o o 5 . o o 4 . 0 0 2 . 0 0 7

. 0 0 6 , 0 0 2 . 0 0 4

, 4 8 7 , 1 9 1 . 1 3 6 ' 0 7 5

. 0 0 6 Z n . O O 7 ' 0 0 9 . 0 1 4

, 6 4 r . 8 9 9 . 9 4 4 1 . 0 4 9

, 0 1 9 . 0 1 r , 0 1 2 ' 0 1 5

. 8 3 8 . 8 7 0 . 8 4 7 . 8 0 9

. 0 0 1

. 0 6 5 . 0 6 2 . 0 5 8 0 4 6

# zn52 2 .us zn ,

5 7 . s 8 2 . 7 8 7 5 9 3 ' 44 2 . 2 4 4 . L 4 3 . 7 4 L 5

r . 9 8 2 r , 9 6 1

. 0 1 3 - O 2 41 . 9 9 5 1 . 9 9 1

. 0 0 4 . 0 0 4

. 0 0 3 . 0 0 3

, 1 5 0 . 0 8 8

. 0 r 8 . 0 2 0

r . 7 1 8 L . 8 2 6

. 0 3 0 . 0 4 8

. 0 3 3 . 0 3 5

. 0 0 6

. 0 0 22 . 0 1 6 2 . O 3 2

9 2 . 3 9 5 5L . 7 1 , 8

r , 1 1 6 r . 7 8 4

.224 2162 . 0 0 0 2 . 0 0 0

. 0 8 2 . 0 6 4

. 0 1 0 . 0 4 5

. 0 0 9 . 0 0 2

. 8 3 2 . 7 4 0

. 0 0 7

. 2 4 0 . 2 2 8

. 0 1 0 . 0 0 5

. 8 4 2 . 9 0 9

. 0 0 1

, 0 4 8 . O 4 9

. 0 0 r2 . 0 8 2 2 . 0 4 2

2 3 . 1 2 4 . O4 3 . 8 4 8 . 3

1 . 8 4 6

. r 5 42 . 0 0 0

. 0 4 6

. 0 0 4

. 001

1 . 5 0 4

. 0 0 4

. 4 3 3

, 0 1 1

. o 4 7

2 , 0 5 0

2 2 , 82 , 3

1 . 9 5 5 1 , 8 9 1

. 0 4 5 . 1 0 92 . 0 0 0 2 . 0 0 0

. 0 4 3 . 0 1 5

. 0 0 3 . 0 0 5

. 0 0 0 , 0 1 4

. 8 4 1 . 7 2 r

. 0 0 4

. 3 5 9 . 3 4 6

. 0 0 8 . 0 0 9

. 7 1 1 . 9 0 0

' 0 0 r

. 0 6 2 , 0 4 0

. 0 0 12 , 0 2 7 2 . 0 5 6

3 0 . 4 3 3 . 03 7 0 4 5 . 5

r . 8 8 4 1 . 9 3 1

. 1 0 9 . 0 6 31 . 9 9 3 1 . 9 9 4

0 0 5 . 0 0 1

. 0 1 9 . 0 0 1

. 7 r 4 . 1 4 2

, o 0 2

, 3 9 7 . 4 0 1

. 0 1 3 , O 2 4

. a 7 7 . 8 6 9

. 0 0 1

. 0 5 0 , 0 1 0

, 0 0 r2 3 7 9 2 n 4 8

J b . )

4 3 . 8 4 2 , 1

I 8 8 0 t . 8 9 5

. 1 2 0 . r 0 52 . 0 0 0 2 . 0 0 0

. 0 0 5 . 0 0 9

. 0 0 6 . 0 0 5

. 0 1 3 . 0 1 0

, 6 3 7 . 6 0 0

. 0 0 6 . 0 0 6

, 4 5 0 . 5 2 r

. 0 0 8 . 0 0 9

. 8 9 0 . 8 3 8

. 0 0 0 . 0 0 r

. 0 5 7 . 0 7 9

. 0 0 0 , 0 0 12 , 0 7 2 2 , O 1 9

4 1 . 8 4 6 . 94 4 . 8 4 2 . 6

1 . 9 0 1

-07 41 . 9 7 5

. 0 0 4

. 0 0 5

. 9 3 9

. 0 0 2

r . 0 7 9

0 1 7

. 0 3 1

. 0 0 1

. 005

.o022 . 0 8 5

5 3 . 91 , 5

r 9 5 3

. o 4 72 . 0 0 0

, 0 0 6

, 0 0 1

. 00r

1 . 3 2 9

. 6 4 1

. 0 2 3

. 0 1 8

2 . O r 9

3 l 3r . 0

. 9 9 0

. 9 3 0

, 0 6 2

. 0 2 8

. 0 0 1

2 . 0 1 8

5 0 . 11 . 4

1 . 9 0 8 1 . 9 C 6

, o 7 2 . 0 4 8r . 9 8 0 1 . 9 5 4

. 0 0 3 . 0 0 3

. 0 0 5

r , r 2 9 1 , 0 4 8

. 0 0 1 . 0 0 9

, 8 9 8 . 9 8 8

. 0 1 9 . 0 2 9

, o 2 2 . O 2 5

. 0 0 0

. 0 0 3 , 0 0 5

. 0 0 1 . 0 0 12 , 0 7 6 2 . 1 1 3

4 4 . a L 9 , 2

1 . 9 5 9

. 0 4 12 , 0 0 0

. 0 0 7

. 0 0 0

** fe = I00(Fe+Mn)/(Fe+MnfMg)// / / ca = 100 cal(Ca+re+Mn+Ms)@ Last three values in lhis rou are for Zn

concerning the behavior of several other elements.

Mn is concentrated in orthopyroxene over augite,

as is Zn in the three specimens analyzed. Cr, Al,

and Ti, with rare exceptions, are strongly preferred

in augite over orthopyroxene. Na, without exception,follows Ca in augite.

Exsolution Lamellae in Augite

C ry s t aIIo grap hic O rientatio n

The relative lattice orientations of host augite andthe three types of unmixed pyroxenes, (100) ortho-pyroxene, "100" pigeonite, and "001" pigeonite,were determined by single crystal precession pho-tography to an accuracy of i0. 1 to 0.2 degrees.Orthopyroxene and "100" pigeonite (Fig. 5A and58, respectively) are oriented so that their b and c

axes are parallel or nearly parallel to b and c,

respectively, of the host augite. These two pyroxenes

thus are oriented with their (010) and (100) planes

parallel to those of augite. The "001" pigeonite

(Fig. 5C) is oriented with its q and b axes parallel

or nearly parallel to a and b axes respectively, of

the augite; the (001) and (010) planes of both

phases being parallel. A slight divergence of the a

or c axes of host augite and exsolved pigeonite (0.25

to 0.45o), due to lattice rotation about the common

b direction, was observed in four specimens (Table

4), and is discussed below.X-ray single crystal precession photographs were

taken of eleven representative specimens from a wide

range of compositions from which lattice parameters,

relative orientations, and relative amounts of lamel-

lar phases were determined (Table 4)' The low con-

l 6

Sec t ionI t o c

Dauobopx Sec l ron

- L t o c

IAFFE, ROBINSON, TRACY, AND ROSS

gAUG

gPIG

oiue ,oirc

uoAnG

tent of pyroxene lamellae in most of the sDecimensnecessitated long exposure times, in some cases upto 2OO hours before reflections of lamellar phasescould be satisfactorily observed. In spite of this,in the more magnesian specimens (447, T52, AZl,Jo-S, Go-4, Po-l), in all of which both ',001" and"100" pigeonite lamellae were detected optically,only the "100" pigeonite lamellae were sufficientlyabundant to yield a measurable diffraction pattern.Similarly, three of the specimens (A21, J223, Ca-6)showed no trace of orthopyroxene (100) lamellaealthough such lamellae could be seen optically.

In nearly all cases the lamellar pyroxenes appearto have an identical b dimension to the host augite,or at least so close to the host that it cannot beseparately resolved. The single exception is speci-men Po-13 in which the b dimension of ortho-pyroxene lamellae is slightly smaller than b of thehost. In those single crystals in which lattice pa-rameters for both "001" and "10O" pigeonitelamellae were obtained (J223, Ca-6, Ca-17, Po-l3,Po-17), the two sets of lamellae may have slightlydifrerent a, c, and F values. It is not knownwhether these differences are due to differences inchemical composition, or represent physical con-straint on the parameters of the lamellae by thehost. In pyroxenes from high temperature lunarbasalts, Ross, Huebner, and Dowty (1973) notedthat the "001" lamellae within host augites and hostpigeonites have their a dimensions constrained toa of the host, whereas the "100" lamellae havetheir c dimensions constrained to c of the host.

M or pholo gical O rientation

All of the unmixed pyroxenes in augite grewas lamellae 0.2 to 3.0 pm thick. the contacts of

Frc. 5. Crystallographic, morphological, and opticalorientations of exsolution lamellae (stippled) in augite.A. Augite with (100) orthopyroxene lamella. dtre A aopx- 16', Doro A Dopx = 0", ceuc I coex - 0". Phaseboundary of orthopyroxene lamella is parallel to (10O) ofaugite. B. Augite with "100" pigeonite lamella. aeuo Ad u c I 3 ' , b - o o n 7 1 b e r c : 0 o , c e r o / \ c e r c t 0 ' . P h a s eboundary of pigeonite lamella is parallel to 6, but at an8" angle to c of augite ("100" I ceuc = 8.). C. Augitewith "001" pigeonite lamella. aeuc A apls - 0", Deuc ADrro = 0', c.ruc A cpro - 3o. Phase boundary of pigeonitelamella is parallel to D, but at a 115. angle to c of augite( " 0 0 1 " n c e r c - 1 1 5 ' ) .

,ooPx

-r^"AUG

F4g6;toe"

l*rn tot' A. J"orn ot'n

B.Sec l ion l l lo (OtO) Sec l ion l l to (OtO)

cauG

Sec t i on l l t o (O tO)

c.Sect ion l t to ( lOO)

lroce (lOO)phose

X+trsq

PIGEONITE EXSOLUTION LAMELLAE IN METAMORPHIC AUGITE t 7

which are referred lo as phase boundaries. The

morphological orientations of the phase boundariesof the three types of exsolved lamellae were de-termined by optical examination of single augite

crystal fragments in immersion oils. Orientationsare most successfully measured in grains orientedwith (010) parallel to the stage and immersed in

an oil of refractive index close to that of the host.

In this position the lamellae are viewed edge on,

indicating, within the accuracy of optical measure-ments, that they are parallel to the D axis of theaugite host. Because the phase boundaries areoriented parallel or nearly parallel to b of theaugite, their absolute orientation may be definedby the angle between the trace of the lamellae asseen in the (010) plane and the c direction of theaugite. These angles are designated "100" A c for"100" pigeonite, and "001" A c for "001" pigeonite.

Orthopyroxenes (Fig. 5A) are oriented so thatthe phase boundary is always p,arallel to the (100)plane of the augite. The "100" pigeonite lamellaepossess phase boundaries that deviate from beingparallel to (100) of the host by 0 to 22" (Table 4).An example of "100" pigeonite with "100" A c =

8o is shown in Figure 58. The "001" pigeonitelamellae have phase boundaries which deviate fromthe (001) plane of augite by 5 to 17o ("001" A c =

111o to 123", Table 4). An example of "001"pigeonite with "001" A c = 115'is shown in Fig-ure 5C. Where "001" pigeonite is in an orientationso that the phase boundary is parallel to (001) ofaugite, "001" A c : B augite.

Measurement of the angles of exsolution lamellaewas accomplished to a precision of f 1o on crystalfragments oriented with (010) perpendicular to theoptical axis of the microscope. Such grains are alsomost suitable for measuring the gamma index ofrefraction and the angle between the Z vibrationdirection and the c axis. The c direction of augitecould usually be found by observing either thetrace of the { 1 10} cleavage or the trace of the ( 100)orthopyroxene exsolution lamellae.

The orientation of the phase boundaries of the

"001" pigeonite exsolution lamellae is defined bythe obtuse angle ("001" A c) measured between

the c axis of the host and the trace of the lamellaeon (010). The orientation of the phase boundariesof the "100" pigeonite exsolution lamellae is de-

fined by the acute angle ("100" A c) measuredbetween the c axis of the host and the trace of the

lamellae on (010). In examples considered in this

paper, the angle between the trace of the "100"

lamellae and the a axis of the host, and the angle

between the trace of the "001" lamellae and the c

axis of the host is greater than the B angle of

augite. Both sets of lamellae thus lie "in the acute

angle B" (Robinson et al, 1971, Fig. 5-1, 5-4). In

some amphiboles and pyroxenes these angles are

less than the B angle of the host so that the lamellae

lie "in the obtuse angle B" (Robinson et al, 1971,

Fig. 5-3, 5-6; Ross, Robinson, and Jaffe, 1972).ln

grains lacking both "100" orthopyroxene lamellae

and {110} cleavage traces, the angle between the

"100" and "001" lamellae ("1.00" A "001") may

be measured as a separate but related angle.

Relation to ComPosition

The exsolution features under consideration were

first studied in augites of intermediate Fe/Mg ratio

from the Hudson Highlands (Table 1) with "001"

angles of 114-116" and "100" angles of 6-10'

(Fig. 68). Pyroxenes of similar composition from

the Adirondacks showed similar or slightly larger

angles (Table 1). With increasing iron content in

the Adirondack specimens the angles of both "001"

and "100" lamellae decrease (Fig. 6C, 6D) until

for the most iron-rich specimen obtained, an augite

for which le = 93.5, the "001" angles are l11o

and the "100" pigeonite lamellae are parallel to the

c axis. Since the most magnesian Adirondack pyrox-

ene pair has augite le : 33, more magnesian

pyroxene pairs, believed to have formed under

ti-ilat conditions, were sought and found in the

Belchertown and Cortlandt Complexes (Table 1).

The most magnesian and most dramatic of these is

specimen 447 (Fig. 6,4,) with angles of 122" and

22o and a total angle between the two sets of

pigeonite lamellae of 144".

The relations between the composition of augite

and angles of pigeonite exsolution lamellae in all

augite specimens from Table 1 are summarized in

Figure 7. The two insets in each part illustrate

schematically typical patterns involving the two

pertinent sets of lamellae for magnesian and iron-

rich compositions. Figure 7 shows moderate scatter

and some interesting difterences between the Hudson

Highlands and Adirondack specimens, and among

the Cortlandt specimens; these may represent either

differences in composition or differences in condi-

tions of formation. Overall the correlation between

1 8 TAFFE, ROBINSON, TRACY, AND ROSS

F E I 35 0 A ^ FE 36

FE 57.5 l 1 2 5 - 3 F E 7 2

composition and angle of exsolution lamellae isexcellent. Indeed, in routine thin section petrographyreasonably careful measurements of exsolutionlamellae in metamorphic augites can be applied toFigure 7 to yield rough estimates of Fe/Mg ratio.

Figure 8 is similar to Figure 7 but shows theangles of pigeonite exsolution lamellae in augite

50 Lrm 5 0 a n

Flc. 6. Tracings from photomicrographs of augites with different iron-magnesium ratios showing patterns and angles of exsolutionlamellae. Scale and directions of augite a and c crystallographic axes indicated. FE indicates value of the ratio 100(Fe'* * Fes+ * Mn)/(Mg * F"t* * Fet* * Mn) from Table 1.A' Specimen 447 from Belchertown Complex. Crystal twinned on (100) with single orthopyroxene lamella on twin plane.

Thin and thick stubby lamellae are pigeonite "001" lamellae at 122" to the c axis. Thin abundant lamellae formingherringbone pattern across twin plane are pigeonite "t00" lamellae at22o to the c axis. (D axis is nearly normal to planeof paper, but slightly misoriented, so that angles shown are slightly less than maximum observed angles of 122" and 22o.)

B. SpecimenJ223ftomHudsonHighlands(originalphotomicrographisFig. lofRobinsonetal, lgTl).Coarsevert icalIamellae are (100) orthopyroxene. Thick stubby lamellae are "001" pigeonite at 7160 to the c axis. Very thin lamellaeare "100" pigeonite at 6o to the c axis.

C ' Specimen Ca-17 from Adirondacks. Vertical lamellae are (100) orthopyroxene. Thick stubby lamellae are "001" pigeoniteat ll2.5o to the c axis. Thin lamellae are "100" pigeonite at 30 to the c axis.

D. Specimen Go-2 from Adirondacks. Thick vertical lamellae are (100) orthopyroxene. Thick tapered lamellaepigeonite at 111.5o to the c axis. Very thin lamellae are "l00" pigeonite at 0.50 to the c axis.

plotted against the composition of the coexistingorthopyroxene. This has some practical value be-cause the composition of orthopyroxene can bemore accurately estimated from optical data. It canbe justified only to the extent that the compositionof the coexisting orthopyroxene resembles the com-position of the pigeonite lamellae that participated

- - + O

r'IIII*

a

t 2 2 - 2 250Atn

"oor" A c

o o J a , . O . '

, : , , : , a t . a oo

a .ooooo

I

TI

n io" t i

aoo

a

"loo"n c

T

I

I

.qB U

a

a

o o '

loO (Fez'* Fe3' * Mn)/(Fez'+ Fe3'*Mn * Mg )

Frc. 7. Angles of pigeonite exsolution lamellae in metamorphic augites plotted against 100(Fe* + Fe"* + Mn)/(Fe" * Fe"* * Mn * Mg) of augite determined by electron probeanalyses (large symbols) or by measurement of the gamma index of refraction (small symbols).Insets to left and right show some exsolution patterns of Mg-rich and Mg-poor augites re-spectively. Closed circles: Mt. Marcy area, Adirondacks. Open circles: Monroe quadrangle,Hudson Highlands. Closed squares: Cortlandt Complex, New York. Open squares: Belcher-town Intrusive Complex, Central Massachusetts.

IAFFE, ROBINSON,

loo(Fe2** Fe3* + Mn )/( Fez*+ Fe3' + Mn + Mg )

Ftc. 8. Angles of pigeonite exsolution lamellae in meta-morphic augites plotted against 100 (Fe* * Feq + Mn)/(Fe'* + Fe* + Mn + Mg) of coexisting orthopyroxene.Plot is justified to the extent that orthopyroxene mimicstho composition of pigeonite that was involved in theexsolution process. Symbols same as in Figure 7.

equally with augite in the exsolution process. Thepigeonite and orthopyroxene lamellae observed inthe present study are much too thin to analyzequantitatively with the electron probe. In the onecase from the Bushveld Complex where probeanalyses could be made of both hosts and lamellae(Boyd and Brown, 1969), host orthopyroxene andpigeonite lamellae in augite have fairly similar com-positions.

Calculated Optimal Phase Boundaries

According to the principle of "optimal phaseboundaries" (Bollmann, 1970; Bollmann and Nissen,1968), when two lattices are superimposed in space

TRACY, AND ROSS

there will be planes of dimensional best fit betweenthe lattices that in nature tend to form the boundarybetween them. In the case of simple monocliniclattices with identical b dimensions and fairly similara and. c dimensions, and B angles, very slight rela-tive rotations about the D axis will produce optimalphase boundaries that are perfect with respect tothe dimensional position on the boundary surfaceof equivalent points in the two lattices (Robinsonet al, L971,). The orientation of these optimal phaseboundaries can be calculated with reasonable ac-curacy from any pair of lattice parameters usingsimple trigonometric equations programmed forcomputer (check Robinson, Jaffe, Ross, and Klein,Erratum, 1,971).

The qualitative effects of relative variations inthe c and c dimensions on the orientation of "001"and "100" lamellae have been covered previously(Robinson et aI, l97l). Based on computer calcu-lations, the quantitative effects of variation in aand c dimensions and of the ,B angle are shown inFigure 9. It should be emphasized that the absolutevalues of the parameters are much less importantthan their differences or misfit (symbolized a). If cand F are held constant at 5.27 A and 1O9o foraugite and 5.22 A and 105" for pigeonite, theangle "001" A c for "001" lamellae increasesmarkedly with aa (Fig. 9, left) whereas the angle"L00" A c for '(100" lamellae hardly changes at all.If a and B are held constant at9.76 A and 109o foraugite and at 9.68 A and 105' for pigeonite, how-ever, "100" A c changes markedly with ac but"001" A c does not (Fig. 9, center). For a and cheld constant, the angles of both sets of exsolutionlamellae-that is. "100" A c and "001" A c-increase markedly as AnB decreases (Fig. 9, right).This last effect was not emphasized in our previouswork, but its geometric sense may be deducible fromstudy of Figure 4 of Robinson et aI (1971).

Figure 10 shows, for analyzed augites, the Aa's,Ac's, and ,ArB's from the lattice parameters (Table4) plotted against the iron-magnesium ratios. De-spite some scatter, the general picture is clear. Themagnesian pyroxenes have larger angles of exsolu-tion lamellae because they have larger a and cmisfits and also because they have smaller AB's(see Smith, 1969, p. 23). Even without the latticeparameters presented in Table 4, this was deducedby calculating "best fit" exsolution angles for thesynthetically matched pure end member pairs di-opside-clinoenstatite (CaM$SirO6-MgzSizOo) (127o,

" roo"n c

* : | ."..

[.r

"roo" n "oor"

. . . O '


5 2 7 8- ^ ^ 9

-<-

o o 2 . o 4 0 6 0 8 l o o 0 2 0 4 0 6 0 8 l o o | 2 3 4 5 6

A or8 ' A c,A, APet

Frc. 9. Quantitative effect of variation of differences rn a, c, and B (La, Ac, AB) on theangles of "001" and "100" pigeonite exsolution lamellae. Pairs of values at double circle denoterepresentative absolute values of the indicated parameter (a at left, c at center, B at right)for augite (upper value) and pigeonite (lower value). These values also were the ones heldconstant when varying a parameter in each of the other columns. Left column: efiect of varia-tion of "a misfit" (Aa), with c and B of augite and pigeonite held constant at 5.27 and 5.22,and at 105" and 109', respectively. Center column: effect of variation of "c misfit" (Ac), witha and P of augite and pigeonite held constant at 9.76 and 9.68, and at 1O5" and 109'. Rightcolumn: effect of variation of AB, with a and c of augite and pigeonite held constant at 9.76,9.68, and 5.27, 5.22, respectively.

2 l

22') and hedenbergite-clinoferrosilite (CaFeSirOu-Fe2Si2O6) (1I7", 4" ) (Robinson et al, l9Tl, Table3, Nos. 1 and 3). It is dramatically shown in plotsof a and c dimensions for synthetic pyroxenes inthe Di-Hed-Clinoen-Clinofs quadrilateral (Turnock,Lindsley, and Grover, I973).

The results of calculations of optimal phaseboundaries are given in Table 4. For each pair oflattice parameters, for example, augite host andpigeonite "100" lamellae, orientations are calculatedboth for "100" lamellae and also for hypothetical"001" lamellae (in parentheses) with identical lat-tice parameters. For augites for which lattice pa-rameters have been measured for both pigeonite"001" and "100" lamellae, two pairs of orientations

are calculated: from pigeonite "100" parameters,"100" actual and "001" hypothetical; from pigeonite

"001" parameters, "001" actual and "100" hypo-thetical. It will be noted that for the many specimensin which only pigeonite "100" lamellae were sum-ciently abundant to obtain lattice parameters, thecalculated hypothetical "001" orientations are closeto the optically measured orientations of "0O1"lamellae.

Figure 11 gives a comparison of calculated andobserved angles of exsolution lamellae in the pyrox-

enes in Table 4. Despite some scatter the correlationis excellent and gives very strong confirmation thatthe lamellae did indeed f;orm on optimal phase

boundaries. Slieht differences between observed and

22 JAFFE, ROBIN.'ON, TRACY, /ND ROSS

TasLe 4. Metamorphic Augites and Included Exsolution Lamellae: Lattice Parameters from X-Ray Single Crystal Photographs,Calculated Optimal Phase Boundaries, and Calculated and Observed Lattice Rotations

,1"B A n g l e A n g l e C a l c . o b s .

"001" LAMELLAE

, . R o t a t l o n * *

i ' IOO" HELLAE

. - R o r a t i o n # / /

Angle Angle CaIc. Obs.

D l f f .f ronAug

Sanp le fe*

J o - 8

1 2 9 A u g l t e h o s t ( 9 0 % ) 9 . 7 2 6 8 . 8 7 4P i g e o n i t e " 1 0 0 r r ( I 0 Z ) 9 , 6 2 Iorthopyroxene(100) ( tr)

2 3 1 A u s i r e h o s r ( 8 0 2 ) 9 , 1 3 8 8 . 8 8 2P l s e o n t c e " 1 0 0 " ( 2 0 2 ) 9 , 6 4 4orthopyroxene(100) (r !)

2 6 . 1 A u g i t e h o s t ( 9 0 2 ) 9 . 7 4 8P i . g e o n i r e " 1 0 0 " ( t 0 Z ) 9 . 6 5 9

3 3 . 0 A u s i t e h o s t ( 8 0 2 ) 9 . j 4 3 L g I 4P i s e o n l t e " I 0 0 " ( 5 2 ) 9 , 6 4 90! chopyroxene (100) ( I5Z) 18. 28

3 6 . 5 A u s i r e h o s t ( 9 0 2 ) 9 . j 5 5 5 . 9 L 4P i s e o n l t e " 1 0 0 " ( 4 2 ) 9 . 6 6 6o r t h o p y r o x e n e ( 1 0 C ) ( 6 2 ) 1 8 . 3 0

3 6 ' 4 A u g i r e h o s t ( 9 0 2 ) 9 , i 7 6 B , B g oP i s e o n l r e " I 0 0 " ( 5 2 ) 9 , 6 9 4P i g e o n i t e ' 1 0 0 1 " ( 2 2 ) 9 6 9 5

4 1 . 8 A u g i r e h o s c ( 8 O Z ) 9 . 7 5 9 8 . 9 3 6P i s e o n i t e " 1 0 0 t ' ( r 0 Z ) 9 . 6 8 2orEhopyroxene(100) (102)18. 33

4 6 . 9 A u g i t e h o s r ( 9 O Z ) g . i 4 9 8 , g I 4P i s e o n i t e " I 0 0 " ( 3 2 ) 9 . 6 j IPiseonire "OOL" (1i() 9 699

5 7 . 5 A u s t r e h o s c ( 7 7 2 ) 9 . 7 7 0 8 . 9 4 3P i g e o n i r e " 1 0 0 , , ( 5 2 ) 9 . 6 9 3P i g e o n i t e " 0 0 I " ( 8 2 ) 9 . j 1 60 r t h o p y r o x e n e ( 1 0 0 ) ( 1 0 2 ) 1 S . 4 O

8 2 . 7 A u s i t e h o s t ( 8 4 % ) 9 . 7 8 2 L g j 6P l g e o n i r e " r 0 0 " ( 4 2 ) 9 , 7 0 6P l s e o n i E e " 0 0 1 " ( 8 2 ) 9 , j 2 4o r t h o p y r o x e o e ( 1 0 0 ) ( 4 2 ) I 8 , 4 2 8 . 9 7 2

9 3 . 4 A u g l c e h o s t ( 8 5 2 ) 9 . 7 8 1 9 . O O 2P i s e o n i t e " 1 0 0 " ( 5 2 ) 9 , 7 4 2P l g e o n l E e " 0 0 1 " ( 5 2 ) 9 , 7 2 5o r t h o p y r o x e n e ( 1 0 0 ) ( 5 2 ) 1 8 , 4 3

( 1 . 6 0 ) # ( 1 6 5 ' ' ( r 2 2 , 6 " ' ) r 2 2 " ( , 4 2 ' )

( r . 8 6 ) ( 1 4 . s ' ) ( \ 2 a . 6 " ' r 2 2 . 5 o ( , 3 9 ' )

( 2 . 3 0 ) ( 1 2 . 0 ' ) ( r 1 7 . 9 " ) 1 2 0 ' ( . 2 9 " )

( 2 . 1 2 ) ( r 2 . 9 " ' ) ( 1 1 8 . 8 ' ) 1 1 7 ' ( . 2 8 ' )

( 2 . 3 4 ) ( r 1 . 7 ' ) ( 1 1 7 . 6 ' ) 1 1 5 ' ( . 2 2 ' '

( 2 e 4 ) ( 9 . s ' ) ( 1 1 5 . 4 " ) ( . 1 6 ' )2 . 6 0 1 0 . 7 " 1 1 6 . 6 ' 1 1 6 ' . 1 9 ' . 2 5 + . 1 '

( 2 . 7 6 ) ( r 0 . 1 ' ) ( 1 r 6 . 0 ' ) t r 7 " ( . 2 0 ' )

( 2 . 7 1 ) ( 1 0 . 3 ' ) ( 1 1 6 . 3 ' ) ( . 2 1 " )4 . 3 2 6 , 6 " 1 1 2 . 7 " r r 5 " . 1 4 ' . 0 + . 2 "

( 3 . s 0 ) ( 8 r " ) ( 1 r 3 . 6 ' ) ( . 1 8 ' )4 9 7 5 . 8 ' 1 1 1 . 3 " 1 r 2 . 5 ' , 1 4 ' . 0 + , 2 '

( 3 . 4 5 ) ( 8 . 2 ' ) ( 1 r 3 . 6 " ) ( . 1 3 " )4 . 3 8 6 . 5 ' 1 1 r . 9 ' 1 r r . 5 " + . r 2 0 . 0 + . 2 0

( s . e o ) ( 4 . 9 " ) ( 1 1 0 . 3 ' ) ( . 0 8 " )3 , 9 7 7 . 2 " 1 r 2 . 6 ' I 1 1 ' + . 1 3 ' . 0 + . 2 "

4 . 6 5 1 9 . 0 ' 1 9 . 5 ' . 4 5 ' . 4 5 + . L '

7 . 9 1 1 2 . 0 ' 1 2 . 5 ' . 2 9 ' . 2 8 + . 0 5 "

9 , 6 1 9 . 9 ' 1 2 ' . 2 4 ' . 0 + r '

1 7 . 5 0 5 , 7 ' 8 ' , 1 3 " . 0 + . 1 '

5 . 2 5 1 1 0 6 . 1 r o5 , 1 9 3 r 0 8 . 6 r '

5 . 2 7 5 1 0 6 , 0 3 05 . 2 0 5 1 0 8 . 6 2 '

5 , 2 6 2 t O 5 9 5 '5 2I4 108,73"

5 , 2 5 5 1 0 5 . 9 5 05 . 2 r 7 I O 8 . 6 2 .5 . 2 2 2

5 . 2 4 9 r 0 5 , 9 1 "5 . 2 2 7 1 0 8 . 5 8 "5 . 2 4 5

5 . 2 5 2 r 0 5 . 9 1 05 . 2 4 8 1 0 8 . 8 3 .5 . 2 3 6 1 0 8 . 5 5 "

5 . 2 6 6 r 0 5 . 9 2 05 , 2 3 9 I 0 8 . 6 3 "5 . 2 5 6

5 . 2 6 4 1 0 6 . 0 2 "5 . 2 3 6 r o a . 7 3 .5 . 2 3 2 r 0 8 . 7 3 .

5 . 2 5 6 1 0 5 . 4 8 "5 . 2 3 6 r 0 8 . 8 0 "5 , 2 2 7 I O A . 1 8 "5 . 2 4 9

5 . 2 5 2 r O 5 . 4 0 '5 . 2 5 6 1 0 8 , 5 0 "5 . 2 3 6 1 0 8 . 4 8 '5 , 2 4 6

5 2 4 5 r O 5 , 4 2 '5 . 2 4 5 1 0 8 . 5 3 '5 , 2 3 3 1 0 8 . 4 0 "5 . 2 4 2

Ca-6

Po-13

P o - I 7

1 1 1 . 3 4 . 9 " 5 " . 0 3 ' . 0 + . 2 "( 2 4 . 3 6 ' G , r ' ) ( . 1 0 " )

r A 5 4 6 . 8 ' 7 " . 1 6 ' , 0 + , I '

1 3 , 9 4 7 . 0 " 5 ' . 1 6 ' . G r . 2 "( 1 2 . s 0 ) ( 7 . 8 ' ) ( . 1 5 ' )

2 4 , 9 a 4 . 0 ' 3 ' . I 0 ' . 0 + . 2 '( 1 7 . 2 r ) ( s . 8 ' ) ( . 1 4 ' )

1 0 0 - 0 . 9 " 0 ' - . 0 2 0 , 0 + , 2 '( 2 9 - s 2 ) ( 3 . 4 ' I ( . 0 7 "

- 0 . 0 " 0 ' , 0 0 ' . 0 + . 2 0< 3 8 . 2 ) ( 2 . 7 " ) ( . 0 5 . )

* comPos lE ion o f aug l te hosc ; fe = lOO(Fe+Mn) / (Fe+Mn+Mg) derern ined by e lecr ron probe ana lys ls (Tabre 2) .

/ / Parentheses ind lcare va lues ca lcu la ted f ron , 'hyporhet lca l Iamel lae , , (see tex t ) .

* * R e l a t l v e c o u n t e r - c l o c k u i s e ! o t a t l o n o f l a t t t c € o f p 1 g e o n t r e " 0 0 t " t a m e l l a e w i r h r e e p e c c r o ( O O t ) o f a u g i r e h o s r .

/ / # R e l a r l v e c l o c k w i s e r o t a t l o n o f l a t t i c e o f p i g e o n i r e " 1 0 0 " t a n e r t a e w i r h r e s p e c r t o ( 1 0 0 ) o f a u g l t e h o s t .

1 Recent oPt tca l re -examlnat ion o f Po- I3 and Po-17 has revea led a very very f ine second se t o f "001" p igeon i re lane l lae or ienred ar 112" and 112.5"resPect lve ly ' These aPPear Eo rePresent a second ep tsode o f exso lu r ion ar lower tempera tu ie , s ln l ta l ro second and ih l rd se ts o f "O0 l ' , Iane l laeo b s e r v e d I n s o n e l g n e o u s p v r o x e n e s ( R o s s , R o b i n s o n a n d J a f i e , 1 9 7 2 ) . T h e o b s e r v e d a n g l e s o f r h e s e l a r e f l n e t e e l l a e o f I I 2 " a n d 1 1 2 . 5 " a g r e en o s E c l o s e l y w l r h a n g r e s o f 1 r 1 . 9 ' a n d r r 2 . 6 ' c a r c u r a r e d f r o n l a t t i c e D a r a n e r € r s .

calculated angles of "100" lamellae cauld be dueto changes in parameters since exsolution, particu-larly reduction of Ac; but at the moment this inter-pretation is highly speculative. From the evidenceat hand the relative lattice parameters appear tohave changed only very slightly since exsolution tookplace.

Lattice Rotation

In our earlier paper (Robinson e/ al, 1971) andabove, we pointed out that slight relative rotationof pyroxene lattices allowed improvement of thefit between lattices. We also presented in detail thequalitative eftect of relative parameters on the direc-tion of relative lattice rotation. In the examples we

evaluated earlier, this rotation as calculated fromlattice parameters was so small, generally 0.15' orless, that it could not be accurately measured inprecession photographs.a It was natural, then, toseek evidence for such lattice rotation in the moremagnesian specimens with large exsolution angles,and in this we were successful.

Detection and measurement of lattice rotation inX-ray single crystal photographs is dependent onsize of crystal used and abundance of lamellae.Small angular separations of spots only show up inphotographs of small undistorted crystals. In many

'Precision in measuring angular relationshipsX-ray precession photographs is 0.05' to 0.2'.

from the

PIGEONITE EXSOLUTION LAMELLAE IN METAMORPHIC AUGITE 23

5 lII T- tD t

"ool" n cIt -- l . J

t " loo"n c

'1

"roo" n "oor"

o\"'1 1l lj o

o lo

lo 20 30 40 50 60

loo (Fez'+ Fe3'+ Mn)/(Fe2'+ Fe3'* Mn * Mg )

Ftc. 10. Aa's, trc's, and AB's from X-ray single crystal datain Table 4 plotted against 100 (Fe% * Fe"' * Mn)/(Fe'* + Fe"* + Mn f Mg) of augite host. Vertical linesconnect data points derived from single crystals in whichlattice parameters for two sets of pigeonite lamellae wereobtained. In Aa row large symbols indicate ..actual,' valuesderived from parameters of "001" lamellae, small symbolsindicate "hypothetical" values derived from parameters of"lfr)" lamellae. In Ac row large symbols indicate .,actual"

values derived from parameters of "100" lamellae, smallsymbols indicate "hypothetical" values derived from param-eters of "001" lamellae. Aa and Ac are larger, and Ap issmaller with lower iron content, resulting in larger anglesof exsolution lamellae.

such crystals exsolution lamellae are not sufficientlyabundant to give measurable X-ray reflections, hencelarger crystals must necessarily be used with resultantloss of definition. For these reasons, in the majorityof cases, small lattice rotations could be neitherdetected nor ruled out, but the evidence for latticerotation is decisive in four cases (Table 4).

It will be seen that for "001" lamellae, calculatedrotations all fit into case 1-1 of Figure 6 (earlierpaper) with relative counter-clockwise rotation ofpigeonite lattices. Of the four augites in which X-rayreflections of "001" pigeonite lamellae were ob-tained, the most magnesian (J223) shows relative

loo (Fez*+ Fe3'+ Mn)/(Fe2'+ Fe3'+Mn + Mg)

Ftc. 11. Comparison of optically observed angles of"001" and "100" pigeonite exsolution lamellae in augiteswith angles calculated from measured lattice parametersfor several different iron-magnesium ratios. Vertical linesconnect observed and calculated values for a single speci-men. Solid circle: Optically observed angle of exsolutionIamellae. Thick cross bar: Calculated angle based on latticeparameters of "actual" lamellae. Thin cross bar: Calculatedangle based on lattice parameters of "hypothetical" lamel-lae. Double cross bars: Calculated angle for "001" n

*1t"

based on combination of "actual" and "hypothetical" re-sults.

counter-clockwise rotation of the "001" pigeonitelattice of 0.25 :L 0.1o (case 1-1 ) in close agreementwith the calculated value of 0.19'. For "100" lamel-lae calculated rotations all fit into case 4-l of Fig-ure 6 (earlier paper) with relative clockwise rotationof pigeonite lattices, except for Po-13 which, by0.02', is in case Gl with counter-clockwise rotation.The augites from the three most magnesian speci-mens (447, T52, and A21) show relative clockwiserotations (case 4-1) of 0.42 -f 0.05o, 0.45 :t O.l'

) 1 IAFFE, ROBINSON, TRACY, AND ROSS

and 0.28 t 0.05' in close agreement with calcu-lated rotations of 0.43o, 0.45o, and 0.29" respec-tively. Thus in the four cases where lattice rotationsare large enough to be detected, they are in re-markably good agreement with rotations calculatedindependently from lattice parameters. This agree-ment is a further confirmation of the optimal phaseboundary theory.

Notation of Phase Boundaries

In this paper we have used the notations "001"and "100" to designate the relative orientation ofthe crystal lattices of the host and unmixed phase.As we have previously stated, the phase boundariesof these "001" and "1.00" lamellae are not usuallyparallel to the (001) or (100) host lattice plane,Thus, in the general case, the 3'001" or "100" phaseboundary plane must be described as being parallelto an irrational lattice plane.

In Table 4, in addition to the calculated angles of"001" and "100" exsolution lamellae and the differ-ences from ,B for "001" lamellae, a number w islisted for each calculated orientation. The value uris the non-integral number that represents the posi-tion in the first row of lattice points above the originthat is common to both lattices (Robinson er a/,I97I, p.923). Another way to look at this is thatw is the irrational intercept of an "001" phaseboundary on the c axis when the intercept is 1 on thec axis or vice versa for "100" boundaries. "Miller"indices for the phase boundaries may be derived fromthese irrational intercepts. Because "001" boundariesmay have either positive or negative intercepts ona, and, "100" boundaries, positive or negative in-tercepts on c depending on the relative valuesof aao* and op1q, and clus and cplq respectively(Robinson et ql, 1971, Fig. 5), the orientation ofthe lamellae with respect to both lattices may beaccurately described in terms of the irrational Millerindices as shown in Table 5. The interested readermay wish to ink in the appropriate Miller indices ineach of the six sections of Fizure 5 in the earlierpaper.

Comment on Equilibrium Relations DuringCrystallization and Exsolution of

Metamorphic Augite

Implicit in the work on igneous augites of Polder-vaart and Hess (1951, p. 483), Hess (1960, p. 40),Preston (1966, p. I23O), Boyd and Brown (1968,p. 358; 1969, p. 212), and Smith (1969, p. 24)

Tenre 5. Relation between Relative Lattice Parameters,Irrational Intercepts, and Irrational Miller Indices of Phase

Boundaries in Monoclinic Pvroxenes

In t e rcepts

Mi l le r Ind ices

In t e rcept s

i, l i I1er Indices

.AUG t tPrc

+ w o l

1 o r i

"-A.uc = tPrG

"AUG '

"Prc

l - w @ I

0 0 l l O u

"AUG t

"PrG "AUG = tPrc teuc '

"Prc

f ' t + " 1 o - 1 @ - w

w 0 1 1 0 0 w 0 1

is the idea that pigeonite exsolves from augite onlyunder temperature conditions where pigeonite isthe stable Ca-poor pyroxene above the pigeonite-orthopyroxene inversion loop. In the case of thespecimens considered here, the metamorphic Ca-poor pyroxene coexisting with augite is orthopyrox-ene. The paucity and fine scale of exsolution lamellaein these specimens show that the initial solid solu-tion was very limited as compared to igneouspyroxene pairs of similar iron-magnesium ratio, in-dicating that the metamorphic recrystallization took'place at much lower temp€ratures with subsequentslight amounts of exsolution during further cooling.There seems no question that the pigeonite andorthopyroxene lamellae exsolved at a temperaturewell below that of the pigeonite-orthopyroxene in-version curve. Two hypotheses are suggested aspossible explanations: (1) Nucleation and growthwithin the monoclinic structure of the augite hostfavors metastable formation of pigeonite underconditions where orthopyroxene is the stable phase.(2) Pigeonite (or clinohypersthene) is again thestable phase at lower temperatures of exsolution(Kuno, 1966; Boyd and England, 1965; Sclar,Carrison, and Schwartz, 1964; Smith, 1969; Brown,1968, 1972; Grover, 1972).

The hypothesis of metastable formation in theaugite host is very attractive. The nucleation andgrowth of pigeonite lamellae, particularly "001"lamellae, requires only migration of the 6- and8-coordinated cations without major disruption ofthe monoclinic tetrahedral network that would berequired for formation of orthopyroxene. The hy-pothesis of metastability allows simultaneous pre-cipitation, for which there is some textural evidence,of pigeonite lamellae on "001" and orthopyroxeneon (100) where the monoclinic and orthorhombiclattices have their best mutual fits respectively.


Simultaneous precipitation of pigeonite on "100" isalso a possibility. This explanation would also beapplicable to metamorphic rocks containing coexist-ing calcic clinoamphibole and orthoamphibole, inwhich the exsolution lamellae in the calcic amphiboleare always monoclinic cummingtonite (Ross, Papike,and Shaw, 1969).

A further argument favoring metastable formationhas to do with the optimal phase boundary theoryitself. Orthopyroxene can have a good fit on augiteonly on (100), and this only provided the c and bdimensions of the two phases are nearly identical.Pigeonite, because of its monoclinic lattice, canachieve nearly perfect dimensional fit with both"001" and "100" lamellae provided it has a bdimension in common with the host. Indeed, thepotential for better fit of pigeonite on irrational"100" planes as compared to poor fit of ortho-pyroxene on (100) could be used to explain therelative abundance of pigeonite '(100" lamellae inmost specimens.

In the case of the most iron-rich specimens(Po-13, Sc-6, Po-17), the argument given in thepreceding paragraph can be reversed. In thesespecimens the fine "100" pigeonite lamellae areessentially parallel to (100) and to coarser (100)orthopyroxene lamellae (Table 1, Fig. 7), so thatthere is little or no misfit of the c dimensions ofpigeonite and augite host. In this case one mightask what advantage monoclinic pigeonite essentiallyon (100) would have over orthopyroxene on (100)and consider again the possibility that the pigeoniteis indeed the stable phase under the lowest exsolu-tion temperatures.

Summary and Conclusions

1. In augites of augite-orthopyroxene pairs believedto have exsolved in a metamorphic environmentthere is a strong correlation between the anglesof pigeonite "001" and "100" exsolution lamel-lae and the iron-magnesium ratio, with the largestangles in the most magnesian augites.

2. The angles of exsolution lamellae with respectto the a and c crystallographic axes are relatedto the misfits of the a and c lattice parametersand differences between the p angles of host andlamellae. In magnesian specimens misfits of aand c arc larger, differences between p angles aresmaller.

3. The optically measured angles of the exsolutionlamellae are in close agreement with the ansles

of optimal phase boundaries calculated fromlattice parameters of augite host and pigeonitelamellae. In the most magnesian specimens rela-tive lattice rotations measured from precessionphotographs are in nearly exact agreement withrelative lattice rotations calculated from latticeparameters. These two facts give strong evidencethat the pigeonite lamellae nucleated on "optimalphase boundaries" and that the relative latticeparameters of host and lamellae have not changeda gteat deal since nucleation of lamellae undermetamorphic conditions. This is in stark con-trast to relations observed in some igteous py-roxenes (Ross el al,1972).

4. A simple notation has been devised making useof irrational Miller indices that succinctly de-scribes the orientation of the phase boundaries.

5. Pigeonite lamellae in augite hosts coexisting withorthopyroxene clearly exsolved below the pi-geonite-orthopyroxene inversion temperature. Itis not clear whether pigeonite nucleated meta-stably in augite hosts because of best fit consid-erations or whether pigeonite is again the stablephase under conditions of low temperature ex-solution.

Appendix: Description of Specimens and Their. Geologic Setting

Hudson Highlands

The Hudson Highlands of New York consist of a belt ofhigh-grade, regionally metamorphosed, essentially concordant,Precambrian gneisses and granulites that represent a north-eastern extension of the Reading Prong and the Blue Ridgegeomorphic and petrographic provinces. According to Jaffeand Jaffe (1973), the gneisses and granulites of the HudsonHighlands represent a eugeosynclinal sequence of sedimentaryand volcanic rocks that attained a largely isochemical meta-morphic equilibrium under conditions of the hornblendegranulite facies. According to these authors metamorphismtook place at temperatures of 700-800oC and pressures in therange of 2-4 kbar under relatively dry conditions with PH,6considerably lower than P.o.,1. Similar estimates were obtainedby Dodd (1965) and Dallmeyer and Dodd (1971) from a studyof mineral assemblages in adjoining terrane. Exsolution inpyroxenes, as well as in feldspars, of necessity commencedbelow these maxima, although not necessarily much belowthese values. A further episode of unmixing in the solid stateat still lower T and P conditions could have occurred inPaleozoic and/or Triassic time associated with orogenic eventsthat mildly affected this region (Hall, 1968). Variation in exsolu-tion textures, such as irregularities in size, distribution, andpercentages of both "001", *100" pigeonite, and (100) hyper-sthene lamellae in grains of host augite, from the same smallspecimen suggest that solid state exsolution may have takenplace in more than one stage.

26 IAFFE, ROBINSON, TRACY, IND ROSS

J-241S. Pyroxene plagioclase granulite, similar to I-437P,below.

J-515. Orthopyroxene-quartz-andesine gneiss, quartz 33Vo,K-feldspar 17o , andesine (An,') 55Vo, orthopyroxene 8%,arsgite 7Vo, biotite l%, magnetite 7Vo.

J-223. Equigranular pyroxene granulite, orthopyroxene30Vo, augite 507o, hornblende 5%, quartz 75%o,

l-437P. Pyroxene plagioclase granulite, andesine (An",)65Vo, orthopyroxene l6Va, augite ll7o, hornblende 17o,magnetite 4Vo, qtartz 7Vo, K-feldspar ZVo.

A tlitotttlucl;s

The Adirondack province of northern New York con-sists of a circular area of Precambrian crystalline rocksthat occupy approximately 30,000 square kilometers (Fig.1). The northeastern quarter of this circular region containsa heart-shaped massif of anorthosite, intimately associatedwith charnockitic gneiss and minor amounts of gneiss andgranulite of sedimentary and volcanic ancestry (Kemp,1921; Buddington, 1939; Crosby, 1969; de Waard, 1970,Davis, 1971). All of the exsolved pyroxenes from theAdirondack region described in this report are from thisanorthosite-charnockite complex (Table 1, Fig. 2).

Metamorphic relations in the anorthosite-charnockitecomplex of the northeast Adirondacks are complicatedby the coexistence of apparent pyroxene-hornfels faciescontact metamorphic assemblages (Turner, 1968, p. 224,235, 242) such as calcite-monticellite-forsterite-augite-garnet-spinel and wollastonite-diopside-grossular with gran-ulite facies regional metamorphic assemblages such asorthopyroxene-augite-garnet-plagioclase and quartz-calcite-diopside-grossular. It is not yet known whether the formerrepresent contact metamorphic assemblages formed byanorthosite magma in contact with argillaceous dolomitethat has survived the later regional granulite facies meta-morphism or whether these assemblages resulted fromspecial localized conditions of low activity of COz thatmay have existed during regional metamorphism (Walter,1963). The extremely iron-rich orthopyroxenes reportedhere suggest, according to the experimental data of Smith(1971), that pressures may have been in excess of 7 kbar.

Jo-8. Anorthositic gabbro, sub-ophitic; andesine (Annu)57Vo, andesine antiperthite I9Vo, artgrte l7Vo, ortho-pyroxene 5Vo, ilmenite f biotite I apatite 2Vo.

Ph-3, IV-AI, MD-1, VI-Nk. Similar to Io-8.Pr-l and CD. Gabbroic anorthosite; andesine (Anc),

garnet, orthopyroxene, augite, hornblende, ilmenite, mag-netite in a matrix of andesine and minor K-feldspar.

Go-4. Pyroxenite; augite 65%, orthopyroxene 25%,andesine 3Va, microperthite 'Vo, quartz 3Vo, magnetitel/2%, biori te 1/2%.

Ma-5. Similar to Go-4.Po-1. Gabbroic anorthosite gneiss; andesine (Ana) mega-

crysts l3Vo in a matrix of andesine (An*) 58%, ortho-clase 6Vo, augite 1Vo, orthopyroxene 3%, hornblende4Vo, garnel 3Vo, llmenite 4Vo, magnetite lVo, apalite 77o.

Sb-2. Sl-5. VH-l. VH-2. Similar to Po-l.Ca-6. Melagabbro granulite; andesine 28Vo, microperthite

5Vo, orthopyroxene 23Vo, augite llVo, garnet 7OVo,ilmenite l3Vo, magnetite 4Vo, apatite 6Vo.

Ma-3. Similar to Ca-6.Ca-17. Microperthite-pyroxene granulite; microperthite

5OVo, orthopyroxene 75Vo, zugite 35Vo.lb-z. Microperthite granulite; blue megacrysts of micro-

perthite in a matrix of oligoclase (An*), orthopyroxene,augite, garnet, magnetite, ilmenite, apatite.

Po-13. Ferromangerite gneiss; microperthite 37Vo, oligo-clase (An.e) 33Vo, qtartz 47o, alu9ite 6Vo, orthopyroxene4Vo, hornblende 4Vo, garnet 77o, magnetite { ilmenite3%, apatita 2%.

Go-2, SC-6, Po-17, Giant, Sb-I. Similar to Po-13.

Belcltertown Complex

The pyroxene-bearing specimens from the BelchertownIntrusive Complex are from the extreme inner portion ofthe batholith that has largely escaped major metamorphichydration during the Acadian kyanite-grade regional meta-morphism that affected the country rocks. The pyroxene-bearing rocks thus represent an igneous mineralogy thathas undergone a slight but as yet uncertain degree ofmetamorphic recrystallization. The fact that pelitic schistinclusions in the hydrated part of the batholith up tohundreds of feet thick and several miles long containsillimanite (and sillimanite pseudomorphs of andalusite)shows that the batholith strongly disturbed the localthermal structure and that metamorphic hydration andrecrystallization took place at metamorphic temperaturesconsiderably above those of the kyanite zone. The speci-mens are from two localities collected by David J. Hall(1973) in a regional study of geophysics and petrogaphy,and the mineralogy and petrology of both are being studiedin greater detail by Lewis D. Ashwal (1974). The ex-tremely magnesian compositions of the pyroxenes fromthe Belchertown monzodiorites as compared to rocks ofcomparable feldspar composition in the Adirondacks appear to be due to the extremely high activity of oxygenunder which they crystallized, as indicated by the pres-ence of primary high temperature titano-hematite.

447. Core of ultramafic inclusion 700 feet across de-scribed by Emerson ( 1898, l9l7) as "cortlandtite;"very coarse a[gfte 5A-75Vo, pargasite lO-4AVo, andorange-brown biotil.e 3-l4Eo enclosing grains of augite,orthopyroxene l-77o, olivine (Fa', ? - 1.694) SVo andrutile.

113, 115, 1lO, A2l. Pyroxene monzodiorites typical ofcore of batholith; clouded pink oligoclase (An+a)4O-47Vo, orthoclase microperthite ll-l8Vo, quartz 4-l3Vo, arugite 7-18%, orthopyroxene zl--8%, hornblende0.5-2%, biotite 6-147o, titanohematite and magnetiteO.5-1.07o.

Cortlandt Complex

The Cortlandt Complex near Peekskill, New York, isa mafic to ultramafic layered intrusion of probable Ordo-vician age, according to Long and Kulp (1962). It isintruded into sillimanite-grado schists and gneisses of thenorthern Manhattan Prong and displays a thermal meta-morphic aureole. Ratcliffe (1968) suggested that the Com-plex was intruded after most Taconic age deformation andmetamorphism had occurred. Tracy (1970) found features


in the eastern, ultramafic end of the complex indicative oflater metamorphic recrystallization such as alteration ofpyroxenes to hornblende and biotite, and exsolution ofpyroxenes and oxide phases. This was ascribed to anAcadian reheating event, proposed by l-ong and Kulp(1962) to account for hybrid K-Ar ages in the Man-hattan Prong.

T62, T58, T52, T65. Olivine pyroxenites; orthopyroxene2O4OVo, augite 30-50%, olivine (Fa^) 5-35Vo.

T25. Hornblende-biotite pyroxenite; orthopyroxene 40Vo,angita 30Vo, hornblende 76Vo, biotite 87o, minor ser-pentine probably after olivine.

Acknowledgments

Optical work, computations, and manuscript preparationat the University of Massachusetts and electron probeanalyses at the Institute of Materials Science, University ofConnecticut, and at the Department of Earth and PlanetaryScience, Massachusetts Institute of Technology, were sup-ported by National Science Foundation Grant GA-31989(to Jaffe and Robinson). X-ray studies were done at theU. S. Geological Survey. Electron probe analyses at theU. S. Geological Survey were done by J. Stephen Huebnerand Nelson Hickling, and data reduction was carried outby Mary Woodruff. Field work in the Hudson Highlands(Jaffe and Jaffe) was supported by the New York StateMuseum and Science Service. Geological information andspecimens from the Belchertown Complex were obtainedunder the guidance of David J. Hall and Lewis D. Ashwal.David B. Stewart and Karen Wier Shaw provided percep-tivo reviews of the manuscript. To each of the abovepersons and institutions we express our grateful acknowl-edgment.

References

AsHwAL, L. D. (1974) Metamorphic Hydration ol Augite-O rthopy roxene M onzodiorite to Hornblende G ranodior iteGneiss, Belchertown Batholith, Massachusetts. M.S.Thesis, University of Massachusetts, 116 p.

Bot-rueNN, W. (1970) Crystal Defects and Crystalline In-t e rf ac e s. Springer, Berlin.

eun H. -U. NrsspN (1968) A study of optimalphase boundaries: the case of exsolved alkali feldspars.Acta Crystallogr. A24, 546-557.

Bovr, F. R., lNo G. M. BnowN (1968) Electron-probestudy of exsolution in pyroxenes. Catnegie Inst. Ilash.Year Book, 66,353-359.

AND - (1969) Electron-probe study of py-roxene exsolution. Mineral. Soc. Am. Spec. Pa:p. 2, 2ll-216.

eNp J. L. ExclrNo (1965) The rhombic enstatite-clinoenstatite inversion. Carnegie Inst. llash. Year Book,64, 116-120.

BnowN, G. M. (1968) Experimental studies on inversionrelations in natural pigeonite pyroxenes. Carnegie Inst.Wash. Year Book, 66, 347-353.

(1972) Pigeonite pyroxenes: A review. Geol. Soc.Am. Mem. 132. 523-534.

BuoorNcton, A. F. (1939) Adirondack igneous rocks andtheir metamorphism. GeoI. Soc. Am. Mem.7r 354 p.

Cnossv, Pnncv (1969) Petrogenetic and statistical implica-tions of modal studies in Adirondack anorthosite. N.Y.State Mu:s. Sci. Seru. Mem. lEr 289-344.

DeLrrrevrn, R. D., lNu R. T. Dooo (1971) Distributionand significanc€ of cordierite in paragneisses of theHudson Highlands, southeastern New York. Contrib.Mineral. Petrol. 33, 289-308.

Devrs, B. T. C. (1971) Bedrock geology of the St. Regisquadrangle, New York. N. Y. State' Mus. Sci. Sero., Mapand Chsrt Ser. No. 16, 34 p.

DeBn, W. 4., R. A. HowrE, eNo Jlcr ZussueN (1963)Rock-Forming Minerals, Yol. 2, Chain Silicates. lohnWiley and Sons, New York, 379 p.

DE WAARD, DIm (1970) The anorthosite-charnockite suiteof rocks of Roaring Brook Valley in the eastern Adiron-dacks (Marcy Massif). Am. Mineral. 55,2A63-2075.

Doro, R. T., Jn. (1965) Precambrian geology of thePopolopen Lake Quadrangle, southeastern New York.New York State Mus. and Sci. Seru., Map and ChartS e r . N o . 6 , 3 9 p .

Elraensow, B. K. (1898) Geology of Old HampshireCounty, Massachusetts comprising of Franklin, Hamp-shire, and Hampden Counties. U.S. Geol. Suro. Mon.29, 790 p.

(1917) Geology of Massachusetts and RhodeIsland. U.^9. GeoI. Suru. Bull. 597,289 p.

Gnovrn, JouN (1972) The stability of low-clinoenstatitein the system Mei,SLOe{aMgSLOu. Trars. Am. Geophys.Union, 53, 539.

Gutunte, J. O. (1972) Geology of the northern portionof the Belchertown Intrusive Complex, west-centralMassachusetts. Geol. Dept., Unia. Mass., Contib. No. t,1 1 0 p .

AND PETER RonrNsoN (1967) Geology of thenorthern portion of the Belchertown Intrusive Complex.Guidebook for Field Trips in the Connecticut Valley olMassachusetts, 59th Annu. Meet., New England Inter-collegiate Geol. Conl., Amherst, Mass., p. 143-153.

Hrrr, D. I. (1973) Geology and Geophysics of the Belcher-town Batholith, llest-Central Massachusetts. Ph.D.Thesis, University of Massachusetts, 110 p.

Heu, L. M. (1968) Times of origin and deformation ofbedrock in the Manhattan Prong. In, E-an Zen, and W. S.White, Eds., Studies of Appalachian Geology: Northernand Maritime. John Wiley and Sons, New York. ll7-127.

Hrss, H. H. (1949) Chemical composition and opticalprollerties of clinopyroxenes. Am. Mineral. 34, 621-666.

(1952) Orthopyroxenes of the Bushveld type, ionsubstitutions, and changes in unit cell dimensions. Am.I. Sci., Bowen Vol., 173-187.

(1960) Stillwater igneous complex, Montana. Geol.Soc. Am. Mem. EOr 21O p.

Hown, R. A. (1955) The geochemistry of the charnockiteseries of Madras. Trans. Roy. Soc. Edinburgh, 62,725-768.

Jenne, H. W., .rNo E. B. Jennr, (1973) Bedrock geologyof the Monroe quadrangle, New York. N.Y. StateMuseum and Sci. Seru., Map and Chart Ser. No. 20,Albany, N.Y.

28 IAFFE, ROBINSON, TRACY,,4ND ROSS

PEren RonrNsoN, AND ConNr,I-Is Kt-stx, Jn. (1968)Exsolution lamellae and optic orientation of clinoam-phiboles. Science, 160, 776-778.

Krup, J. F. (1921) Geology of the Mount Marcy quad-rangle, New York. N.Y. State Mus. Bull. 229-230, 86 p.

Knrrz, R,uru (1963) Distribution of magnesium and ironbetween orthopyroxene and calcic pyroxene in naturalmineral assemblages. l . Geol. 71,773-785.

KuNo, Hrsasal (1954) Study of orthopyroxenes fromvolcanic rocks. Am. Mine.ral. 39, 3U46.

-(1966) Review of pyroxene relations in terrestrialrocks in the light of recent experimental work. Mineral.I . (Tokyo) , 5 ,21-43 .

LrNosrpv. D. H.. H. E. KrNc. Jn.. exo A. C. TunNocr(1974) Compositions of synthetic augite and hypersthenecoexisting at 810'C: Application to pyroxenes fromlunar highlands rocks. Geophys. Res. Lett. l, 134-136.

LrNosrEy, D. H., I. D. MecGnecoR, AND B. T. C. Devrs( 1964) Synthesis and stability of ferrosilite. Carnegie Inst.Wash. Year Book, 63, 174-176.

Loxc, L.E'. , eNo J. L. Kurp (1962) Isotopic age study ofthe metamorphic history of the Manhattan and ReadingProngs. Geol. Soc. Am. Bull. 73r 969-996.

Murn, I. D., eNo C. E. TlrrEy (1958) The compositionsof coexisting pyroxenes in metamorphic assemblages.Geol. Mag. 95, 403-408.

Por-oEnvlenr, ARrE, AND H. H. HEss (1951) Pyroxenesin the crystallization of basaltic magma. l. Geol. 59,472-489.

PnesroN, JouN (1966) An unusual hourglass structure inaugite. Am. Mineral. 51, 1227-1233.

Rlrcurrr, N. M. (1968) Contact relations of the Cort-landt Complex at Stony lPoint, New York, and theirregional implications. Geol. Soc. Am. Bull. 79, 777-786.

RorrNsoN, Pnrrn, H. W. JeEEr, Mercorlr Ross, eNlConNnrs KnIN, Jn. (1971) Orientation of exsolutionlamellae in clinopyroxenes and clinoamphiboles: con-sideration of optimal phase boundaries. Am. Mineral.56, 909-939.

(1971) Er ra tum. Am.Mineral. 56, 1836.

Ross, MelcoLM, J. J. Perrxr, eNn K. W. Sn.lw (1969)Exsolution textures in amphiboles as indicators of sub-

solidus thermal histories. Minerol. Soc. Ant. Spec. Pap.2, 275-299.

Psren RosrNsoN, AND H. W. J.IEEE (1972)

Pigeonite lamellae in Bushveld augite sequentially ex-solved on phase boundaries at angles controlled by dif-ferential thermal contraction. Geol. Soc. Amer' Abstractswith Programs 1972, 644-645.

J. S. HuEsNpR, AND ERIc Dowrv (1973) Delinea-tion of the one atmosphere augite-pigeonite miscibilitygap for pyroxenes.,from lunar basalt l2t2l. Amer.Mineral. 5t, 619-635.

SlxENe, S. K. (1971) Mg2+-p.r+ order-disorder in ortho-pyroxene and the Mg"*-Fe'* distribution between coexist-ing minerals. Lithos, 4, 345-356.

Scren, C. B., L. C. ClnnrsoN, AND C. M. Scswln:rz (1964)High-pressure stability field of clinoenstatite and theorthoenstatite-clinoenstatite translation (abstr.). Trans.Am. Geophys. Union. 45, 121.

SneNo, S. J. (1942) Phase petrology of the CortlandtComplex, New York. Geol. Soc. Am. Bull. 53, 409428,

Slurn, Doucres (1971) Stability of the assemblage iron-rich orthopyroxene-olivine-quartz. Am. I. Sci.27l, 374-3E2.

Srvrrrn, J. V. (1959) Crystal structure and stability of theMgSiOa polymorphs; physical properties and phase rela-tions of Mg, Fe pyroxenes. Mineral. Soc. Am. Spec.Pap. 2, 3-29.

Srr,pnENsoN, D. A., C. B. Scun, AND J. V. Srunr (1966)

Unit cell volumes of synthetic orthoenstatite and lowclinoenstatite. Mineral. Mag. 35, 838-846.

Tnecv, R. J. (1970) The Pe'trology of the Eastern End ofthe Cortlandt Complex, Westchester County, New York.M.Sc. Thesis, Brown University, 45 p.

TunNsn, F. I. (1968) Metamorphic Petrology. McGrawHill, New York, 403 p.

TuRNocK, A. C., D. H. LrNosrrv, AND J. E. Gnovrn (1973)Synthesis and unit cell parameters of Ca-Mg-Fe py-roxenes. Am. Minerol. 5E' 50-59.

Werrrn, L. S. (1963) Experimental studies on Bowen'sdecarbonation series. lm. l. Sci. 261,773J79.

Manuscript receiued, Iuly 31, 1973; accepted

lor publication, August 9, 1974.

Orientation of Pigeonite Exsolution Lamellae in Metamorphic Augite ...

Documents

Transcript of Orientation of Pigeonite Exsolution Lamellae in Metamorphic Augite ...