American Mineralogist, Volume 66, pages 5l-69, l98I

Introduction

A classic pattern of regional metanlorphism is pre-sented in Maine by the distribution of isograds whichdelineate zones of metamorphic grade ranging fromsub-chlorite in the extreme northeastern part of thestate to sillimanite-K feldspar in the southwest(Doyle, 1967). Important contributions to the studyof metamorphism of pelitic rocks in Maine have beenmade by Guidotti (1963, 1970a. 1970b, 1974) in theregion near Rangeley and by Osberg (1968, 1971,1974) n the Waterville-Vassalboro area. Guidotti'sstudies have shown that the regional isograds are ac-tually composite features, formed by more than onemetamorphic episode. These overlapping eventscomplicate detailed petrologic studies and make ac-

rPresent address: Air Force Geophysics Laboratory, HanscomA.F.B., Massachusetts 01731.

0003-004x/81/0102-0051$02.00 5l

Metamorphic petrology, mineral equilibria, and polymetamorphismin the Augusta quadrangle, south-central Maine

Jauns M. Novm'AND M. J. HoloewaY

Department of Geological SciencesSouthern Methodist (Jniversity, Dallas, Texas 75275

Abstract

A study of pelitic metamorphism in the western 70 percent of the Augusta quadrangle,

Maine, has shown that this metamorphic terrane consists of products of a series of over-

lapping thermal events. Prograde isograds have been delineated on the basis of discontinuous

reactions. Continuous reactions are also described, based on Fe-Mg-Mn phase relations andgarnet zoning patterns.

Two approaches have been used to estimate the pressure and temperature of the lower sil-

limanite zone. Fe-Mg Ko's from garnet-biotite pairs were used to estimate the temperature

of metamorphism. Reactions involving staurolite were also used after adjusting their experi-

mentally determined reaction curves for variations in mineral compositions and the composi-

tion of the fluid phase. Assuming the presence of an independent fluid phase, calculationsshow that X(H2O) was on the order of 0.75i0.10. Results indicate that the average temper-

ature and pressure of the lower sillimanite zone were 570+40oC, 3.8+l kbar.Field relations indicate that the metamorphism, both prograde and retrograde, was closely

associated with the emplacement of felsic plutonic bodies. The sequence of events postulated

for the Augusta area is as follows: (l) a widespread andalusite-biotite-staurolite-producingevent of uncertain cause, was followed by (2) a sillimanite-producing event whose heat source

was southeast of the area. This event produced widespread slight retrogressive metamor-phism to the north. (3) The final metamoprhic event was also sillimanite-producing, with a

heat source supplied by a group of plutons west of the Augusta quadrangle. This event

caused localized intense retrogressive metamorphism along a NNE-trending zone in the west-

ern part of the Augusta quadrangle.

curate estimates of pressure-temperature relationsdifficult.

We report the results of a moderately detailedstudy of metamorphism in the western 70 percent ofthe Augusta l5-minute quadrangle in south-centralMainq (Fig. l). Mineral assemblages in the peliticrocks indicate that amphibolite facies conditions pre-vailed throughout most of the area. The results showthat polymetamorphism played an important role inshaping the metamorphic history. Various reactionshave been deduced, based on AFM topologychanges, garnet zoning patterns, and element distri-bution between minerals. Reaction isograds havebeen drawn on the basis of these reactions. We hopethat this work will help provide insights into themetamorphism in Maine through a better under-standing of the types of processes which occurred inthis area.

52 NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY, SOUTH-CENTRAL MAINE

monzonite. Dallmeyer and Vanbreeman (1978) re-port Rb Sr whole-rock isochron ages for the Hallo-well Pluton at Augusta and the Togus pluton to theeast of Augusta as 387+ll and 394t8 m.y. respec-tively. Possibly the Skowhegan pluton at the north-eastern edge of the map area is of comparable age.

44 30 Moench and Zartman (1976) report a Rb-Sr whole-rock age of 379+6 m.y. for the Mooselookmeguntictwo-mica granite about 60 km northwest of the maparea.

Petrology

In discussing the petrology of pelitic rocks, we willfirst summarize the petrographic characteristics of

qq"E, the minerals, then discuss the mineral chemistry, andfinally relate the textures and mineral assemblages tothe metamorphic pattern in the area. Poorly exposed

Fig. l. Locality map of study area. Random line pattern :plutonic igneous rocks, Si : high-grade side ofsillimanite isograd,Z = localities where zoisite or clinozoisite occurs in calcareousrocks. Osberg (1968) worked mainly on pelites in the northwestpart of the indicated area; Ferry (1976a,b) worked mainly oncalcareous rocks in the southeast part ofthe area. Distribution ofplutonic rocks is based on Preliminary Lithologic Map of Maine(unpublished, compiled by P. H. Osberg, 1973, Maine GeologicalSurvcy).

During the summer of 1977 reconnaissance geo-logic mapping and more detailed specimen sampling(Fig.2) were undertaken. The degree ofoutcrop andthe reconnaissance nature of the mapping did not al-low us to make any meaningful contribution to thestratigraphy and structure of the area (Novak, l9?8).

Geologic setting

The Augusta area lies on the southeast limb of theMerrimack Synclinorium, one of the major structuralfeatures of New England. The general geology andstratigraphy of the Merrimack Synclinorium inMaine have been described by Osberg et al. (1968)and more recently by Pankiwskyj et al. (1976). Anoverview of the deformation, plutonism, and poly-metamorphism in western Maine is given by Moenchand Zartman (1976).

Our samples were collected from pelitic layers ofthe Silurian Waterville and Mayflower Hill Forma-tions (Osberg, 1968; Novak, 1978). The calcareousVassalboro Formation is locally exposed within thestudy area. Igneous rocks consist mainly of two plu-tons of biotite-muscovite-garnet-bearing qnaraz

7o'oo'

, /

N /

Ia l

. "1

' /4

/ v

l o\9- -

to

2 ' n ,

l o

4

- 5

t 7 5 /

1,

I M I L E

7 0 " 0 0 '

Fig. 2. Pelitic mineral assemblage and isograd map of thewestern 70 percent ofthc Augusta lS-minutc quadrangle. Randomline pattern : biotite-muscovite-gamet quartz monzonite, heavystipple : zone ofpartially retrograded rocks, italicized numbers =localities of specimens whose minerals were analyzed. Mineralabbreviations for this and other figures: Ga : almandine-richgarnet, Ch : chlorite, St : staurolite, ful : andalusite, Cd =cordierite, Si : sillimanite, Qu : quanz, Bi : biotite, Mu -muscovite. Metamorphic zones: G. Z. Earnret zone, ST. Z. :staurolite tnne, L. S. Z. = lower sillimanite zone, U. S. Z. : uppersillimanite zone.

t"tt i ienlt lSSetuternOeS*

O Go-Ch '

I S l - G o - C h '

l o S l - G o - C h - A n '

l b S l - C h - A n - C d

l c C h - A n

l d C h - A n - C d

2 s i - s r - c o '

2 0 S i - 5 1 - G o - A n '

2 b S i - S l - A n - C d

J S i - G o

3 0 S i - G o - A n

+ P l u s O u - B i - M u

e G o o r C h h i s s i n g i n

s o m e s p e c r m e n s

NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY, SOUTH.CENTRAL MAINE

rocks of the garnet zone occur in the northeasternpart of the area (Fig. 2). From here, grade increasesto the west and south. Around the garnet zone is abroad staurolite zone. At the western edge are rocksof the lower sillimanite zone (staurolite present).Similar grade rocks are exposed in the south-centralpart. [n the southeastern and northwestern parts areupper sillimanite-zone rocks (without staurolite).Parallel to the western sillimanite isograd is a zone ofpartial retrograde metamorphism within the stauro-lite zone (Fig. 2).

Petrography

In all pelitic rocks brown or red-brown biotite andmuscovite are dominant matrix minerals, definingschistosity as fine imperfectly oriented grains. Withincreasing grade some biotite takes on a slightly por-phyroblastic habit with weaker preferred orientation.Chlorite occurs as cross-cutting plates larger than thesurrounding micas. Chlorite is present in the garnetand staurolite zones; the only chlorite seen in eithersillimanite zone is associated with cordierite in asingle specimen. Chlorite is most abundant in the ret-rograde zone, where it commonly rims staurolite.

Staurotte is a common mineral of the stauroliteand lower sillimanite zones. It occurs as subhedral toeuhedral porphyroblasts, locally exhibiting sieve tex-ture due to abundant quartz inclusions. Pelitic layersshow twinned crystals on the weathered surface.Staurolite crystals are 2 mm to 2 cm long and showno sign of rotation during growth.

Garnet occurs as subhedral to euhedral grainsabout I mm in diameter. It is present in most peliticrocks, regardless of grade. Quartz, ilmenite, andgraphite inclusions are common and in some casesare so abundant they form a continuous boundarybetween the rim and core (Boone, 1973, Plate 2OB).Garnet in one specimen shows evidence of rotationduring growth. Most garnet crystals show post-kinematic growth, with very minor rotation orstretching due to late deformation associated with thefinal stages of metamorphism.

Andalusite is widely distributed throughout thestaurolite and lower sillimanite zones and is evenseen in some specimens from the upper sillimanitezone. It occurs as large irregular porphyroblasts com-monly highly sieved by quartz and micas. Highlysieved cordierite with chlorite rims was found in twospecimens (Fig. 2).

Sillimanite occurs in the southern and far westernpart of the area as fibrolite with biotite and as coarserprismatic crystals. The occurrence of sillimanite con-

trasts with that of andalusite in that every specimenof the lower and upper sillimanite zones contains sil-limanite, while only half of the specimens in thestaurolite and lower sillimanite zones contain an-dalusite (Fig. 2). As discussed in a subsequent sec'tion, andalusite and cordierite are believed to be me-tastable relict minerals from a metamorphic episodeprevious to the events which produced sillimanite-grade metamorphism in the western and southernparts of the area. (Holdaway has collected a numberof upper sillimanite-zone specimens from the Wayneand Fayette 7Vz-minule quadrangles immediatelywest of the Augusta l5-minute quadrangle.)

Ilmenite, occurring in every thin section, is themost abundant accessory mineral. Osberg (1974) de-scribes magnetite from Mayflower Hill and Water-ville Formation specimens northeast of the area stud-ied. These rocks appear to be marly pelites. To checkfor magnetite, 12 polished sections of pelites contain-ing opaque minerals were examined on the micro-probe using qualitative Ti and Fe analyses. Between20 and 40 grains were checked on each specimen. Il-menite was the only opaque oxide.

Graphite occurs in most pelitic rocks as smallamounts of finely disseminated specks generally dis-tributed evenly throughout, but it sometimes concen-trates in thin layers in an otherwise normal peliticrock. The presence of graphite is best demonstratedin a polished section by a combination of fine opaquespecks when viewed with transmitted light and tinypits in the same area when viewed with reflectedlight. Graphite is either less abundant or occurs inlarger crystals (or both) in sillimanite-grade rocks. Insome specimens one cannot be sure that graphite ispresent; however, it is never possible to demonstrateits absence in these rocks. Traces of pyrite and/otpyrrhotite are present in most rocks, and chalcopyriteis locally associated with pyrrhotite. Other accessoryminerals include tourmaline, apatite, zircon, and ru-tile.

Mineral analyses

Minerals in twelve polished thin sections (Fig. 2,Table l) were analyzed with a MAC-400 three-spec-trometer automated electron microprobe at Massa-chusetts Institute of Technology. A 15 kV accelera-ting potential was used with a 0.03 microamperebeam current and a beam diameter of about 2 mi-crons. Muscovite, biotite, staurolite, garnet, chlorite,and cordierite were analyzed for Si, Al, Ti, Fe, Mg,Mr., Zn, Ca, K, Na, and Ba. Silicate standards wereused for most of the elements, and analytical data

NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY. SOUI:H.CENTML MAINE

Specinen

Table l. Model analyses of microprobe specimens* pend on the mineral assemblage and distance fromthe isograd.

Micas. Chemical analyses of muscovite and biotiteare given in Tables 2 and,3 respectively. As noted byTracy (1978), pelitic muscovites closely approximatethe ideal composition with two octahedral cations.Biotites in these high-Al rocks are aluminous in na-ture and comparable to those analyzed by Guidotti(1974). Because of the prevalence of ilmenite andgraphite and the absence of magnetite in the peliticrocks, most of the Fe in the silicates is assumed to bein the divalent state. A plot of FelMg in biotite vs.muscovite (Fig. 3) shows nearly equal distributionbetween the minerals. Ti tends to concentrate in bio-tite over muscovite, but both minerals show a slighttrend toward Ti enrichment with increasing grade.

Garnet. Garnets were analyzed by making a tra-verse across a grain with an analysis at least every150 microns, depending on the size of the grain.Chemical analyses (Table 4) give core and rim com-positions. Cores are commonly enriched in MnO rel-ative to rims. Cores range from 8 to 25 mole percentspessartine while rims range from 7.5 to 19 mole per-cent. Cores are also enriched in grossular and de-pleted in almandine relative to the rims. Plots of al-mandine, pyrope, spessartine, and grossular contentfor two zoned garnets are given in Figure 4.

Staurolite. Staurolite compositions are given inTable 5, with the formula unit based on that of Grif-fin and Ribbe (1973). The range in Fel(Fe+Mg) is0.80 to 0.87. ZnO shows an inverse relationship withthe amount of staurolite in the rock (Tables l, 5), im-plying that if all the ZnO is in staurolite, the peliticrocks contain about 0.02Vo ZnO. The Fe-Mg distri-bution between staurolite and biotite (Fig. 5) shows areasonably consistent pattern, suggesting that equi-libriunr was approached in the specimens analyzed,.

Chlorite. Analyses of chlorite are given in Table 6.In all three analyses chlorite has Fel(Fe+Mg) lowerthan that of the coexisting biotite, suggesting that itgrew in equilibrium with the biotite (Guidotti, 1974).

Fisher (1941) describes a chloritoid phyllite, the lo-cality of which is "one quarter of a mile northwest ofthe town of Winthrop," and traces it "northwardalong strike into the Belgrade Lakes area." We havecollected many specimens from this area and exam-ined them in thin section, but no chloritoid has beenidentified. As the area described by Fishei corre-sponds to the area of retrogressive metamorphismshown in Figure 2, it must be concluded that some ofthe chlorite was misidentified as chloritoid.

Cordierite. Table 6 gives a single analysis of cor-

7 7 1 5 7 \ H - I 4 t9 1

M u s c o v i t e

B i o t i t e

Garn et

S t a u r o l i t e

C h l o r i t e

C o r d i e r i t e

S i l I i m a n i t e

A n d a l u s i t e

Q u a r t z , p l a g

I lneni te

Specinen

3 3

5 2 I

4 l

1 . 5

t r

8 . 0

2 8 9

t T

r 7 5

t 1 5

3 6 . 3

t r

t r

6 7

t r

9 . 2

3 0 I

s l

1 0 . 5

3 8 . 3

t r

t r

7 . 4

l 5 4

4 4 0

t r

5 9

! r

3 6

2 7 0

2 5

1 0 2

2 0 6

l 0

2 1

0 8

3 l

5 3 5

l 4

4 1 . 9

1 7

0 5

t 8 6

3 1 . 5

1 2

Erv- 3

l 7

4 9 . 4

t r

t 2

2 5

4 3 5

t I

2 0 5

2 7 . 9

4 . 9

8 0

1 2 3

2 7 2

2 . 0

MA-4

7 3

4 t 8

1 9

t r

t T

3 8

4 3 I

t 7

t 2 l

4 0 . 5

t r

7 . 0

1 9

t r

3 J 1

t r

1 0 7

2 . 2

3 9 s

t r

t r

l 9 4

3 6 . 9

t r

2 9 . 8

l 3

5

3 0 3

3 4 5

t r

7 6

t r

4 l

. I 8 8

5 0

S p e c i n e n s l i s t e d i n a p p r o x i n a t e o r d e r o f i n c r e a s i n g g r a d e R e p l a c i n gc h l o r i t e n o t c o u n t e d . A c c e s s o r y g r a p h i t e , a p a t i t e , t o u r m a l i n e , z i r -c o n , p y r i t e , p y r r h o t i t e , a n d c h a l c o p y r i t e .

were reduced using the method of Bence and Albee(1968). Each analysis was for 30 seconds or 60,000counts. Estimated uncertainty is 2 percent of theamount present for major elements and 5 percent forminor elements. All analysis points for a given speci-men were within 2 mm of each other. For each min-eral except garnet, I to 5 grains were analyzed. A to-tal of 3 to 5 analyses of each mineral was averaged.Homogeneity within the analysis area is indicated byaverage standard deviations for FeO and MgO of0.25 and 0.10 for biotite, 0.21 and 0.04 for staurolite,0.30 and 0.24 for chlorite, and 0.21 and 0.09 for cor-dierite. These figures imply a maximum variation oftl mole percent in Fel(Fe+Mg). For each garnet-bearing rock, a single zoned garnet was analyzed 6 to10 times at regular intervals across the grain.

Many of the analyzed specimens were collectedfrom the higher grade part of the lower sillimanitezone near the staurolite-out isograd (Fig. 2), and theycontain similar mineral assemblages. This was donein order to evaluate compositional effects on Fe-Mgexchange reactions and to estimate the conditions ofthe lower sillimanite zone. More detailed sampling isneeded from adjacent areas in order to define system-atic mineralogical changes, but in a general way theminerals show element variation trends which de-

NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY, SOUTH-CENTRAL MAINE

Table 2. Chemical analvses of muscoviter

55

7'l 1 0 79 I trl, l-3 I\1A-4

s i o .

A 1 ^ 0 ,

T i 0 2

F e O

IlgO

I\ ln0

K : O

N a 2 0

B r O

' l o t a 1

S i

A I

A I

T i

F e

Mg

M n

I

K

N A

Ba

I

F e / F e + l i g

K / K + N a

A s s e n b I a g e

4 1 1 , 5 4 7 3 0

3 6 7 7 3 6 . 9 5

0 . 3 7 0 4 2

l 1 6 1 . 0 4

0 . 4 8 0 s 0

0 . 0 0 0 0 4

9 6 8 9 7 1

r r 7 1 . 7 2

0 , 1 9 0 . 4 7

9 7 2 7 9 1 5 2

6 1 3 1 6 r 3 2

1 - 8 6 9 1 8 6 8

3 . 7 6 6 3 7 7 5

0 034 0 038

0 1 2 5 0 . 1 1 0

0 0 9 1 0 . 0 9 3

0 . 0 0 0 0 0 0 1

4 . 0 1 6 4 . 0 1 1

1 . 6 0 2 1 6 0 3

0 . 2 9 3 9 . 2 7 9

0 . 0 2 2 0 0 0 2

1 . 9 1 7 1 . 9 0 4

0 5 7 9 0 . 5 4 2

0 . 8 4 5 0 8 5 2

1 a 1

4 7 1 7 4 7 . , 1 8

3 5 7 8 3 5 5 5

0 4 5 0 . 3 7

0 9 8 1 1 4

0 . 6 0 0 6 6

0 0 1 0 . 0 0

1 0 . 0 1 I 3 3

0 9 5 0 9 8

0 9 4 0 . 2 5

9 6 8 9 9 6 2 6

6 1 8 5 6 2 2 8

I 8 1 5 1 . 7 7 2

3 7 t 4 3 7 2 5

0 042 0 034

0 1 0 5 0 . 1 2 3

0 . 1 1 4 0 . 1 2 7

0 0 0 0 0 . 0 0 0

3 9 7 5 4 . 0 0 9

1 671 r 644

o 2 4 1 0 . 2 5 0

0 0 4 6 0 . 0 1 1

1 9 5 8 1 . 9 0 5

0 4 8 0 0 . 4 9 2

0 8 7 4 0 8 6 8

1 a ] a

4 6 3 9 4 7 0 7

5 6 7 5 3 6 . 3 3

o 2 7 0 . 5 8

0 9 4 0 l t 7

0 . 4 5 0 . 5 8

0 0 0 0 . 0 0

, q . s 3 9 1 6

I 2 9 t . 2 6

0 4 7 0 3 6

9 6 8 9 9 5 9 9

4 6 3 5 4 6 5 1 4 6 . 7 7

36 36 36 30 36 .32

0 3 9 0 4 2 0 . 5 1

0 8 6 0 9 5 0 . 9 6

0 5 2 0 4 4 0 4 2

0 . 0 0 0 0 0 0 0 1

9 . 2 2 9 . 0 2 9 6 2

1 . 5 5 1 5 1 I . 3 4

0 9 7 0 . 0 4 0 2 8

9 6 7 0 9 5 7 0 9 6 . 2 3

3 . 7 5 5 3 8 0 7 3 7 6 1

0 . 0 2 7 0 0 3 9 0 . 0 4 9

0 0 9 2 0 . 1 0 2 0 1 0 4

0 . 0 9 9 0 . 0 8 4 0 . 0 8 0

0 0 0 0 0 0 0 0 0 . 0 0 0

3 9 8 3 , 4 0 3 2 3 9 9 4

1 . 5 4 0 ] . 5 1 1 1 . 6 0 8

0 . 3 8 7 0 3 8 3 0 . 3 3 7

0 0 4 7 0 . 0 0 1 0 . 0 1 2

1 . 9 7 4 I 8 9 5 r J 5 7

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

0 7 9 9 0 7 9 8 0 . 3 2 7

2b 2a 2a

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

3 6 6 8 3 5 4 3 3 5 . 1 5

0 . 5 4 0 . . 1 3 0 . 5 0

1 . 0 8 1 . 0 5 0 7 1

0 5 8 0 . 7 2 0 . 5 4

0 . 0 0 0 . 0 0 0 . 0 0

9 . 7 9 1 0 1 1 9 . 6 9

0 8 6 0 . 8 6 0 7 a

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

96.67 95 . 0 .3 94 .90

6 0 9 2 6 . 1 4 1 6 . 2 3 7

1 . 9 0 8 1 . 8 5 9 1 . 7 6 3

3 . 7 5 7 3 . 7 1 7 3 . 7 1 9

0 . 0 4 9 0 . 0 3 9 0 . 0 3 9

0 1 2 3 0 . 1 1 3 0 1 0 1

0 . 1 0 7 0 . 1 3 7 0 . 1 0 4

0 0 0 0 0 . 0 0 0 0 . 0 0 0

4 0 3 6 4 . 0 0 6 3 . 9 7 3

1 . 6 5 8 1 - 7 2 0 1 . 6 3 5

0 2 2 2 0 . 2 3 8 0 . 2 6 7

0 . 0 2 1 0 0 2 3 0 0 1 9

1 9 0 1 1 . 9 8 1 1 . 9 2 1

0 . 5 3 5 0 4 5 2 0 . 4 9 3

0 . 8 8 2 0 . 8 7 8 0 . 8 6 0

2 3 a 3 a

6 . 1 2 5

I 8 7 5

3 . 7 8 3 3 . 7 8 2

0 . 0 2 5 0 . 0 3 5

0 r 0 0 0 0 9 3

0 . 0 8 5 0 . 1 1 0

0 0 0 0 0 . 0 0 0

3 9 9 3 4 . 0 2 0

1 6 3 6 I . 5 3 0

0 3 2 5 0 . 3 1 8

0. a22 0 . 0 l s

1 . 9 8 3 1 . 8 6 3

0 . 5 4 1 0 4 5 8

0 8 3 4 0 . 8 2 8

t a I

Fornu la based on 22 oxygens

6 . 1 7 0 6 1 3 9 6 l l 0 6 1 4 0

I B 3 0 1 . 8 6 1 1 . 8 9 0 1 . 8 6 0

* L o c a l i t i e s a n d n i n e r a l a s s e m b l a g e s g i v e n i n F i g u r e 2

f l o l d a w a y . T o t a l F e g i v e n a s F e O .

dierite which has Fel(Fe+Mg) of 0.30. The specimenis mineralogically similar to that which occurs atWest Sidney in the northeastern part of the Augustaquadrangle (Osberg, l97l), except that the West Sid-ney locality is near the beginning of the staurolitezone and the present specimen (175) occurs in thelower sillimanite zone (Fig. 2).

Petrologic analysis

Figure 2 shows that the map area can be dividedinto five types of areas on the basis of the dominantmineral assemblage. These are (l) a garnet zonecharacterized by garnet without staurolite, (2) astaurolite zone characteized by staurolite withoutsillimanite, (3) a partly retrograded area of thestaurolite zone characterized by chlorite rims onstaurolite and locally by retrograde pseudomorphsafter staurolite and andalusite, (4) a lower sillimanitezone characterized by sillimanite and staurolite, and

N u n b e r s : a n t l y z e d b y N o v a k ; l e t t e r - n u n b e r c o n b i n a t i o n : a n a l y z e d b y

(5) an upper sillimanite zone characterned by sil-limanite and muscovite with neither staurolite nor Kfeldspar. The staurolite isograd is not well defineddue to poor exposure and will not be discussed in de-tail. Andalusite may be present in rocks of any gradeexcept presumably those of the garnet zone.

All minerals in pelitic rocks may be described interms of the system Al'O'-FeO-MgO-MnO-HrO af-ter accounting for the component oxides contained inquartz, muscovite, plagioclase, and ilmenite presentin all specimens. In the discussions which followpresence of a fluid phase is ossum€d;

Most specimens in the staurolite zone containchlorite. Two assemblages which are widely devel-oped throughout this zone (Fig. 2) are biotite-stauro-lite-garnet-chlorite and biotite-staurolite-garnet-chlorite-andalusite. Most of the remaining localitiesare subassemblages of these two. The existence ofgarnet in these assemblages is consistent with their

56 NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY, SOUT:H-CENTRAL MAINE

Table 3. Chemical analyses of biotite*

MA-4 107EW- 31021 7 54 l9 1AH- 17 7 157147

sio2

Al 2o3Tio2

FeO

Mgo

MnO

Kzo

Na20

BaO

Tota l

S i

A I

A 1

T1

Fe

Mg

Mn

t

K

Na

Ba

x

3 5 . 6 3

1 9 . 8 7

1 . 6 5

2 0 . 0 6

9 . 3 8

o . 0 7

9 . 2 s

0 . 2 2

0 . 1 8

9 6 . 3 1

5 . 3 7 1

2 . 6 2 9

0 . 8 9 9

0 . 1 8 5

2 . 5 3 0

2. ro9

0 . 0 0 8

5 . 7 3 1

I . 7 7 7

0 . 0 6 1

0 . 0 0 8

1 . 8 4 6

3 6 . 4 1

1 9 . 9 1

r . 6 2

1 7 . 5 8

1 0 . 7 8

0 . 1 4

8 . 8 0

0 . 3 2

0 . 1 5

9 5 . 6 8

s 4 . 9 6

2 0 . 0 2

1 . 9 s

2 1 . 1 3

8 . 2 9

0 . 0 2

8 6 7

0 . 3 2

0 . 0 7

9 5 . 4 3

5 . 3 3 1

2 . 6 6 9

0 . 9 2 6

0 . 2 2 1

2 . 6 9 3

1 . 8 8 3

0 001

5 . 7 2 4

1 . 6 8 5

0 . 0 8 9

0 . 0 0 3

1 . , 7 7 7

0 . s 8 9

0 . 9 5 0

3 6 . 0 9 3 6 . 9 3

1 9 . 5 8 1 9 5 8

I . 7 5 1 . 5 8

1 8 3 4 1 7 . 0 5

1 0 . 5 9 I t . 2 6

0 . 1 4 0 . 1 1

9 . 0 8 9 . 2 6

0 . 1 4 0 . 1 8

0 . 1 2 0 . 2 4

9 5 . 7 0 9 6 . 1 7

5 . 4 1 1 5 . 4 7 8

2 . 5 8 9 2 . s 2 2

0 . 8 6 8 0 . 8 9 9

0 . 1 9 5 0 . I 7 4

2 . 2 9 7 2 . I L 4

2 . 3 6 5 2 . 4 9 0

0 . 1 7 0 0 . 0 0 9

5 . 8 9 3 5 . 6 8 6

I . 7 3 6 t 7 5 1

0 . 0 3 8 0 . 0 s 0

0 . 0 0 1 0 . 0 1 1

r . 7 7 5 1 . 8 r 2

0 . 4 6 6 0 . 4 5 9

0 9 7 9 0 . 9 7 2

1 l a

3 6 . 2 1 3 5 4 9

1 9 . 6 7 1 9 . 6 4

r . 6 8 t . 6 4

1 8 . 2 5 Z O . l 3

1 0 . 3 9 9 3 7

0 . 1 1 0 . 0 4

9 . 0 9 9 . 0 3

0 1 8 0 3 3

0 . 0 5 0 1 0

9 5 . 6 1 9 5 , 7 7

3 6 . 5 7 3 5 . 3 7

1 9 . 9 7 t 9 . 7 4

I . 4 5 I . 6 4

1 6 . 7 7 2 0 . 1 5

1 1 . 7 0 9 . 3 1

0 . 1 3 0 . 0 6

8 . 8 5 I . r 2

0 . 2 8 0 . 2 7

0 . 0 5 0 . 0 4

9 5 . 7 7 9 5 . 7 0

3 6 . 2 4 3 5 . 4 3 3 6 . 0 0

2 0 . 0 8 1 9 . 6 2 1 9 . 1 1

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

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

9 . 4 1 9 . 9 5 1 0 s l

o . 2 3 0 . r 4 0 . 1 3

8 . 7 8 9 . 1 8 9 . 2 2

0 . 6 0 . 0 2 6 0 . 2 6

0 0 4 0 0 3 0 . 1 6

9 5 . 7 8 9 4 . 3 7 9 5 . 0 5

s . 4 4 6 s 3 8 6 5 . 4 4 2

2 . 5 5 4 2 . 6 1 4 2 . s 5 8

1 . 0 0 0 0 . 9 0 2 0 . 8 4 4

0 . 2 4 2 0 . 2 5 0 0 . 1 8 9

2 . 3 0 2 2 . 2 3 1 2 . 2 7 1

2 . 0 9 2 2 . 2 s 4 2 . 3 6 6

0 . 0 1 3 0 . 0 1 8 0 , 0 1 4

5 . 6 4 9 5 . 6 S s 5 . 6 8 4

1 6 7 8 L . 7 8 I r . 7 7 5

0 0 4 1 0 . 0 8 5 0 . 0 7 3

0 . 0 0 1 0 . 0 0 0 0 . 0 0 7

r . 7 2 0 1 , 8 6 4 1 . 8 5 5

0 . s 2 4 0 . 4 9 7 0 . 4 9 0

0 . 9 7 6 0 . 9 5 6 0 . 9 6 1

2 3 a 3 a

Fornula based on 22 oxygens

5 . 4 3 1 5 . 3 7 6 5 . 4 3 4 5 . 4 2 7

2 . 5 6 9 2 . 6 2 4 2 . 5 6 6 2 . s 7 3

5 . 3 5 9

2 . 6 4 1

F e l F e + M g 0 . 5 4 5

K/ K+Na 0 .967

Assemblage Ia

0 . 9 0 8 0 . 8 8 1

0 . 1 8 9 0 . 1 8 4

2 . 2 8 8 2 . s 4 9

2 . 3 1 9 2 . I t 6

0 . 0 1 3 0 . 0 0 2

t . / r / 5 . / . 5 1

1 . 7 3 8 | 7 4 3

0 . 0 4 8 0 . 0 9 5

0 . 0 0 0 0 . 0 0 3

t . 7 8 6 l . 8 4 1

0 . 4 9 7 0 . 5 4 6

0 . 9 7 3 0 . 9 4 8

l a l a

0 . 9 2 0 0 . 8 8 8

0 . 1 6 0 0 . 1 8 4

2 . 0 8 0 2 . 5 5 5

2 . 5 8 6 2 . 7 0 3

0 . 0 1 4 0 . 0 0 5

5 . 7 6 0 5 . 7 3 5

t . 6 7 3 1 . 7 6 1

0 . 0 7 8 0 . 0 7 7

0 . 0 0 2 0 . 0 0 4

| . 7 5 3 r . 8 4 2

0 , 4 4 6 0 . s 4 9

0 9 5 6 0 . 9 5 8

2b 2a

0 . 9 3 5

0 . 1 7 8

2 . t 9 3

2 . 3 9 5

0 . 0 1 4

5 . 7 1 5

t . 6 7 3

0 . 0 9 5

0 . 0 0 5

\ . 7 7 3

o . 4 7 8

0. 946

2

S e e T a b l e 2 f o r n o t e s

high Mn content. The 4-phase assemblage is tri-variant lP, T, X(}J2O)1, while the S-phase assemblageis divariant and becomes a surface in P,T,X(H.O)space. Normally such an assemblage would mark anisograd, as a result ofthe reaction ofstaurolite, chlo-rite, and muscovite to biotite, andalusite, and water.However, a simple isograd is not possible in this in-stance since it must separate areas of biotite-stauro-lite-garnet-chlorite from areas of biotite-staurolite-garnet-andalusite. If the locally developed cordieriteoccurrences are considered, the variance is further re-duced and there is still no possible pattern of iso-grads. In our opinion the andalusite and local cor-dierite represent an earlier metamorphism, at whichtime some staurolite and garnet formed. Then duringa second metamorphism the whole staurolite zonewas retrograded as andalusite and biotite reacted tochlorite and more staurolite. With the exception ofandalusite and cordierite, all the minerals equili-

brated with this later event, as shown by consistentKo values. The heat source for this second metamor-phism was probably to the southeast.

P. H. Osberg (personal communication) has ar-gued for a bed-to-bed variation of X(HrO) as an ex-planation for the distribution of rnineral assemblagesdiscussed above. During retrograde metamorphismX(H,O) could drop in the less permeable layers andallow the metastable preservation of andalusite afterpartial reaction. However, a widespread pattern oflarge variations in X(HrO) during a single progradeevent appears improbable. Both the staurolite iso-grad (Osberg, 1968) and the sillimanite isograd (Fig.2) involve dehydration, and yet each is a very regularfeature. The studies of Ferry (1976a, b) demonstrateirregular isograd patterns in impure carbonates, pre-sumably because both HrO and CO, are involved inthe equilibria and the calcareous rocks of the area(Fig. l) are intimately interbedded with pelites. Ac-

NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY, SOUTH-CENTRAL MAINE s7

M 9 / F e

roi l |_ l i = o sa

O 0 4 O : 8 1 2B I O T I T E

Fig. 3. Mg,/Fe Kp plot for muscovite-biotite pairs from thcAugusta quadrangle. Dots = staurolite-zone localitics, circles :lower sillimanite= zone localities, pluses : upper sillimanite-zonclocalities. Line givcs average Kp.

cording to our model the andalusite isograd of thefirst metamorphism approximately coincides with thestaurolite isograd of the second metamorphism. Inthe Farmington area to the northwest (Holdaway,Novak and Guidotti, in preparation), the two iso-grads are clearly distinct.

ln the western part of the staurolite zone chlorite ismore abundant than elsewhere. The stippled part ofFigure 2 is an area in'which chlorite is abundant andhas textures which clearly indicate its retrogressiveorigin. As previously indicated for the whole stauro-lite zone, chlorite cuts across the schistosity formedby muscovite and biotite. In some sections in the ret-rograde area chlorite plates include smaller staurolitegrains. Partial to complete pseudomorphs afterstaurolite and andalusite are common throughout thearea. Toward the northern limits of the retrogradedarea fine-grained white mica pseudomorphs after an-dalusite are most cornmon. while staurolite alterationto chlorite is confined to the rims. Euhedral garnetsin former andalusite are unaffected. Further south,pseudomorphic reactions after staurolite are more

t 2

o 8UJF

o

af

= o 4

Table 4. Chernical analyses ofgarnetr

t 4 7 7 7K I N L O I E K 1 N

A H - 1 9 tCore Rin Core Rin Core

41 5 r02Rin Core R in Core R im

.EW-3 MA-4 lCore Rin Core Rin Core

t 0 7R i n

sio2

Ar203

TiO2

Fe0

MnO

Ca0

Tota l

S i

A I

3 7 . 4 4

2 1 1 0

0 , 1 0

6 . 2 1

1 9 1

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

2 t 0 6 2 r . 4 0 2 1 4 2 2 0 3 6 2 1 0 5

0 1 0 0 0 2 0 0 0 0 . 0 9 0 0 0

3 4 7 9 2 9 1 6 3 t 0 0 2 8 . 1 3 3 t 1 l

2 3 9 2 . 6 3 2 a 4 2 3 0 2 5 4

3 6 3 8 3 5 1 4 1 8 4 9 6 3 7

1 5 3 1 6 9 1 3 8 2 , 5 7 I 8 9

3 7 , 4 3 3 7 , 7 3 3 7 5 1 3 t , 1 4

2 0 7 7 2 0 - 9 5 2 1 . 0 8 2 0 9 6

0 _ 0 3 0 0 0 0 0 3 0 0 1

3 2 , 2 9 3 4 0 2 3 0 8 0 3 0 7 2

2 1 5 2 . 2 4 3 2 0 2 7 8

6 0 1 3 9 3 6 7 9 7 1 1

1 , 6 3 t - 7 1 I 0 4 I t 6

3 7 2 8 3 7 t 4 3 7 . 3 6 3 7 8 A

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

0 0 0 0 0 0 0 0 0 0 0 0

3 2 6 5 3 4 5 6 3 5 8 1 3 6 4 3

2 4 9 2 3 t 2 2 4 2 2 0

5 6 1 4 6 8 3 3 9 3 3 0

1 , 2 9 0 . 7 7 0 7 3 0 . 5 3

3 7 5 0 3 7 2 2 3 6 6 3

2 t . 2 9 2 1 . 3 7 2 0 , 9 0

0 0 1 0 0 0 0 1 5

3 0 1 7 3 0 ? 9 2 6 1 6

2 6 5 2 , 3 3 2 0 8

6 2 8 6 . 6 5 t r . 1 2

2 3 9 2 2 0 2 9 5

3 6 9 3 3 7 , 6 8 3 7 6 3

2 1 0 5 1 9 8 2 1 9 4 l

0 1 3 0 0 0 0 0 1

2 9 2 7 2 9 . 3 7 2 9 6 r

2 7 2 2 8 8 2 2 6

7 9 7 7 . 8 7 8 2 6

2 2 0 1 . 3 9 1 7 4

1 0 0 . 6 3 1 0 0 . 6 3 1 0 0 . 5 4 1 0 1 6 5 9 9 8 3 1 0 0 , 3 8 1 0 0 3 0 1 0 0 . 7 3 1 0 0 4 4 9 9 8 7 1 0 0 . 6 1 1 0 0 7 6 1 0 0 . 4 0 t 0 1 . 3 3 I 0 0 2 8 1 0 0 . 5 4 9 9 9 8 1 0 0 , 2 6 9 9 0 1 9 8 . 9 2

Formula based on I2 oxygens

2 - 9 6 8 2 9 8 7 2 . 9 8 8 2 9 8 5 3 . 0 4 9 3 0 0 7 3 0 I 8 3 . 0 2 2 2 0 0 6 3 0 0 1 2 9 9 1 2 9 8 7 3 , 0 2 7 3 0 2 ' 7 3 , 0 0 0 2 9 8 6 2 9 6 7 2 9 7 3 3 0 6 4 3 , 0 8 2

0 0 3 2 0 . 0 1 3 0 . 0 1 2 0 0 1 5 0 . 0 0 9 0 0 I 3 0 0 1 4 0 0 3 3 0 0 2 7

A 1 1 . 9 5 8 0 9 8 r 2 0 0 8 1 . 9 8 9 1 9 2 9 1 9 9 3 1 9 7 3 t 9 ? 7 r 9 9 0 1 9 9 5 2 . 0 0 4 2 0 o S 1 0 6 4 1 9 7 8 2 0 0 ? 2 0 0 6 1 9 6 2 1 9 6 8 1 8 9 9 1 8 7 1

T i 0 0 0 4 0 , 0 0 5 0 . 0 0 0 0 0 0 0 0 . 0 3 7 0 0 0 0 0 0 0 0 0 . 0 0 0 0 0 0 1 o o 0 o o 0 o o 0 0 0 0 0 . 0 0 0 0 0 0 0 0 . 0 0 0 0 0 0 0 0 , 0 0 6 0 0 0 7 0 , 0 0 0 0 , 0 0 0

: 1 , 9 6 2 1 9 8 6 2 0 0 8 1 . 9 8 9 1 9 6 6 t 9 9 3 1 9 7 3 t - 9 7 7 1 . 9 9 1 1 9 9 5 2 o O 4 2 0 0 5 1 9 6 4 1 9 7 8 2 o O 7 2 0 0 6 1 9 6 8 1 9 7 5 1 8 9 9 1 8 7 1

F e 2 . ) . 1 A 2 3 3 7 r . 9 5 3 2 , 0 5 7 1 8 9 3 2 0 8 9 2 t ' 1 7 2 2 7 8 2 0 6 3 2 . 0 7 5 2 1 9 0 2 3 2 3 2 4 1 7 2 4 3 4 2 o t 7 2 0 6 5 7 7 7 2 1 9 6 9 1 9 9 6 2 0 2 8

M g O 2 6 7 0 2 8 7 0 . 3 1 3 0 3 5 5 0 . 2 7 5 0 3 0 5 0 2 5 7 0 , 2 8 3 0 3 6 1 0 . 3 3 4 0 2 9 6 0 , 2 7 5 O 2 6 9 O 2 6 t O 3 I 4 O 2 7 7 0 2 5 0 0 3 2 4 0 , 3 4 a 0 2 7 5

M n 0 4 2 0 0 2 4 5 0 5 6 0 0 . 4 9 7 0 , 5 7 8 0 4 4 3 0 . 4 0 9 0 2 6 5 0 4 6 0 0 . 4 8 5 0 3 8 0 0 . 3 1 8 0 2 3 6 0 2 2 2 0 4 2 5 O 4 4 g 0 7 6 2 0 5 4 2 0 5 4 0 0 5 7 2

c a 0 . 1 6 2 0 1 3 1 0 1 4 3 0 . 1 1 6 0 2 2 0 0 1 6 1 0 l 4 o 0 1 4 5 0 0 8 8 0 0 9 9 0 1 0 9 0 0 6 4 o 0 6 2 o o 4 4 0 2 0 4 0 . 1 8 7 0 2 5 5 o l 8 9 0 . 1 9 9 o 1 5 t

| 2 9 6 7 3 . 0 0 0 2 9 6 9 3 . 0 0 5 2 9 6 6 2 9 9 8 2 9 8 3 2 , 9 7 1 2 . 9 7 2 2 g g 3 2 g t , 2 g 8 o 2 g A 4 2 . 9 6 1 2 9 6 0 2 , 9 7 8 3 o 3 g 3 . 0 2 4 3 o O 3 3 0 2 6

l v 1 0 1 e % A l n 7 1 , 4 7 7 9 6 5 8 6 8 , 5 6 3 8 6 9 7 7 3 0 7 6 7 6 9 , 4 6 9 3 7 3 6 7 7 9 8 1 0 a 2 2 6 8 . 1 6 9 3 5 8 , 3 6 S . t 6 6 6 6 7 0

M o l e % P y r 9 . 0 9 6 1 0 . 5 1 1 1 9 . 5 1 A 2 8 , 6 9 . 5 1 2 t t t 2 9 . 9 g 2 9 0 8 8 1 0 6 9 3 a 2 1 O 7 t r 6 9 l

M o l e ? S p e s s 1 4 2 8 2 1 8 . 9 1 6 5 1 9 5 1 4 8 1 3 7 8 9 l 5 , S t 6 2 1 1 . 8 1 0 7 7 g 7 5 7 4 3 1 5 1 2 5 0 1 7 9 1 8 0 1 8 9

M o l e % G r o s s s 4 4 3 4 , 8 3 9 7 , 4 , 5 3 4 . 7 4 g 3 o 3 3 3 7 2 2 2 t 1 . 5 7 o 6 . 3 8 . s 6 i i . 9 s , o

F e l F e + M g 0 8 8 8 0 8 9 1 0 8 6 2 0 . 8 6 0 0 8 7 3 0 8 7 3 0 8 9 4 0 - 8 9 0 0 8 5 1 0 8 6 1 0 . 8 8 1 0 8 9 4 0 9 0 0 0 , 9 0 3 0 8 6 5 0 8 8 2 0 8 7 6 0 8 5 9 0 8 5 2 0 8 8 1

A s s e n b l a g e l a I l a I a 2 2 a 2 a 2 3 e 3 ^

*See Tab le 2 fo r no tes

HOLDAWAY: METAMORPHIC PETROLOGY. SOUTH.CENTML MAINE

I O O ! m i n t e r v o l

Fig. 4. Rim to rim compositional profiles for 8-fold sitecontents ofgarnets from two specimens. Lines connect data pointstaken at stated intervals. For specimen 91, depletion of Mn + Catoward rim indicates garnet growth. For specimen 107, increase ofMn + Ca towards rim indicates garnet reaction.

complete, resulting in crystals without fresh surfaces,and consisting of fine-grained muscovite and chlo-rite. Within the same specimens, fine-grained whitemica pseudomorphs after chiastolite or andalusiteare found. Chiastolite crystals are confined to verythin graphite-rich layers, similar to those observed byGuidotti and Cheney (1976, Fig. l). Toward thesouthern limit of the area, just north of Winthrop,the effects ofthe retrogressive event appear to be at amaximum. In addition to staurolite, garnet, whichwas not altered toward the north, has been chlori-tized by the low-temperature episode. Pseudomorphsmaintain the euhedral outline of the host crystals andcommonly contain only core remnants of the originalmineral. These textures are consistent with a staticthermal event accompanied by little deformation,Minor evidence of shearing was noted in a few speci-mens. West of the retrograde area staurolite-bearingrocks are followed by staurotte-sillimanite rockswith no retrogressive efects.

We interpret the textures of the retrograde area asresulting from the third and latest metamorphicevent, with increasing temperature to the west. Theparallelism of the western sillimanite isograd with theretrograde area suggests a hinge effect such that inthe retrograde area the rocks were being adjusted to

lower temperatures while at the western edge of themap area temperature was increasing.

The southern limit of the staurolite zone is definedby the first appearance of sillimanite in traceamounts. The sillimanite isograd, as drawn, separatesall pelitic rocks which have staurolite without sil-limanite from those which have staurolite and sil-limanite. With the exception of a single cordierite-chlorite-bearing rock, chlorite is absent from theserocks. The sillimanite first encountered is of the dis-harmonious type described by Vernon and Flood(1977), Le., growing along quartz and plagioclaseboundaries at high angles to grain contacts. This im-plies that the sillimanite grew after equilibriumamong grain contacts had been established. As sil-limanite increases in amount, much of it grows on orwithin biotite. Along a zone which approximatelycoincides with the southern sillimanite isograd areseveral occurrences ofgarnet with discontinuous zon-ing. Such garnets have been interpreted by Boone(1973) as representing more than one stage of garnetgrowth.

The zone of coexisting sillimanite and staurolite isdesignated the lower sillimanite zone (Guidotti,1974). Staurolite, garnet, and andalusite are commonthroughout the zone in the southern paft of the maparea. One specimen (Fig. 2) contains cordieriterimmed by chlorite, muscovite, and fine staurolitecrystals, which suggests that the cordierite is a retictmineral from an earlier metamorphism. There is nocorrelation between andalusite and sillimanite occur-rence in these rocks. Sillimanite formed as easily inrocks with no andalusite as in rocks containing an-dalusite. Andalusite becomes more inclusion-filledand skeletal toward the southern edge of the lowersillimanite zone. In some specimens containing bothAl silicates, andalusite seems to be selectively re-placed by large muscovite grains, as was observed byChinner (1973). Staurolite also decreases in modalabundance toward the southern limit of the zone.

The upper sillimanite zone in the southern part ofthe area is characterized by the absence ofstaurolite,the near absence of andalusite, and a marked in-crease in modal sillimanite. Toward the southernlimit of staurolite stability, coarse-grained biotite andmuscovite pseudomorphous after staurolite are com-mon. Locally only traces of Zn-ich staurolite remain(Tables l, 5), and these are completely engulfedwithin large (0.3 mm) grains of muscovite. In con-trast to pseudomorphs previously described, thismineral replacement is prograde in nature (Guidotti,1968). The mineral assemblage sillimanite-alman-

t o 7

7 6

7 2

6 8

F r a

-

t[--------------'

"'--------\]

--------rr.

_r/\

,t

NOVAK AND HOLDAIIAY: METAMORPHIC PETROLOGY, SOUTH-CENTRAL MAINE

Table 5. Chemical analyses of stauroliter

59

4 l9 1 Elll- 3

sio2Al 2O3

Tio2

Fe0

Mgo

Mn0

ZnQ

Tot a l

S i

AI

A 1

T i

E

Fe

Mg

Mn

Zn

t

Fe,/ Fe+Mg

As senb lage

2 7 . t 7

5 4 . 7 0

0 . 3 1

1 3 . 9 9

1 . 4 1

0 . 1 5

0 . 7 2

9 8 . 5 4

7 . 6 9 0

0 . 3 1 0

1 7 . 9 3 7

0 . 0 6 6

1 8 . 0 0 5

3 . 3 0 5

0 . 5 8 9

0 . 0 3 3

0 . 1 4 9

4 . 0 7 6

0 . 8 4 9

2 7 . 9 7

5 4 . 4 3

0 . 5 2

1 2 . 1 r

1 . 5 6

0 . 5 8

1 . 8 0

9 8 . 7 9

7 . 8 5 9

0 . 1 4 1

L 7 . 8 8 7

0 . 1 0 7

t 7 . 9 9 4

2 . 8 4 5

0 . 6 5 r

0 . 0 9 0

o . 3 7 |

0 . 8 1 4

I

2 8 . 1 3

5 2 . 8 8

0 . s 9

1 3 . 1 5

1 . 6 0

0 . 6 1

0 . 4 3

9 7 . 3 5

8 . 0 1 1

1 7 . 7 5 7

0 . 1 2 3

I 7 . 8 8 0

3 . 1 2 7

o . 6 7 7

0 . t 4 2

0 . 086

4 . 0 3 r

0 . 8 2 2

28 .52

5 2 . 9 5

0 . 5 0

1 3 . 1 0

l . 6 l

0 . 2 9

1 . 0 2

97 .99

2 8 . 0 8

5 4 . 1 1

0 . 5 0

i 3 . 9 5

I . 4 3

0 . 1 5

0 . 3 7

9 8 . 5 8

2 7 . 6 4

5 4 . 7 7

0 . 5 8

1 4 . 3 5

r . 4 3

0 . 2 4

0 . 3 4

99 . 35

7 . 7 4 8

o . 2 5 2

1 7 . 8 5 3

0 . 1 1 8

7 7 . 9 7 1

3 . 361

0 . 594

0 . 054

0 . 066

4 . 0 7 5

0 . 8 5 0

Fornu la based on 47 oxygens

8 . 0 7 6 7 . 9 0 9 7 . 8 5 5 7 . 9 2 7

0 . 0 9 1 0 . 1 4 5 0 . 0 7 9

2 8 . 0 3

5 4 . 8 6

0 . 5 8

1 2 . 9 3

1 . 6 1

0 . 4 2

0 . 1 8

98 61

2 8 . 0 5

53 .87

0 . 5 5

1 ,2 . 47

7 . 7 6

0 . 4 4

1 . 0 8

9 8 . 2 0

1 7 . 9 7 6 1 7 . 8 5 4

0 . t l g 0 . 1 1 1

t 8 . 0 9 5 1 7 . 9 6 5

3 . 0 2 9 2 . 9 4 3

0 . 6 6 6 0 . 7 3 7

0 . 097 0 . 099

0 . 0 3 1 0 . 2 1 9

3 . 8 2 3 3 . 9 9 8

0 . 8 2 0 0 . 8 0 0

2 2 b

2 8 . 1 9 2 8 . 5 3

5 4 . 0 5 5 5 . 8 6

0 . 5 5 0 . 5 0

1 3 . 3 8 1 0 . 5 8

1 . 3 6 0 . 8 8

0 . 1 5 0 . 4 2

0 . 4 7 I . 1 6

9 8 . l s 9 8 . 0 0

7 . 9 5 6 7 . 9 9 2

0 . 0 4 4 0 . 0 0 8

1 7 . 9 3 8 1 8 . 3 9 6

0 . 1 1 4 0 . 0 9 8

1 8 . 0 5 2 1 8 . 4 9 4

3 . 1 5 8 2 . 4 6 9

0 . 5 6 8 0 . 3 6 0

0 . 0 3 1 0 . 0 9 8

0 . 0 9 3 0 . 2 3 7

3 . 8 5 0 3 . ) , 6 4

0 . 8 4 8 0 . 8 7 3

1 7 . 6 9 2

0 . 1 0 0

L 7 . 6 9 2

3 . r 0 2

0 . 6 7 4

0 . 066

0 . 2 0 7

4 . 0 4 9

o . 8 2 2

7 a

1 7 . 8 7 3

0 . 1 0 2

3 .280

0 . 5 9 7

0 . 0 3 1

0 . 0 7 1

3 . 9 7 9

0 . 8 4 6

" See Tab le 2 fo r no tes .

dine-biotite-muscovite-quartz-plagioclase-ilmeniteis ubiquitous in the upper sillimanite zone (Fig. 2).

We interpret the lower and upper sillimanite zonesas the result of the second metamorphism to affectthe area. Thermal gradients here increased to thesoutheast. The absence of a well-defined zone of ex-tensive retrograde metamorphism parallel to thesouthern sillimanite isograd suggests that (l) stauro-lite-zone conditions were much more widespreadduring this second metamorphism than during thethird metamorphism which affected the rocks alongthe western edge of the area, and (2) the rocks werealready hot at the inception of the second meta-morphism, which implies a relatively short time spanbetween the first and second metamorphic episodes.

To summarize, much of the area was affected byan early metamorphism which produced andalusite,garnet, staurolite, and some cordierite. A secondmetamorphism partly destroyed cordierite and partlyconverted andalusite-biotite to staurolite-chlorite inthe central and northern part of the area, and up-

o 2lrJF

El

F<t,

o . l

o o . 4 0 . 8 t . 2B I O T I T E

Fig. 5. MglFe Kp plot for staurolite-biotite pairs from theAugusta quadrangle. Dots : staurolite zone, circles : lowersillimanite zone. Line gives average Kp.

M g / F e

xol, '_ f i = o.zr

60 NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY. SOT]TH-CENTRAL MAINE

Table 6. Chemical analyses ofcblorite and cordierite+ son, 1957) in which no change in the mineral assem-blage is apparent. Garnet zoning patterns add an-other dimension to the understanding of continuousreactions. Because diffusion in garnets is slow undermedium-grade conditions (Tracy et al.,1976), a con-tinuous record of element distribution with time ispreserved. This distribution, along with the Fe-Mg-Mn variation of the other mineral species, is usedhere to deduce the important equilibria and mineralreactions, both continuous and discontinuous.

Garnet zoning

Figure 6 shows an AFM plot which includes thetotal range of garnet rim composition in terms of Fe/(Fe+Mg). The limiting points also exhibit the maxi-mum range of spessartine and grossular componentsfor rims (Table 4). This maximum range is encom-passed in specimens which contain sillimanite andstaurolite, but the range in Fel(Fe+Mg), spessartine,and grossular components is comparable in thestaurolite-zone rocks. The garnets as a whole arecharacteraed by (l) an absence of compositionaltrends as a simple function of grade, (2) a tendencytoward higher Fel(Fe+Mg) in garnet with lowerspessartine plus grossular component, and (3) deple-tion in spessartine plus grossular from core to rim ac-companied by little change or slight increase n Fe/(Fe+Mg). Because garnet is the main carrier of Mn,the reduction in Mn from core to rim suggests that in

Chlor i te

1 5 7

C o r d i e r i t e

1 7 51.75

si0I

Al 203

Tio2

Fe0

MEO

MnO

Na2O

Tota l

s i

A 1

Ti

Fe

Mg

Mn

Na

x

F e l F e + M g

Assenblage

2 5 . 0 1

2 3 . 1 7

0 . 0 8

2 3 . l O

1 6 . 0 4

0 . 2 4

8 7 . 6 3

4 9 . 0 4

3 2 8 2

0 . 0 3

6 . 3 6

8 . 3 7

0 . 3 2

0 . 8 0

9 7 . 7 4

18 oxygens

5 . 0 3 5

0 . 9 6 5

2 5 . 9 3 2 5 . 4 5

2 2 . 7 4 2 3 . 7 6

0 . 0 7 0 . 0 1

2 1 . 5 5 2 0 . 2 6

1 6 . 7 9 1 8 . 2 1

0 . 2 6 0 . 1 8

8 7 . 3 4 8 7 . 8 6

Fornula based on 28 oxygens

s . 1 9 7

2 . 8 0 3

2 . 8 7 3

0 . 0 0 9

4 0 1 3

4 . 9 7 0

0 . 0 3 8

1 1 9 0 3

0 . 4 4 7

t

2 . 8 S I

0 . 0 0 8

3 . 7 0 9

5 . 1 5 3

0 . 0 4 4

t 1 . 7 6 5

0 . 4 1 9

5 336 3 I92

2 . 6 6 4 2 . 8 0 8

2 . 9 0 3 3 0 0 6

0 . 0 0 0 0 . 0 0 r

3 . 4 5 3 0 . 5 4 4

5 . 5 3 3 1 . 2 7 9

0 . 0 2 8 0 . 0 2 6

- 0 . 1 5 7

1 1 . 9 1 7 5 . 0 1 3

0 . 3 8 4 0 . 3 0 0

2b 2b

" S e e T a b l e 2 f o r n o t e s .

graded rocks in the southern part to sillimanitegrade. A third metamorphism produced a NNE-trending zone of more extensive retrogressive effectsand upgraded the western edge of the area. In thediscussions which follow, the presence of andalusiteor cordierite interpreted to be relict will be ignored.

Garnet zoning and mineral reactions

J. B. Thompson (1957) first introduced the conceptof defining isograds on the basis of discontinuous re-actions, in order 16 aflimize original bulk-composi-tional effects. The applicability of this concept hasbeen repeatedly demonstrated in the field (e.g.,Green, 1963; Albee, 1968; Carmichael, 1970). Win-kler (1979, p. 66) futher emphasizes this point anduses the term reaction-isograd. Index-mineral-de-fi.ned isograds in the classical sense are still useful forprovidi-ng a general framework for establishing meta-morphic conditions, but a knowledge of specific reac-tions is necessary in order to deduce the completemetamorphic history of an area.

In addition, detailed chemical analyses of mineralsprovide evidence for continuous reactions (Thomp-

A

/ t \,,\

Fig. 6. AFM plot giving extremes of garnet rim compositionstable with sillimanite, staurolite, and biotite. As shown" spessartineand grossular contents s16 highg5l in low-Fe garnets and lower in

high-Fe garnets. The ranges shown for minerals are also appli-cable to the staurolite zone.

NOVAK AND HOLDAIIAY: METAMORPHIC PETROLOGY, SOUTH-CENTRAL MAINE 6 l

most instances garnet was growing during equilibra-tion of the rim (Tracy et al., 1976). Figure 7 gives aschematic pressure-temperature diagram for equi-libria involving almandine, biotite, staurolite, mus-covite, and sillimanite. [f we assume that the rocks allcrystallized at about the same pressure, then the pres-ence of garnet-muscovite-biotite-staurolite in therocks requires that the sum ofspessartine and grossu-lar decrease as Fe/(Fe+Mg) increases, in order forconditions to stay approximately parallel to thenearly flat curve for

Staurolite + Biotite p Almandine

* Muscovite + HrO2 (l)

Once sillimanite is produced, two additional equi-libria must prevail:

Staurolite * Muscovite ? Biotite

+ Sillimanite * H,O (2)

Staurolite A Sillimanite * Alrnandine * H,O (3)

As staurolite breaks down over a temperature range,equation 2 requires that all minerals become moreFe-rich because Mg fractionates more strongly intobiotite than into the other minerals. At the sametime, more garnet is being produced and its spessar-tine and grossular contents must decrease to main-tain equilibrium for equation l. In summary, therelationship between Fel(Fe+Mg) and the sum ofspessartine and grossular in garnet must hold for allbut the two highest-grade specimens which have nostaurolite. However, once sillimanite begins to form,each garnet must increase in Fel(Fe+Mg) and de-crease in spessartine and grossular as temperaturerises. For these rocks, the composition of garnet is abulk-compositional effect related to Fe,/(Fe+Mg) ofthe dominant biotite and to temperature, once sil-limanite has formed. Finally, because all but two ofthe rocks contain staurolite, garnet grows during in-creasing temperature. The continuous reaction

Almandine + Muscovite ? Biotite * Sillimanite(4)

must remain at equilibrium but cannot begin to de-stroy garnet until all the staurolite has reacted. Thefacts that Mn depletion requires garnet growth andthat garnet growth is a prograde phenomenon inthese rocks indicate that the garnet rims are in mostcases prograde effects.

T E M P E R A T U R EFig. 7. Schematic pressure-temperature diagram for thc system

FeO-AI2O3-SiO2-K2O-H2O involving excess-SiO2 and excess-H2O reactions about the invariant point involving garnet-biotite-staurolite-muscovite-sillimanite. Relative slopes are based ondata of Thompson (1976).

D is c o nt inu ou s r e ac tions

The isograds of Figure 2 may each be related to aparticular discontinuous reaction. The absence ofchloritoid in the region supports a reaction such as

Almandine * Chlorite + Muscovite €Staurolite + Biotite + HrO (5)

for the first appearance of staurolite. Based on otherareas (Osberg, 1968; Guidotti, 1970b) the first ap-pearance of staurolite is at a more or less constantgrade, but atnandine and chlorite can persist to-gether to higher temperatures due to Mn and Ca inthe garnet.

The first sillimanite must form by the reaction

Staurolite + Chlorite + Muscovite aBiotite + Sillimanite + HrO (6)

as implied by AFM topologies on either side of thesillimanite isograd (Guidotti, 1974). ln chlorite-freerocks and in rocks where chlorite has reacted away,sillimanite forms by the continuous reaction 2. Reac-tion 6 is sharply defined in the field (Fig. 2) and re-sults in the immediate disappearance of chlorite. Thereaction

Andalusite = Sillimanite (7)

is not considered responsible for the first appearance

ld(r3a(t,IJOE(L

l ol(9+

6

2 Quartz is omitted from all reactions for simplicity

NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY. SOUTH-CENTRAL MAINE

Table 7. Garnet-biotite KD temperatur€ estimates

Specinen I , l ineral Xp" /X11* KD

Staurolite disappears by the reaction

Staurolite + Muscovite eBiotite + Sillimanite + Garnet + H,O

the combination of reactions 2 and 3. However, thisreaction is not strictly discontinuous, due to Mn andCa in the garnet. These minerals coexist in mostrocks once sillimanite has formed. As temperaturerises, reaction 2 dominates and progressively in-creases Fel(Fe+Mg) in all minerals (Fig. 7). As reac-tion progresses the growing garnet becomes depletedin spessartine and grossular until staurolite even-tually disappears. Once staurolite has reacted away,the continuous reaction 4 may occur. A few thin sec-tions of highest-grade rocks show reaction of garnetby pseudomorphs of biotite or biotite and sillimaniteafter garnet.

Conditions of metamorphism

Estimates of temperature and pressure for the Au-gusta area during the last metamorphism were madewith the intent of avoiding use of any of the AlrSiO,triple points, due to the lack of complete agreementon this system.

Garnet-biotite Ko

The ubiquitous presence of biotite and garnet inthe pelitic rocks allows the use of the recently cali-brated leothermometer based on Fe-Mg partitioningin natural assemblages (Thompson, 1976) as well asin synthetic phases (Ferry and Spear, 1978). For thetemperature range involved here (Table 7), the twocalibrations agree quite well.

Garnet rim compositions were measured about 5microns from the actual rim to avoid effects of graincurvature from diferential hardness during polish-ing. These determinations involve the assumptionthat the core garnet equilibrated with biotite of thesame composition as that now in the rock (Tracy etal., 1976). Biotite is the dominant Fe-Mg phase,being 6 to 25 times as abundant as garn€t and 3.5 to35 times as abundant as staurolite (Table l). Ilmenitecomponents dissolving in the biotite with increasingtemperature might be expected to drive the biotite to-ward slightly more Fe-rich compositions with in-creasing grade. More important is the Fe enrichmentwhich should result as staurolite begins to breakdown in the lower sillimanite zone. This implies thatcore temperatures from the sillimanite zone are prob-ably maximum values. Large changes in biotite com-position are not possible within a single specimen,because the relatively small amounts of staurolite

T O C * T o C '(8)

1 4 7 G a r n e t c o r eG a r n c t r i nB i o t i t e

7 7 G a r n e t c o r eG a r n e t r i nB i o t i t e

A I I - 1 G a r n e t c o r eG a r n e t r i nB i o t i t e

91 GarnetGarnetB l o t l t e

41 GarnetGarnetB i o t i t e

7 . 9 3 3 6 6 1 38 1 4 3 6 . 7 8 6r . 2 0 0

6 . 2 4 4 6 . 4 2 46 1 4 0 6 . 3 2 20 . 9 7 1

6 . 8 8 4 6 9 7 76 8 4 9 6 . 9 4 20 . 9 8 7

8 . 4 7 1 7 . 0 3 28 . 0 5 0 6 . 6 8 21 . 2 0 5

5 . 7 1 5 6 . 2 4 16 . 2 1 3 6 . 7 8 50 . 9 1 6

7 . 5 9 9 6 0 9 08 . 4 4 7 6 9 5 5I 2 1 5

8 . 9 8 5 6 . 2 8 29 . 3 2 6 6 . 5 2 rI . 4 - 3 0

6 . 4 2 4 5 . 8 3 87 4 5 5 6 . 7 7 51 . 1 0 0

7 . 0 8 8 7 . 1 6 r6 0 7 7 6 . 1 4 00 . 9 9 0

5 . 7 3 6 5 . 9 7 67 . 3 7 5 7 . 6 8 30 . 9 6 0

s 1 5s235 1 6

c o r eI 1 n

c o r er r n

528 532532 537

5 0 9 5 0 8s l 0 5 0 9

5 0 7 s 0 55 1 9 5 2 0

5 3 5 5 4 15 1 5 5 1 6

5 4 1 5 4 95 1 0 5 0 9

s 3 3 5 3 95 2 5 5 2 7

551 5625 1 6 5 1 6

s 0 3 5 0 05 3 9 5 4 6

s 4 5 5 5 5492 481

5 Garnet coreG a r n e t r i mB l o t i t e

102 Garnet coreG a r n e t r i mB i o t i t e

E W - 3 G a r n e t c o r eG a r n e t r i nB i o t i t e

M A - 4 G a r n e t c o r eG a r n e t r i nB i o t i t e

I 0 7 G a r n e t c o r eG a r n e t r i nB i o t i t e

. . G a . . B it ^ = I F e r l F e

y b a y b l''l4g "Mg

* T h o m p s o n ( 1 9 7 6 )

" F e r r y a n d S p e a r ( i 9 7 8 )

of sillimanite for the following reasons: (l) Tex-turally, sillimanite appears to be a product of a latermetamorphic episode, as shown by its growth alongundisturbed quaftz and plagioclase boundaries. (2)Andalusite and sillimanite are rarely seen in contact,and when they are, sillimanite needles cut across pre-existing andalusite crystals, leaving them undis-turbed with no sign of a reaction relationship (Ver-non and Flood, 1977, Fig. la). (3) The zone of dis-continuously zoned garnets near the southernsillimanilg isograd is consistent with more than oneperiod of metamorphism. (4) The decrease in stauro-lite and disappearance of chlorite are consistent withreaction 6. (5) Sillimanite appears in andalusite-freerocks at the same grade as it appears in andalusite-bearing rocks. Andalusite is gradually converted tosillimanite throughout the lower sillimanite zone andthe beginning of the upper sillimanite zone.

NOYAK AND HOLDAI{AY: METAMORPHIC PETROLOGY. SOUT:H.CENTRAL MAINE

must disappear before having a major effect on bio-tite composition. Valid geothermometry also requirescomplete diffusion in biotite if it changes composi-tion and absence of ditrusion in garnet as it changescomposition. The absence of zoning in biotite andthe generally reasonable temperature values (Table7) suggest that this is a good approximation.

A I mole percent change in biotite Fel(Fe+Mg)changes temperature by about l0o while a I molepercent change in garnet Fel(Fe+Mg) changes tem-perature by about 20o. Thus analytical error in-troduces a potential t30o uncertainty while calibra-tion error adds another !20o.

Temperature plots may be made for each point ofan analytical traverse across a garnet. Such plotsgiven in Figure 8 show a prograde and a retrogradeprofile. Analysis of the data of Table 7 in light of theabove discussion, and the fact that Mn*Ca depletiontoward rims results from prograde growth, indicatesthat all but four of the garnets showed approximatelylevel or increasing temperature during growth. Threespecimens, 41, EW-3, and 107 have retrograde rimsas indicated by Mn*Ca increase, and core to rimtemperature decrease of 25" to 74o. Specimen 5shows a retrograde Ko and a prograde Mn*Cachange. This could have resulted from retrogradeFe-Mg exchange (Tracy et al., 1976). The averagetemperature of the staurolite zone was 519+l2oc' in-cluding core and rim temperatures, 52ltl2oC in-cluding only rim temperatures. The average temper-ature for the lower sillimanite zone was 528+20"Cincluding cores and rims, 537+13'C including rimsonly. The four retrograde rims were excluded. Forcomparison, Osberg estimated a temperature of500'C for the West Sydney locality in the staurolitezone (Osberg, 1971) and 550oC for the sillimaniteisograd (Osberg, 1974).

Thefluid phase

Among other important variables which affect theconditions of metarnorphism is the composition ofthe fluid phase. This is an especially important vari-able in graphite-bearing pelitic rocks in which thewater expelled during progressive dehydration reac-tions becomes diluted with other species, mainlyCHo, CO2, Hr, and CO (French,1966), thus loweringP(H,O) and affecting the physical stability ranges ofmineral assemblages (see for example Holdaway,le78).

ln our treatment we assume P(fluid) : P(total).

l O O , u m i n l e r v o l

l o7

5 4 0

4 6 0 1 2 5 r m i n l e r v o l

Fig. 8, Rirn to rim temperature profiles for the garnets shownin Fig. 4. Specimen 9l shows prograde rims while specimen 107shows retrograde rims. Note that the core temperatures forspecimen 107 in the sillimanite zone could have been afected by

I changing biotite composition (see text).

Important reactions to consider are (Eugster andSkippen,1967):

(Jo

lrJE.f

F

EUJo-

UJF

c+o,eco,co++o,=co,H2+ +O2c HrO

CHo + 2O|?CO2+2HzO

With independent estimates of temperature and pres-sure, the amounts of the diluting gas species can becalculated. Data for equilibrium constant calcu-lations were taken from Ohmoto and Kerrick (1977,Table 1). Fugacity coefficients used are from the fol-lowing sources: Hr, CO, CHo, Ryzhenko and Volkov(1971);COr,Zhaikov et al. (1977); HrO, Burnham e/al. (1969). QFM buffer data are from Huebner( le7 l ) .

As a first approximation, calculations were madeusing an average temperature of 550"C for thestaurolite-out isograd (Table 7) and a total pressureestimate of 4.0 kbar, approximating the maximumpressure estimate of Ferry Q967a) for the Water-ville-Vassalboro area. Partial pressures of the various

(e)(10)

( l l )

(r2)

3 One standard deviation of data, Ferry and Spear calibration.

NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY, SOUTH-CENTML MAINE

H:0

coz

cH+

ll2

c0

O2

XHzo

Tablc 8. Partial pressures of gas species with graphite, 550'C,4000 bars

P c o 2 = P c t 1 4 fq, = 0FI1

or X(HrO) was less than the values assumed here,then these estimates will be the maximum values oftemperature and pressure (Table ll) possible for theconditions of metamorphism. Reaction 3 has beenstudied in the pure Fe system with P(H,O) :

P(total), thus both mineral solid solution and dilu-tion of the fluid phase will have an effect on actualstability relations in the natural system. Mineralcompositions used by Hoschek (1969) approximatethose determined by microprobe analyses for thisstudy. Reaction 2 has therefore been adjusted for di-lution of the fluid phase only. Reaction boundariesmay be adjusted with a temperature or a pressureshift.

Temperature adjustments for reaction 3 weremade using the relationship:

rQ) - r(1) : aT(z) : Rr(2)lnK/aS (14)

for assumed ideal solid solutions in mineral speciesand:

r(3) - r(2): ar(b) : RZ(3)hr@,o)/I-^S (15)

for the fluid phase (Holdaway, 1978). f(l) is the ini-tial temperature for each reaction while T(2) andI(3) are the final temperatures after mineral andfluid composition adjustments respectively. f(HrO)is the mole fraction of water raised to the number ofmoles of water in the balanced reaction.

In a similar fashion, reaction 2 was adjusted inpressure using the relation:

P(2) - P(l) : AP: -RZlnx(H,O)/LV (16)

Entropy values were calculated following the proce-dure of Holdaway (1978). Molar volume data aremainly from Robie et al. (1978) and Burnham et al.(1969). Entropy and volume data are shown in Table9 along with AS for reaction 3 and LV for reaction 2.The calculated changes in pressure and temperatureare given in Table l0 for each reaction, with an ex-ample for specimen 102 shown diagrammatically inFigure 9. The intersections of reactions 2 and 3 (reac-tion 8) for all_specimens containing graphite areshown in Table I I for three different values ofx(H,o).

These calculations show that in the lower sillima-nite zoire temperature estimates vary from about590"C at X(H,O) : 0.75 to a maximum of 655"C atX(HrO) : 0.866. At these same conditions pressureestimates range from 3.2 to 5.1 kbar. These values arein good agreement with the P-Z estimates of Guid-otti (1970a) for the lower sillimanite zone.

0 4 5 9

4 l 9

4 0 5

2 0 4

3 s 9

5 4 6 5 b a r s

2 6 0 b a r s

2 6 0 b a r s

I 5 b a r s

= 1 q - 2 2 l 5 " 1 t

0 8 6 6

3327 t lars

5 4 7 b a r s

i l 5 b a r s

9 b a r s

2 b a r s

f ^ ^ = l 0 - 2 I 8 3 a r n

0 8 3 0

foz

* S e e t e x t f o r r e f e r e n c e s

gas species were then calculated by solving the abovegas equations simultaneously on two different as-sumptions. One set of determinations assumes equalmolar production of CHo and CO, by the reaction:

2 C + 2 H , O c C O , + C H 4 (13)

which yields a maximum estimate for X(H'O)(French, 1966). A second calculation was done atQFM (quartz-fayalite-magnetite) buffer conditions,similar to those found by Osberg (1971) in the Wa-terville-Vassalboro area for coexisting ilmenite andmagnetite. Calculated partial pressures of the gasspecies are shown in Table 8 along with X(H,O). Thesulfur fluid species have been omitted due to spo-radic occurrence of sulfides. However, sulfur speciescould have locally produced lower values of X(H'O).Ideal mixing in the fluid phase has been assumed,following Ohmoto and Kerrick (1977).

Reactions inv o lving stauro lit e

From the maximum value of X(HrO) given inTable 8 and two lower values of 0.80 and 0.75, an ap-proximation of the P-7 conditions was made forlower sillimanite zone rocks based on two differentexperimentally determined reaction equilibria. Thetwo sets of mineral equilibrium data which are appli-cable to rocks ofthe Augusta area are:

Staurolite e Sillimanite * Almandine * HrO (3)

as determined by Richardson (1968), and the

Staurolite * Muscovite = Biotite + Sillimanite+ H,O (2)

reaction boundary of Hoschek (1969). Both curvescan be adjusted for mineral composition andX(HrO). The intersection of the two reaction curvesdefines a unique point in P-T space for a given set ofmineral compositions and value of X(H'O), assumingP(total) : P(fluid). If a fluid phase was non-existent

NOVAK AND HOLDAIAAY: METAMORPHIC PETROLOGY. SOUTH-CENTML MAINE

Interpretation of results

Previous estimates of the pressure-temperaturerelationships in the Augusta-Vassalboro area by Os-berg (1974) and Ferry (1976b) have relied on the useof the aluminum silicate triple-point diagram as de-termined by Holdaway (1971). Although Holdaway'sslope for the andalusite-sillimanite transition hasbeen found to be most consistent with thermochem-ical data (Anderson et al., 1977). use of the Al,SiOjpolymorphs to accurately estimate pressures of meta-morphism may be inaccurate (A. B. Thompson,1976). The andalusite-sillimanite transformation isknown for its sluggish nature, which allows andalu-site to persist unreacted well outside of its field of sta-bility. Also, as postulated here, sillimanite is prob-ably the product of a later metamorphism at higherpressures than the andalusite-producing event. Dur-ing this later metamorphism the pressure may havebeen higher due to increased overburden at a latertime.

For these reasons, one might consider the 550oC,2.8 kbar estimates of Osberg (1974) for his sillimaniteisograd to be lower limits of tempbrature and pres-sure for the Augusta area. A maximum pressure esti-mate in the 550'-600oC range, based on absence ofkyanite, would be between 4.5 and 6.0 kbar (Newton,1966; Richardson et al.,1969; Holdaway, l97l).

As shown, the procedure used to estimate P-Z

Table 9. Entropy and volume data+

trli ner a I S 9 6 O o K ( J / o K ) V o l u n e ( J / b a r )

Table 10. Ternperature and pressure shifts for reactions 3 and 2

S p e c i n e n R e a c t i o nA T b a t 4 0 0 0 b a r s

xHto= o 366

4 I

5

i 0 2

Eiv-3

a l I

spec rnens

- 4 1 2 5

- 5 - 2 6

- 4 8 - 2 5

" H z o

0 8 6 6

0 . 8 0

0 7 5

- 4 9

- 4 9

- 5 1

- 4 4

AI ' at 875oK

+ 2 8 9

+ 4 4 8

+ 5 7 8

5

3

3

3

- 3 8

- 3 8

-44

- 3 8

6 Stauro l i te + 11 Quar tz = 4 A lnand ine + 23 S i l l i nan i te+ 6 H 2 o ( 3 )

o S t a u r o l i t e + 4 l l u s c o v i t e + 7 t u r e r f - = 1 l S i l l i m a n i t e+ 4 B io t i te + 6HZO (2)

A S - = 2 5 3 . 1 6 J / o KJ

LYt = Z1 '6gn t 'O^ t

A T a : e f f e c t o f m i n e r a l c o m p o s i t i o n

A T b : e f f e c t o f d i l u t i o n o f u a t e r

E s t i n a t e d e r r o r i n d T a n , l A P i s 2 5 %

conditions in this study is subject to variations inX(HrO). Ferry (1976b) notes that X(H'O) can be ex-pected to vary on a bed-to-bed scale, due to fluidcompositional differences resulting from proximity tosources of CO, and HrO such as interbedded carbon-ate-bearing layers and granitic intrusions respec-tively. Even among the samples studied in detail,some were characterized by absence of clearly detect-able graphite, which further suggests that fluid com-positional differences can exist on any scale.

Abundant pegmatites throughout the area suggestthat the magmas generating the various plutons werewater-saturated. The spatial relationship betweenzoisite in the carbonate rocks and the outcrop patternof plutons (Fig. l) also suggests the presence of a wa-ter-rich fluid, as zoisite-bearing rocks are indicativeof a water-rich environment (Kerrick, 1974). Thus if

QFM conditions were approximated, the distributionof the fluid species could not have varied signifi-cantly from those shown in Table 8.

Considering the errors involved, the agreement be-tween the garnet-biotite geothermometry and thetemperatures indicated from staurolite equilibriumstudies is reasonable. Values of X(H,O) less than 0.70

Table ll. Interscctions ofreactions 2 and 3r

S p e c i n e n0 8 6 6

Quar tz 108. 35

F e S t a u r o l i t e 1 4 9 2 . 2 2 4

Almand ine 805. 964

S i l l i n a n i t e 2 7 4 . 9 I

Water (4000 bars ) I l3 .Zeb

l' luscovite

F e B i o t i t e

2 2688

22 .8 t46c

L t . s 2 a 2 d

4 . 9 9 0 0

z .s l tab ls tsoy l

14 -0770

t s . 4 4 6 7 e

^H2o

a

b

c

d

e

D a t a f r o n R o b i e e t a l . ( 1 9 7 8 J e x c e p t w h e r e i n d i c a t e d

S t r u c t u r a l a n a l o g m e t h o d , H o l d a w a y ( 1 9 7 8 )

B u r n h a n e t a 1 ( I 9 6 9 ) , 4 0 0 0 b a r s

G r i f f i n a n d R i b b e [ 1 9 7 3 ) t i n e s 4 7 / 4 6 ( s e e t l o l d a w a y , 1 9 7 8 )

S k i n n e r ( 1 9 5 6 )

H e w i t t a n d W o n e s ( 1 9 7 5 )

41.

5

1 0 2

EW- 3

3 . 8 k b 6 0 7 " c 3 . 4 k b s 9 2 ' c a 1 k b s 8 o " c

5 . s k b 6 5 6 ' C 5 . r k b 6 4 0 " C 4 8 k b 6 2 8 ' C

4 . 0 k b 6 1 4 " C 3 . 6 k b s 9 9 ' C 3 3 k b 5 8 7 ' C

3 . 7 k b 6 0 5 ' C 3 . 3 k b 5 9 0 " C 3 0 k b 5 7 8 " C

* D a t a f r o n T a b l e l 0 a n d e x p e r i n e n t a l s t a b i l i t y c u r v e s

f o r X g r 6 = l , s h o m i n F i g u r e q

66 NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY. SOUTH-CENTRAL MAINE

5 0 0 6 0 0T E M P E R A T U R E O C

Fig. 9. Intersections of reactions 2 and 3 adjusted for mineral and fluid phase compositions (Table ll) and garnet-biotitegeothermometry results for the lower sillimanite zone (Table 7). Dots - conditions for X(H2O) : 0.75, circles = conditions for X(H2O) :0.80, vertical lines = garnet-biotite tcmperatures, box = overall estimate, including error, for lower sillimanite-zone conditions. Dashedlines show corrected curves for reactions 2 and3 applicable to specimen 102 at X(H2O) = 0.75. Reactions at X(H2O) : I are based onRichardson (1978) and Hoschek (1979) modified by Winkler. Aluminum silicate diagram is that of Holdaway (1971).

d):<

lrlE 4

(t,6ulE.L

place all the intersections in the andalusite stabilityfield for any Al silicate determination. Temperaturesabove 600"C are inconsistent with those obtained byCheney and Guidotti (1979) for the sillimanite-Kfeldspar zone in the Puzzle Mountain area, NWMaine. Such conditions are not realized within theAugusta Quadrangle.

Combining all methods of P-T determination, abest average estimat€ for the lower sillimanite zone isX ( H , O ) : 0 . 7 5 1 0 . 1 0 , P : 3 . 8 t 1 k b a r , T :570140'C. Thus, the Holdaway (1971) triple pointappears most consistent with the pressure-temper-ature conditions estimated for the Augusta area. Thesmall temperature discrepancies between the twomethods and the need for X(H'O) somewhat lessthan calculated (Table 8) suggest that the availablestaurolite experimental results may be 25 to 50o toohich.

Conclusions

Regional relationship s

The metamorphism observed in the Augusta areahas several characteristics in common with the vari-ous episodes described by Guidotti (1970b), suggest-ing the possibility that at least a few of the eventswere coeval or sinilar in style in the two areas. Someof the similarities include the observations that:

(l) pseudomorph-producing events are similar;(2) cordierite is rimmed by coarse chlorite, sug-

gesting it is a metastable relict from a previous event,possibly coeval with andalusite;

(3) isograds, both prograde and retrograde, arcspatially related to the distribution of igneous bodies;

(4) sillimanite appears to be the result of a latermetamorphic episode, also related to the outcrop pat-tern of plutons.

Guidotti's M(2) event is an andalusite-biotite-staurolite-producing event, and, like the one in theAugusta area, its origin is uncertain. Possibly andalu-site and locally cordierite were the result of a wide-spread (lower pressure) event, the effects of whichwere expressed on a regional scale. M(3) (Guidotti) isrepresented in the Augusta area by two later sillima-nite-producing events closely associated with the dis-tribution of plutons, but which further from the con-tacts become staurolite-biotite-chlorite events. whichbegin to destroy andalusite and cordierite.

Dallmeyer and Vanbreeman (1978) suggested thatthe sillimanite-grade rocks are indicative of a higherpressure regime. They cite a regional southwest re-duction in ooAr/'nAr mineral ages as being duelargely to differences in metamorphic grade, theyounger ages corresponding to higher grades andlonger cooling periods. (Zartman et al., 1970, hadpreviously explained the younger ages as the result of

/ =I

l *l ol(,

l

lr,I /t / "t/.

t /t /

i l i l|

NOVAK AND HOLDAWAY: METAMORPHIC PETROLOGY, SOUTH.CENTRAL MAINE 67

a Permian thermal event.) The concordance of min-eral ages between plutons and high-grade countryrocks suggests to Dallmeyer and Vanbreeman thatthese plutons, which include the Hallowell plutonnear Augusta, were emplaced at deeper crustal levelsthan those which intrude lower-grade rocks farther tothe north. Though this may be the case, there areother factors to consider. One of these includes thehypothesis postulated here that the sillimanite-grademetamorphism represenls later metamorphic epi-sodes, associated with higher pressures. Mineral agescould have been partly to totally reset due to theselater events, whose heat sources were located to thewest and southeast.

Other evidence which suggests that the sillimaniteoverprint was possibly at higher pressures relative tothe other events in the area involves the character ofthe sillimanite-producing events. The sillimanite-pro-ducing event whose source was located to the west ofthe town of Winthrop produced a widespread area ofretrogression which included the production ofstaurolite and andalusite pseudomorphs along withthe formation of large amounts of chlorite. The sil-limanite isograd to the southeast shows only minorretrograde products. For this reason, the two sillima-nite isograds are believed to have been produced sep-arately, and seem related only because they intersectsouthwest of Winthrop. In fact, the geometry of thesillimanite isograd alone suggests that more than onemetamorphic episode was involved in its formation.The amount of retrogression associated with theevent to the west suggests that the country rocks af-fected were cooler than those afected by the south-ern event, imptying that the southern event moreclosely followed M(2) than the western event.

Cause of metamorphism

Several of the igneous bodies in western andsouthern Maine, including the Mooselookmegunticpluton and the Sebago pluton, have been proposed tobe subhorizontal sheets as opposed to bulbous or cy-lindrical masses (Kane and Bromery, 1968; Moenchand Zartman, 1976). Kane et al. (1972) and Nielsenet al. (1976) have correlated these sheet-like bodieswith regions of high-grade rocks, and suggested thattheir shapes result from rheological properties of therocks at high temperatures. Gravity contours west ofWinthrop (Kane and Bromery, 1966) suggest thepossibility that the plutons found here may also fol-low the same pattern of intrusive style. The numer-ous small igneous bodies west of the study area (Fig.l) may actually be cupolas of a larger granitic sheet

at depth, whose eastern linit lies just west of the Au-gusta quadrangle. These plutons acted as sources ofboth heat and water, contributing to the sillimanite-grade metamorphism as well as imposing lower tem-perature conditions on the former andalusite-biotiteterrane. This model agrees well with other studies inMaine which relate various metamorphic episodes togroups of igneous bodies as well as to individual plu-tons (Guidotti, 1970b, 1974; Boone, 1973; Moenchand Zartman, 1976; Ferry, 1976a).

The ultimate heat source responsible for the distri-bution of regional isograds in Maine is unknown. Invarious pulses throughout the time span covering theAcadian orogeny, however, granitic magmas weregenerated which provided local sources of heat andwater, thus forming the polymetamorphic terranenow observed in western and south-central Maine.

AcknowledgmentsThis paper is part of a master's thesis submitted by the senior

author to Southern Methodist University. We thank Walter An-derson of the Maine Geological Survey for allowing access to aer-ial photographs and for use of work space during visits to the Bu-reau of Geology in Augusta. Thanks also go to Philip Osberg foran introduction to the stratigraphy and structure of the Augusta-Waterville-Vassalboro area.

Robert J. Tracy, Philip Osberg, and Charles Guidotti reviewedthe manuscript and offered many helpful suggestions. Financialsupport for electron microprobe analyses was provided by a grantfrom the Institute for the Study of Earth and Man at SouthernMethodist University. Financial assistance for field work andother research-related activities came from NSF erant EAR76-12463 to M. J. Holdaway.

ReferencesAlbce, A. L. (1968) Metamorphic zones in northern Vermont. In

E-an Zen et al., Eds., Studies in Appalachian Geology: North-ern and Maritime, p. 239-341. Wiley-Interscience, New York.

Anderson, P.A.M., Newton, R. C., and Kleppa, O. J. (1977) Theenthalpy change of the andalusite-sillimanite reaction and theAl2SiO5 diagram. American Journal of Science, 277, 585-593.

Bence, A. E. and Albee, A. L. (196E) Empirical correction factorsfor th€ electron microanalysis of silicates and oxides. Journal ofGeology, '16,382403.

Boone, G. M. (1973) Metamorphic stratigraphy, petrology andstructural geology of the Little Bigelow Mountain map area,western Maine. Maine Geological Survey Bulletin 24.

Burnham, C. W., Holloway, J. R., and Davis, N. F. (1969) Ther-modynamic properties of water to l000oC and 10,000 bars.Geological Society ofAmerica Special Paper 132.

Carmichael, D. M. (1970) Intersecting isograds in the WhestoneLake area, Ontario. Journal ofPetrology, ll, 147-181.

Cheney, J. T. and Guidotti, C. V. (1979) Muscovite-plagioclaseequilibria in sillimanite + quartz bearing metapelites, PuzzlcMountain area, northwest Maine. American Journal of Science,279,4tr434.

Chinner, G. A. (1973) The selective replacement of the aluminum

NOVAK AND HOLDAIIAY: METAMORPHIC PET:ROLOGY, SOUTH-CENTRAL MAINE

silicates by white mica. Contributions to Mineralogy and Petrol-ogy,41, 83-87.

Dallmeyer, R. D. and Vanbreeman, O. (1978) qr'lr/1eAt and Rb--Sr ages in west-central Maine: their bearing on the chronologyof tectono-thermal events. (abstr.) Geological Society of Amer-ica Abstracts with Programs, 10, 38.

Doyle, R. G. (Ed.) 1967) Preliminary geologic map of Maine,scale l:500,000. Maine Geological Survey.

Eugster, H. P. and Skippen, G. B. (1967) Igneous and metamor-phic reactions involving gas equilibria. In P. H. Abelson. Ed.,Researches in Geochemistry, II, p. 492-520. John Wiley andSonst New York.

Ferry, J. M. (1976a) Metamorphism of calcareous sediments in theWaterville-Vassalboro area, south-central Maine. AmericanJournal of Science. 276.841-882.

Ferry, J. M. (1976b) P, T, flH2O) during metamorphism of cal-careous sediments in the Waterville-Vassalboro area, south-central Maine. Contributions to Mineralogy and Petrology, 57,l 19-143.

Ferry, J. M. and Spear, F. S. (1978) Experimental calibration ofthe partitioning of Fe and Mg between biotite and garnet. Con-tributions to Mineralogy and Petrology, 66, ll3-117.

Fisher, L. W. (1941) Structure and metamorphism of Lewiston,Maine, region. Geological Society of America Bulletin, 52, 107-160.

French, B. M. (1966) Some geological implications of equilibriumbetween graphite and a C-H-O gas phase at high temperaturesand pressures. Reviews of Geophysics, 4,223-253.

Green, J. C. (1963) HighJevel metamorphism of pelitic rocks innorthern New Hampshire. American Mineralogist, 48, 991-1023.

Griffin, D. T. and Ribbe, P. H. (1973) The crystal chemistry ofstaurolite. American Journal of Science, 273A,479495.

Guidotti, C. V. (1963) Metamorphism of the pelitic schists in theBryant Pond quadrangle, Maine. American Mineralogist, 48,7't2-79t.

Guidotti, C. V. (1968) Prograde muscovite pseudomorphs afterstaurolite in the Rangeley-Oquossoc areas, Maine. AmericanMineralogist, 53, 1368-1376.

Guidotti, C. V. (1970a) The mineralogy and petrology of the tran-sition from the lower to uppcr sillimanite zone in the Oquossocarea, Maine. Journal of Petrology, 11,271-336.

Guidotti, C. V. (1970b) M€tamorphic petrology, mineralogy, andpolymetamorphism in a portion of N.W. Maine. In G. M.Boone, Ed., 1970 New England Intercollegiate Geological Con-ference, 62nd Annual Meeting. Field trip B-1, l-29.

Guidotti, C. V. (1974) Transition from staurolite to sillimanitezone, Rangeley quadrangle, Mainc. Geological Society ofAmerica Bulletin. 85. 47 5490.

Guidotti, C. V. and Cheney, J. T. (1976) Margarite pseudomorphsafter chiastolite in the Rangeley area, Maine. American Miner-alogist,6l, 431-414.

Hewitt, D. A. and Wones, D. R. (1975) Physical properties ofsome synthetic Fe-Mg-Al trioctahedral biotites. American Min-eralogist, 60,85+862.

Holdaway, M.J. (1971) Stability of andalusire and the aluminumsilicate phase diagram. American Journal of Science, 27I,97-l 3 l .

Holdaway, M. J. (1978) Significance of chloritoid and staurolite-bearing rocks in the Picuris Range, New Mexico. GeologicalSociety of Amcrica Bulletin, 89, I4M-1414.

Hoschek, G. (1969) The stability of staurolite and chloritoid and

their significance in metamorphism of pelitic rocks. Contribu-tions to Mineralogy and Petrology, 22,208-232.

Huebner, J. S. (1971) Buffering techniques for hydrostatic systemsat elevated pressures. In G. C. Ulmer Ed., Research Techniquesfor High Pressure and High Temperature, p. 123-177. Springer-Verlag, New York.

Kane, M. F. and Bromery, R. W. (1966) Simply Bouguer gravitymap of Maine. U. S. Geological Survey Geophysical Investiga-tions Map GP-580.

Kane, M. F. and Bromery, R. W. (1968) Gravity anomalics inMaine. In E-an Zen et al., Eds., Studies of Appalachian Geol-ogy: Northern and Maritime, p. 415423. Wiley-Interscience,New York.

Kane, M. F., Bromery, R. W., Simmons, G., Diment, W. H., Fitz-patrick, M. M., and Joyner, W. B. (19'12) Bouguer gravity andgeneralized map of New England and adjoining areas. U. S.Geological Survey Geophysical Investigations Map GP-839.

Kerrick, D. M. (1974) Review of metamorphic mixed-volatile(HrO-CO2) equilibria. American Mineralogist, 59,'129-762.

Moench, R. H. and Zartman, R. E. (19?6) Chronology and stylesof multiple deformation, plutonism, and polymetamorphism inthe Merrimack synclinorium of western Maine. Geological So-ciety of America Memoir, 146,203-238.

Newton, R. C. (1966) Kyanite-andalusite equilibrium from 700oto 800oC. Science, 151. l7Fl72.

Nielson, D. L., Clark, R. G., Lyons, J. B., Englund, E. J., andBorns, D. J. (1976) Gravity models and mode of emplacementof the New Hampshire Plutonic Series. Geological Society ofAmerica Memoir, 146, 301-318.

Novak, J. M. (1978) Metamorphic petrology, mineral equilibria,and polymetamorphism in the Augusta quadrangle, south-cen-tral Maine. M. S. Thesis, Southern Methodist University, Dal-las, Texas.

Ohmoto, H. and Kerrick, D. M. (1977) Devolitization equilibria ingraphitic systems. American Journal of Science, 277, l0l3-l0/y'..

Osbcrg, P. H. (1968) Stratigraphy, structural geology, and meta-morphism of the Waterville-Vassalboro area, Maine. MaineGeological Survey Bulletin 20.

Osberg, P. H. (1971) An equilibrium model for Buchan-type rneta-morphic rocks, south-central Maine. American Mineralogist, 56,570-586.

Osberg, P. H. (1974) Buchan-type metamorphism of the Water-ville pelite, south-central Maine. In P. H. Osberg, Ed., 1974New England Intercollegiate Geological Conference, 66th An-nual M€eting. Field trip B-6,210-222.

Osberg, P. H., Moench, R. H., and Warner, J. (1968) Stratigraphyof the Merrimack synclinorium in west-c€ntral Maine. In E-anZen et al., Eds., Studies of Applachain Geology: Northern andMaritime, p. 241-253. Wiley-Interscience, New York.

Pankiwskyj, K. A., Ludman, A., Griffen, J. R., and Berry, W. B.N. (1976) Stratigraphic relationships on the southeast limb ofthe Merrimack synclinorium h central and west-central Maine.Geological Society of America Memoir, 146,263-280.

Richardson, S. W., (1968) Staurolite stability in a part of the sys-tem Fe-Al-Si-O-H. Journal of Petrology, 9, 467488.

Richardson, S. W. Gilbert, M. C., and Bell, P. M. (1969) Experi-mental determination of kyanite-andalusite and andalusite-sil-limanite equilibria: the aluminum silicate triple point. AmericanJournal of Science. 26'l, 259-272.

Robie, R. A., Hemingway, B. S., and Fisher, J. R. (1978) Ther-modynamic properties of minerals and related substances at298.15K and I bar (l0s pascals) pressure and at higher temper-atures. U.S. Geological Survey Bulletin 1452.

NOYAK AND HOLDAWAY: METAMORPHIC PETROLOGY. SOUTH.CENTML MAINE 69

Ryzhenko, B. N. and Volkov, V. P. (1971) Fugacity coefficients ofsome gases in a broad range of temperatur€s and pressures.Geochemistry International, 8, 468-48L

Skinner, B. J. (1956) Physical properties of end-members of thegamet group.American Mineralogist, 4 1, 428436.

Thompson, A. B. (1976) Mineral reactions in pelitic rocks: II. Cal-culation of some P-I-X (Fe-Mg) phase relations. AmericanJournal of Science, 276,425454.

Thompson, J. B., Jr. (1957) The graphical analysis ofmineral as-semblages in pelitic schists. American Mineralogist, 42, E42-858.

Tracy, R. J. (1978) High grade metamorphic reactions and partialmelting in pelitic schist, west-central Massachusetts. AmericanJournal of Scicnce, 279, lSG-l'|8.

Tracy, R. J., Robinson, P., and Thompson, A. B. (1976) Garnetcomposition and zoning in the determination of temp€ratureand pressure of metamorphism, central Massachusetts. Ameri-can Mineralogist, 61, 7 62-71 5.

Vernon, R. H. and Flood, R. H. (1977) Interpretation of metamor-phic assemblages containing fibrolitic sillimanite. Contributionsto Mineralogy and Petrology, 59,227-235.

Winkler, H. G. F. (1979) Petrogenesis of Metamorphic Rocks, 5thedition. Springer-Verlag, Heidelberg.

Zaflma\ R. E., Hurley, P. M., Kruegcr, H. W., and Giletti, B. J.(1970) A Permian disturbance of K-Ar radiometric ages in NewEngland-its occurrence and cause. Geological Society ofAmerica Bulletin, 81, 3359-337 4.

Zharikov, V. A., Shmulovich, K. I., and Bulatov, V. K. (1977) Ex-perimental studies in the system CaO-MgO-AI2O3-SiO2-CO2-H2O and conditions of high temperature metamorphism. Tec-tonophysics, 43, 145-162.

Manuscript received, August 2, 1979;

acceptedfor publication, fune 17, 1980.