Structural and metamorphic relations in the southwest Seward Peninsula, Alaska: Crustal extension and the unroofing of blueschists

Kimberly A. Hannula* Elizabeth L. Miller Department of Geology, Stanford University, Stanford, California 94305 Trevor A. Dumitru } Jeffrey Lee* Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, California 91125 Charles M. Rubin Department of Geology, Central Washington University, Ellensburg, Washington 98926

ABSTRACT role in the exhumation of the blueschist-fa- these interpretations of similar structures is cies rocks but may have followed an earlier an indication of the difficulty of distinguish- An oblique crustal section on the south- period of partial erosional unroofing of ing structures formed during subduction western Seward Peninsula, Alaska, exposes these rocks. from those formed during exhumation of polydeformed, polymetamorphosed rocks blueschists. Careful determination of the that record extensional thinning of the crust INTRODUCTION age of structures and associated metamor- in the Late Cretaceous. The rocks experi- phic grade, along with consideration of tim- enced an early (M1a) high-pressure/low- The preservation, uplift, and exhumation ing of regional events (i.e., brittle supra- temperature metamorphism (pumpellyite- of high-pressure/low-temperature metamor- crustal faulting, basin formation, igneous actinolite facies in the upper part of the phic rocks have long been considered prob- activity, etc., synchronous with blueschist ex- section; blueschist facies at deeper struc- lematic (e.g., Ernst, 1988). Numerous mod- humation), is necessary to understand the tural levels) followed by a greenschist-facies els have been proposed to explain how significance of different generations of struc- overprint (M1b) and accompanied by D1 de- blueschist-facies rocks might be returned to tures. In this paper, we discuss new struc- formation. A second deformational event, the surface: return flow in a melange (Cloos, tural, metamorphic, and geochronologic

D2, is responsible for the prominent, gently 1982), a combination of thrusting and ero- observations relating to the exhumation dipping foliation, northwest-southeast– sion (Suppe, 1979; Rubie, 1984), upper- history of high-pressure/low-temperature trending stretching lineations, and abun- crustal extension synchronous with subduc- metamorphic rocks found on the Seward dant recumbent isoclinal folds throughout tion and underplating (Platt, 1986), and Peninsula, Alaska. the section. Metamorphism during D2 (M2) postorogenic extensional collapse (Dewey, Multiply deformed blueschist- to green- varied from extremely low-grade at the shal- 1988; Platt and Vissers, 1989). It is difficult schist-facies rocks of Paleozoic and presum- lowest structural levels to upper amphibo- to test these models in the field because ably Precambrian age (the Nome Group of lite- to granulite-facies within the Kigluaik rocks that have undergone blueschist-facies Till et al. [1986]) underlie a large portion of gneiss dome. Apatite fission-track ages from metamorphism have usually experienced a the rolling hills of the Seward Peninsula the shallowest structural levels indicate complex structural history. The multiple (Fig. 1). The Nome Group flanks several that cooling below ϳ120 ؇C took place at ca. generations of ductile structures commonly Cretaceous high-temperature gneiss domes 70–100 Ma, during the same time span as found in blueschists have been explained in (the Kigluaik Group) exposed in the Kig- high-grade metamorphism and pluton in- several ways: progressive, evolving shear luaik, Bendeleben, and Darby Mountains trusion at depth. Significant vertical atten- during a single event (e.g., Ridley, 1984; and is structurally overlain by the lower uation of the crustal section apparently took Harris, 1985; Helper, 1986; Patrick, 1988); grade slate of the York region (Sainsbury, place during D2 deformation, resulting in two phases of deformation, with D1 occur- 1972) and thin-bedded Paleozoic (?) silty close spacing of both M1a and M2 isograds ring during subduction and D2 resulting limestone (Fig. 1). Weakly metamorphosed and overall thinning of the crust. This ex- from thrusting during uplift (e.g., Bell and to little metamorphosed Ordovician to Mis- tensional deformation played an important Brothers, 1985; Jayko and Blake, 1989; Phil- sissippian (A. Harris, 1991, written com- ippot, 1990; Joyce, 1992); and two phases of mun.) carbonate rocks lie in low-angle fault

deformation, D1 during subduction and D2 contact above the more-deformed silty lime- *Present address: Hannula: Geology Depart- resulting from ductile extension during up- stone and slate of the York region (Sains- ment, Middlebury College, Middlebury, Vermont 05753; Lee: Department of Geology, Central lift (e.g., Lister et al., 1984; Anderson and bury, 1972) (Fig. 1). Washington University, Ellensburg, Washington Jamtveit, 1990; Ave´ Lallement and Guth, Previous studies of the structural and met- 98926. 1990; Little et al., 1992). The variation in amorphic history of the Nome Group have

Data Repository item 9524 contains additional material related to this article.

GSA Bulletin; May 1995; v. 107; no. 5; p. 536–553; 11 figures.

536 STRUCTURAL AND METAMORPHIC RELATIONS, SEWARD PENINSULA

This paper reports the results of a de- tailed investigation undertaken to deter- mine the polyphase structural and metamor- phic history of the rocks of the southwestern Seward Peninsula (Fig. 1). Our work fo- cused on the question of which fabric ele- ments developed during the blueschist-fa- cies metamorphism and which developed during later events related to the uplift and exhumation of these rocks.

STRUCTURAL AND METAMORPHIC HISTORY (OVERVIEW)

Previous detailed work on structures within the Nome Group focused on areas south of the eastern Kigluaik Mountains (Thurston, 1985; Patrick, 1988). They iden-

tified two main foliations: an earlier S1 fo- liation subparallel to lithologic layering, and

an S2 foliation axial planar to isoclinal folds of S1. They interpreted both of these folia- tions as having formed during a single de- formational event synchronous with high- pressure metamorphism. On the basis of our observations north and west of the Kigluaik Mountains, we disagree with their interpre- tation and believe that the two foliations they observed are the result of two tempo- rally distinct deformational events. North of the Kigluaik Mountains, meta- sedimentary rocks dip gently, generally northward or northwestward, and display two main tectonic foliations (Fig. 2). The

oldest recognized foliation (S1) predates a strongly developed second foliation, which

Figure 1. Geologic map of the Seward Peninsula (simplified from Sainsbury [1972], is the dominant one in the map area. S1 ap- Sainsbury et al. [1972], and Till et al. [1986]). Rocks assigned to the slate of the York region pears to postdate an early high-pressure met- in the Nome 1:250 000 quadrangle by Sainsbury et al. (1972) are shown as part of the Nome amorphism (M1a) that varied from pumpel- Group based on a redefinition of the Nome Group by Till et al. (1986) to include all lyite-actinolite facies near Teller, where blueschist-greenschist-facies rocks of the Seward Peninsula. North-trending fold axes are original bedding and sedimentologic fea- inferred from foliations and outcrops in Sainsbury et al. (1972) and Till et al. (1986). tures are preserved, to blueschist-facies far- ther south, across a map distance of about 25 km and across a structural section of concluded that the regionally pervasive pen- ulite-facies metamorphism of the Kigluaik about 5–10 km (Figs. 2, 3, 4, and 11, below). etrative subhorizontal fabrics and north- Group as the result of post-tectonic relax- S1 formed synchronous with a greenschist- south stretching lineations in these rocks ation of formerly depressed isotherms fol- facies overprint (M1b) on the early high- formed as a consequence of top-to-the- lowing the cessation of subduction. How- pressure assemblages. north overthrusting and continental (A- ever, Miller et al. (1992), Calvert (1992), and The most readily measured foliation in type) subduction (Patrick, 1988). Armstrong Amato et al. (1994) have shown that signif- the field is S2, which developed during the et al. (1986) attempted to date the blue- icant deformation in the Kigluaik Group second recognized deformational event schist-facies metamorphism using the K-Ar took place synchronous with the younger (D2) and is responsible for the present dis- and Rb-Sr methods; the scatter of the re- high-temperature metamorphism. These tribution of map units (Fig. 2). S2 is axial ported ages (ranging from 96 to 194 Ma) was later studies raised the question of how planar to tight to isoclinal recumbent folds interpreted as the result of a partial over- much of the deformation in the overlying of both bedding and S1 everywhere in the print by Cretaceous high-temperature met- Nome Group took place in a subduction map area. S2 foliation surfaces (solution amorphism, and 160 Ma was interpreted to zone synchronous with high-pressure/low- cleavage surfaces in the northern part of the be a minimum age for the blueschist-facies temperature metamorphism and how much area and a syn-metamorphic foliation to the metamorphism. Patrick and Lieberman deformation took place at a later time, per- south) display a weak northwest-southeast

(1988) interpreted the amphibolite- to gran- haps related to the exhumation of the rocks. stretching lineation (Ls) first seen in more

Geological Society of America Bulletin, May 1995 537 3 elgclSceyo mrc ultn a 1995 May Bulletin, America of Society Geological 538

Figure 2. Geologic map of the southwestern Seward Peninsula showing foliations, lin- eations, and stereonet data. Gray regions are outcrops of metabasites. Cross sections A–A؅ .and A؅–A؆ are shown in Figure 5. Note duplication where cut Geological Society of America Bulletin, May 1995 539 HANNULA ET AL.

Figure 3. Map of the southwestern Seward Peninsula show- ing metamorphic isograds, glaucophane, crossite, and biotite lo- calities; sample localities for probed amphiboles; and locations of samples shown in photomicrographs.

phyllitic units, which becomes better defined defined stretching lineations southward in- from sub-greenschist facies near Teller to southward in all rocks except the metaba- dicate a north to south strain gradient asso- greenschist facies west and south of the sites (Fig. 2). The better development of a ciated with the development of S2. The sec- Kigluaik Mountains. Along the flanks of more penetrative foliation and the concom- ond deformational event was accompanied the Kigluaik Mountains, M2 metamorphic itant development of increasingly better- by a metamorphic event, M2, which ranges grade increases dramatically to amphibolite

540 Geological Society of America Bulletin, May 1995 STRUCTURAL AND METAMORPHIC RELATIONS, SEWARD PENINSULA

.sillimanite ؍ biotite; Sil ؍ garnet; Bt ؍ pumpellyite; Grt ؍ Figure 4. Cross sections (locations shown in Fig. 2). Pmp

facies over a set of closely spaced Barrovian east–trending axes of F2 folds (Fig. 2b). A bedding and are crenulated by S1 and S2. isograds (Miller et al., 1992) (Fig. 5). D2 fab- second, shallowly northwest-dipping cleav- They are interpreted as detrital mica grains. ric elements S2 and Ls become even better age (S2, Fig. 2c) is developed axial planar to In thin section (Fig. 5A), S2 can be observed defined along the flanks of the Kigluaik tight to isoclinal F2 folds. The S1 orienta- to be a closely spaced crenulation cleavage Mountains, obliterating all trace of S1 tions plotted in Figure 2b are taken mainly defined by an insoluble opaque residue. It

(Fig. 2) (Calvert, 1992; Miller et al., 1992). from hinge zones of F2 folds, where S1 and appears to have formed largely by the dis-

S2 can be clearly distinguished. This causes solution of calcite and quartz, resulting in

STRUCTURAL DATA F2 folds to appear more open on the stereo- the variable rotation of M1b fabrics and min-

net than they are. Intersections between the erals into parallelism with S2 along the edges

The structural data upon which the above S2 cleavage and S1 and S0 (L1ϫ2 and L0ϫ2, of microlithons. Little or no new growth of general observations were based are sum- Fig. 2b) trend mostly northeast-southwest. minerals is apparent in this younger cleav- marized in stereonets in Figure 2. Here we Poles to S1 define a fold axis (␲ axis, Fig. 2b) age (Fig. 5A). The formation of S2 by the describe the data from the shallowest to that is parallel to L1ϫ2 and L0ϫ2 intersection processes of dissolution and reprecipitation deepest structural levels. The map units are lineations. Depending on the attitude of of calcite and quartz, as well as the lack of 1 described in detail in Appendix A. bedding prior to F2 folding, it is either new micaceous mineral growth in S2, implies

Two distinct deformational events are re- tightly folded or thinned and boudinaged that S2 formed at temperatures Ͻ400 ЊC corded in coastal exposures of platy-weath- during D2 deformation. Prominent steeply (e.g., Gray and Durney, 1979). ering, thin-bedded silty limestone near dipping calcite-filled tension gashes formed The fine-grained gray phyllite unit to the

Teller (map unit Pzsl) (Fig. 2). A strong pen- at a high angle to S2 and indicate a north- south (unit Pzqms) possesses a well-devel- etrative cleavage (S1, Fig. 2b) cuts bedding south extension direction at low-tempera- oped penetrative fabric with a weak north- at generally low angles but is also observed ture conditions (Fig. 2c). These tension northwest–south-southeast stretching linea- axial planar to tight to isoclinal F1 folds of gashes are not folded by F2 and thus are tion (defined by fibrous quartz pressure bedding. Measurements of F1 fold axes inferred to be related to the late stages of D2 shadows on pyrite) (Ls, Fig. 2f) developed

(L0ϫ1, Fig. 2a) are gently plunging and trend or to postdate D2. on its surfaces. On the basis of the gently north-northeast/south-southwest to south- In these northernmost exposures, meta- dipping attitude of this fabric (S2, Fig. 2d) southeast. F1 folds are refolded along with morphic conditions during formation of S1 and rare observations of it axial planar to S1 around south-southwest–north-north- were those characteristic of subgreenschist isoclinal recumbent folds of an earlier foli-

to lower greenschist facies, as evidenced by ation (S1), we interpret it to be the same as the growth of very fine-grained white mica S described above. Lithologic variations in 1GSA Data Repository item 9524, Appendixes 2 A and B, is available on request from Documents and chlorite in S1 (Fig. 5A). Large mica this unit are minor, and original bedding Secretary, GSA, P.O. Box 9140, Boulder, CO flakes are occasionally visible in hand spec- (S0) was only rarely documented with cer- 80301. imen and in thin section. They lie parallel to tainty; where measurable, S0 is gently dip-

Geological Society of America Bulletin, May 1995 541 HANNULA ET AL.

ping and subparallel to measurements of S2 trending to the south (Fig. 6A). L1ϫ2 trends end of the Kigluaik gneiss dome (Fig. 2), the

(Fig. 2d). S2 formed axial planar to recum- northeast to northwest in northern Pzmvu earlier S1 foliation is much less evident in bent isoclinal folds of an earlier cleavage outcrops, but trends more consistently the field. Rare measurements of F2 fold axes and/or bedding; the axes of these F2 folds northwest-southeast to the south (Fig. 6B). trend northwest (FA2, Fig. 2j), subparallel

(FA2, Fig. 2d) are variable in orientation, Southward, a strong north-northwest– to a well-developed stretching lineation (Ls, with the largest number trending southeast. south-southeast stretching lineation (Ls, Fig. 2k). The dominant S2 foliation dips gen-

Intersection lineations between S2 and lith- Fig. 2i) develops on S2 foliation surfaces. A tly to the west (Fig. 2j). Within this map unit, ologic layering (L0ϫ2) plunge gently and plausible interpretation of the changing ori- about 5 km north of the Kigluaik Moun- vary in trend from northeast-southwest to entations of L0ϫ2 and L1ϫ2 as Ls becomes tains, the first evidence for new white mica, increasingly well developed is that with in- chlorite, and albite growth synchronous with northwest. S2 is gently folded by centimeter- amplitude upright, open folds. A rarely ob- creasing strain, linear elements at an angle development of S2 appears (Fig. 3). This to the stretching direction (X axis of the new mineral growth obliterates evidence for served steeply dipping spaced cleavage (S3, Fig. 2e) is locally developed axial planar to strain ellipsoid) rotate into parallelism with an older S1 in chlorite- and white-mica-rich this direction (e.g., Sanderson, 1973). samples (Fig. 5D), although remnants of S these folds. The intersections between S2 1 Steeply dipping, northeast- to northwest- can still be distinguished in more quartz-rich and S3 (L2ϫ3, Fig. 2e) trend northwest- southeast to southwest. Calcite-filled ten- striking calcite-filled tension gashes (Fig. 2i) samples. Because new mineral growth dur- sion gashes are again nearly vertical, strike are not folded by F2 folds and are oriented ing the youngest event makes it difficult to east-northeast–west-southwest (approximate- approximately perpendicular to the stretch- read the older history of these rocks, it is ly perpendicular to the stretching lineation), ing lineation. possible that several pre-S2 deformational Oriented thin sections from map unit events may be represented but are impossi- and are not folded by F2 folds. Since F3 folds and the tension gashes were both rarely ob- Pzmvu support field observations of two de- ble to identify in these rocks. served and never seen in the same outcrop, formational events and provide more infor- To the south of the Kigluaik Mountains, mation on the nature of metamorphism and chloritic schists are found interlayered with the relative timing of F3 folding and the for- mation of the tension gashes is unknown. mineral growth accompanying them. Rocks marbles and quartzose mica schists (Figs. 2, of appropriate composition in Pzmvu con- 3, and 4). The various lithologic units are The underlying map unit (Pzmvu) con- tain M or M micas that increase in grain exposed in approximately north-south– tains metabasite bodies (Pzmb) intercalated 1a 1b size southward, as well as porphyroblasts of trending belts, which commonly pinch out in the metasedimentary section (Fig. 2). albite and titanite, the growth of which along strike (Fig. 2). The disappearance of Both S and S are present in the metasedi- 1 2 clearly predated S (Figs. 5B and 5C). As in units along strike may be the result of map- mentary rock types in this map unit, but the 2 the silty limestone unit, S developed mostly scale isoclinal folding and/or boudinage, as more resistant and/or thick-bedded rock 2 by the processes of dissolution/reprecipita- similar structures have been observed at the types, such as metabasite bodies, thick ho- tion; not much new growth of minerals other outcrop scale. Lithologic layering in these rizons of tuffaceous sediments, and gray- than quartz or calcite is apparent in S units, as in rocks north of the Kigluaik wacke, are not penetratively deformed and 2 (Figs. 5B and 5C). These relations are best Mountains, has generally been transposed preserve original igneous or sedimentary observed in thin sections cut perpendicular into parallelism with the dominant S2 folia- structures. to the axes of crenulations. Thin sections cut tion. As is the case immediately north and Within the phyllitic metasediments in the perpendicular to foliation but parallel to west of the Kigluaik Mountains, the domi- Pzmvu unit, S2 is the dominant cleavage stretching lineations do not exhibit obvious nant foliation (S2) is generally penetrative in (Fig. 2h). It is gently dipping and axial pla- asymmetry or well-developed kinematic some units (chloritic schists and marbles) nar to abundant thin section to outcrop- indicators. but forms a zonal crenulation cleavage that scale tight-isoclinal recumbent folds of S1 Within the underlying chloritic schist map is axial planar to tight to isoclinal recumbent and bedding (subparallel to one another). unit (Pzcs), which wraps around the western folds of an earlier foliation (S1) in the more F1 folds are either rare or are more difficult to discern because of the greater strain as- sociated with S2. Their presence, however, is ã suggested by the spread in orientations of S0 (Fig. 2g), the poles to which define a south- Figure 5. Locations of samples are shown in Figure 3. (A) Interlayered phyllite and west-trending fold axis. Poles to S define a 1 marble from map unit Pzsl. S1 is defined by fine-grained white mica and chlorite and is at northwest-trending fold axis (Fig. 2g). Both a low angle to the S0 lithologic layering. S2 is a spaced cleavage axial planar to an isoclinal compositional layers and S must have orig- 1 fold of S1 and S0. Long direction of photo is 7.2 mm across. (B) Schist from map unit Pzmvu. inated at high angles to S2 in order to have S1 is defined by white micas and titanite porphyroblasts and is at about right angles to the been ubiquitously tightly folded and crenu- S2 spaced crenulation cleavage. Long direction of photo is 3.6 mm. (C) Schist from map unit lated by this second event. Intersection lin- Pzmvu. Note pre-S2 growth of albite (Ab). Long direction of photo is 1.44 mm. (D) Schist eations defined by the intersection of bed- from map unit Pzcs. Only one foliation, synchronous with the growth of chlorite, white ding and S1 with S2 (L0ϫ2 and L1ϫ2, Fig. 2h) mica, and albite, can be observed. This foliation is parallel to S2 observed in other rocks. are gently plunging to subhorizontal and are Long direction of photo is 3.6 mm. (E) Mica schist from south of Kigluaik Mountains. Both quite variable in trend. L0ϫ2 varies from remnants of S1 micas and new, S2 micas can be seen. Long direction of photo is 7.2 mm. southwest to northwest trending in the (F) Mica schist from southwest of Kigluaik Mountains. Note growth of dark biotite (Bt) in northern part of the Pzmvu unit but be- albite pressure shadows and parallel to the dominant foliation. Long direction of photo is comes consistently northwest-southeast 7.2 mm.

542 Geological Society of America Bulletin, May 1995 A B

C D

E F

Geological Society of America Bulletin, May 1995 543 HANNULA ET AL.

S2 is cut by a northwest-striking, moderately the evidence for these three metamorphic northeast-dipping extensional crenulation events.

cleavage (S3) (Fig. 2n) with a top-to-the- The northernmost exposures of metaba- north shear sense. sites commonly preserve their original igne- Foliation within Nome Group units im- ous mineralogy: coarse pyroxene, tabular mediately north, west, and south of the Kig- plagioclase, ilmenite, and apatite. In some luaik gneiss dome parallels foliation within cases, the pyroxene is partially replaced by the underlying higher grade rocks (Fig. 2). brown hornblende, which presumably is the

D2 fabric elements become extremely well result of late-stage magmatic processes. In developed in the region of the isograds, most rocks, original igneous minerals are which ring the Kigluaik Mountains, persist- partially replaced by a finer-grained meta- ing across the Barrovian isograds into the morphic assemblage: pyroxene and horn-

sillimanite zone (Miller et al., 1992). S2 dips blende are replaced by actinolite and chlo- gently to moderately west on the west flank rite, ilmenite is replaced by titanite, and of the range (Fig. 2o) and south on the south plagioclase is replaced by a very fine-grained

flank (Fig. 2p). Stretching lineations (Ls, mixture of epidote, albite, white mica, and Figs. 2o and 2q) defined by mineral streak- rare pumpellyite. The assemblage pumpel- ing, pressure shadows, and aligned silliman- lyite ϩ actinolite ϩ chlorite ϩ albite ϩ epi- ite trend northwest-southeast in biotite- dote can be found together in mutual con- to sillimanite-grade rocks. Quartz-chlorite– tact in some samples. Epidote appears to filled tension gashes strike east-west and dip replace pumpellyite in several cases. Al- moderately northward (Fig. 2p). though reaction relationships among the fine-grained metamorphic minerals are dif- METAMORPHISM ficult to identify, this assemblage suggests

that this area may have undergone M1a met- The southwestern Seward Peninsula has amorphism in the pumpellyite-actinolite fa- experienced three identifiable phases of met- cies (Ͻ375 ЊC, 3–6 kbar) (pumpellyite ϩ ac- amorphic recrystallization, each of which tinolite ϩ chlorite ϩ albite ϩ quartz, Liou et

varies in grade across the area. The first met- al., 1985; Evans, 1990) followed by M1b met- amorphic event (M1a) is the high-pressure/ amorphism in the greenschist facies (350– low-temperature metamorphism identified 500 ЊC, Ͻϳ6 kbar) (actinolite ϩ albite ϩ Figure 6. Stereonets showing the rota- by previous workers (Sainsbury et al., 1970; chlorite ϩ epidote ϩ quartz, Liou et al., Forbes et al., 1984; Thurston, 1985; Patrick 1985). The occurrences of pumpellyite are tion of L0؋2 and L1؋2 from north to south within map unit Pzmvu. The orientation of and Evans, 1989). M1a assemblages are most restricted to a few samples in the northern all stretching lineations from the Teller distinctive in metabasites and are only rarely part of the Teller A3 quadrangle. We have area is shown for comparison. preserved in our study area. M1a grade drawn an approximately located M1a ranges from pumpellyite-actinolite facies in pumpellyite-out isograd (Fig. 3) to the south the northern part of the study area to epi- of these sample localities. quartz-mica-rich units. In rocks in which S2 dote blueschist facies in the southern Teller To the south of this rough isograd, most forms a zonal cleavage, variable new growth A-3 quadrangle (Fig. 1; Fig. 3). The second metabasites contain actinolite, epidote, of chlorite ϩ white mica (generally Ͼ50% of metamorphic event (M1b) is a greenschist- chlorite, albite, quartz, and titanite Ϯ white the rock volume) occurs parallel to the S2 or epidote-amphibolite-facies overprint on mica Ϯ stilpnomelane in addition to relict foliation (Fig. 5E). In rare localities, new M1a assemblages and can be distinguished igneous pyroxene, plagioclase, ilmenite, and growth of biotite can also be observed within from M1a in metabasites that preserve M1a brown hornblende. We have found one oc- S2 (Figs. 3 and 5F). Stretching lineations (Ls, assemblages. The S1 foliation in schists and currence of crossite in interlayered mafic to Fig. 2l), defined by elongate pressure shad- rare, weakly foliated metabasites appears to intermediate and calcareous schist within ows about porphyroblasts and by mineral have formed during M1b metamorphism. A unit Pzmvu (LD-13, Fig. 3) near the streaking, are somewhat more scattered qualitative increase in M1b grade from north pumpellyite-out isograd. The crossite occurs than stretching lineations north of the Kig- to south in the Teller A-3 quadrangle is sug- in more mafic layers along with epidote, luaik Mountains but trend generally north- gested by the increasing size of chlorite and chlorite, albite, quartz, titanite, stilpno- northwest–south-southeast. The scatter in white micas down-section. M1a and M1b may melane, and calcite. It forms ragged relict orientation of Ls may be the result of broad both be parts of a single metamorphic cycle grains, many of which lie with their long di- post-D2 folding, as indicated by the concen- (as described by Forbes et al. [1984] and mension at a high angle to the dominant tration of Ls orientations within a small cir- Thurston [1985]). The third metamorphic foliation (Figs. 7A and 7B). Chlorite, whose cle about the fold axis defined by poles to S2 event (M2) varies from subgreenschist facies preferred orientation helps define the foli- (Figs. 1, 2l, and 2m). F2 fold axes (FA2, in the northern Teller A-3 quadrangle to ation in the rock, wraps around and forms Fig. 2m) are generally subparallel to the greenschist facies west and south of the Kig- pressure shadows around the crossite. It is stretching lineation. Rare quartz-filled ten- luaik Mountains to amphibolite to granulite not clear whether the foliation defined by sion gashes occur perpendicular to the av- facies in the core of the Kigluaik gneiss the chlorite is S1 or S2. Adjacent metasedi- erage stretching direction (Fig. 2n). Locally, dome. In the paragraphs below, we describe mentary samples contain no micaceous min-

544 Geological Society of America Bulletin, May 1995 STRUCTURAL AND METAMORPHIC RELATIONS, SEWARD PENINSULA

eral growth in S2, suggesting that the chlo- bro is preserved. These two samples indi- Garnets south and west of the Kigluaik rite foliation in LD-13 is S1. In either case, cate that rocks in this area experienced M1a Mountains are primarily almandine-grossu- it is clear that the dominant foliation in this epidote-blueschist-facies metamorphism fol- lar solutions and are similar in composition sample postdates M1a crossite growth and lowed by M1b greenschist- to epidote-am- to those reported by Thurston (1985). Gar- occurred during M1b retrogression of cross- phibolite-facies metamorphism. nets west of the Kigluaik Mountains gener- ite to chlorite. Epidotes exhibit quartz ϩ Other metabasite samples from the area ally show evidence of retrogression to chlo- chlorite pressure shadows, and albite por- consist of actinolite ϩ epidote ϩ chlorite ϩ rite and/or biotite (Figs. 7E and 7F). In phyroblasts contain inclusions of chlorite albite Ϯ stilpnomelane Ϯ garnet Ϯ quartz. some samples, resorbed garnet porphyro- and exhibit chlorite pressure shadows. Ti- This mineral assemblage might possibly blasts are found as inclusions in albite or tanite occurs rimming opaques (ilmenite?) have crystallized under the same conditions actinolite, suggesting that garnet growth and also occurs in thin layers parallel to the as the glaucophane-garnet–bearing rocks may have preceded growth of the green- foliation. The composition of crossite is (the differences in mineral assemblage being schist-facies assemblage. In some mafic shown in Figure 8. The occurrence of cross- the result of difference in bulk-rock compo- schist samples, chlorite and biotite form ite ϩ epidote, although not a precise indi- sition, as suggested by Patrick and Evans asymmetric pressure shadows aligned in the cator of metamorphic grade, is consistent [1989] elsewhere on the Seward Peninsula). S2 foliation around garnet (Figs. 7E and 7F), with M1a metamorphic conditions that are However, the extent of the greenschist-fa- implying that the retrogression of the garnet transitional from pumpellyite-actinolite to cies overprint in the two samples that do occurred before or during D2 deformation. blueschist facies (Brown, 1977). The re- contain blue amphibole (blue amphibole is The replacement of garnet by S2 biotite in placement of crossite by chlorite indicates present only as inclusions in chlorite- mafic schists provides some of the evidence that metamorphism later progressed to rimmed garnet in one sample; in the second for a greenschist-facies metamorphic event greenschist-facies conditions during M1b. sample only a single ragged grain partially (M2) synchronous with D2 deformation. Ap- Progressing southward, metamorphic replaced by chlorite and oxychlorite is proaching the Kigluaik Mountains, garnets minerals in the metabasites coarsen and present) suggests that the mineral assem- in mafic rocks become progressively more make up a progressively greater proportion blages in other metabasites in the area may retrograded until they disappear entirely of each sample (as opposed to relict igneous have crystallized under greenschist-facies 1–3 km from the M2 biotite-in isograd in minerals). About 16 km from the northern conditions and that evidence for the early pelitic rocks (Fig. 3). boundary of the map area, scattered meta- blueschist-facies history of these rocks has Metamorphism in the Kigluaik gneiss basites contain almandine-grossular garnet been obliterated. dome has been described by Lieberman

(Al39Sp15Py1Gr45 based on electron micro- The sample with the best-preserved glau- (1988), Miller et al. (1992), Calvert (1992), 2 probe data reported in full in Appendix B ), cophane in the Teller A-3 quadrangle is a and Amato et al. (1994). M1a and M1b as- suggesting an increase in temperature dur- quartz-rich rock collected from rubble-crop semblages are not preserved within the Kig- ing M1a metamorphism (Evans, 1990). about 4 km (2.5 miles) north of the southern luaik gneiss dome, although the rocks are These samples are the basis for the approx- border of the Teller quadrangle (HD-1b, believed to have undergone high-pressure/ imately located M1a garnet-in isograd Fig. 3). The glaucophane prisms lie within low-temperature metamorphism along with (Fig. 3). Garnet from one metagabbroic and help define a foliation that we believe is the overlying Nome Group (Patrick and sample from this area (LM-12b, Fig. 3) con- likely to be S1. Little to no metamorphic Lieberman, 1988). M2 metamorphic grade tains inclusions of glaucophane and epidote mineral growth occurs parallel to S2 in ad- increases from the flanks to the core of the (Figs. 7C and 7D). The matrix of this sample jacent schists. The glaucophane coexists range through a series of Barrovian pelitic consists of pale green actinolite rimmed by with quartz, white mica, epidote, titanite, isograds. The distance between the bi- darker blue-green hornblende, weakly foli- and calcite. The composition of glaucophane otite-in and sillimanite-in isograds is Ͻ2km ated, fine-grained dark blue-green horn- from this sample is shown in Figure 8. (Miller et al., 1992; Calvert, 1992) (Fig. 3). blende, epidote, titanite, apatite, chlorite, Farther south, metabasites contain no Peak metamorphism within the isograds and opaque oxides, and minor amounts of bi- remnants of M1a assemblages or of their ig- the core of the dome occurred synchronous otite and calcite. The composition of am- neous mineralogy. Metabasites west and with the formation of the dominant foliation phiboles in LM-12b is plotted in Figures 8 south of the Kigluaik Mountains typically (S2) (Miller et al., 1992; Calvert, 1992; and 9. Garnets typically are rimmed by chlo- contain actinolite and/or hornblende, epi- Amato et al., 1994). Peak metamorphic rite and/or biotite. A second metagabbroic dote, albite, chlorite, and titanite Ϯ garnet, pressure-temperature (P-T) conditions de- sample from a nearby locality (collected by biotite, white mica, quartz, and calcite. Ac- termined in the eastern Kigluaik Mountains B. Evans and B. Patrick, 1980 and 1984, and tinolite is generally rimmed by blue-green by Lieberman (1988) range from T ϭ 525– now residing in B. Evans’s Alaska blueschist hornblende (compositions plotted in Fig. 9). 575 ЊC, P ϭ 3.5–4.5 kbar in the staurolite collection, University of Washington) con- One sample in the southwestern part of our zone to T ϭ 800–850 ЊC, P ϭ 8–10 kbar for tains a single grain of ragged blue-purple study area (HC-36, Fig. 3; Fig. 9) contains granulites in the core of the range. amphibole included in epidote and partially barroisitic amphibole. Thurston (1985) re- replaced by chlorite, oxychlorite, and albite. ported actinolite rimmed by hornblende or GEOCHRONOLOGY/ The rest of the sample consists of pale blue- barroisite south of the eastern Kigluaik THERMOCHRONOLOGY green amphibole (sodic actinolite?), chlo- Mountains (Fig. 1). The compositions of the rite, epidote, albite, titanite, and apatite. actinolite from our study area are similar to Apatite fission-track, 40Ar/39Ar, and The original igneous texture of the metagab- his, but the hornblende rims on our amphi- U-Pb dating place important constraints on boles reach lower amounts of Si than those the timing of metamorphic and deforma- 2See footnote 1. Thurston (1985) reported. tional events on the southwestern Seward

Geological Society of America Bulletin, May 1995 545 HANNULA ET AL.

Peninsula. The details of the data are re- time between ca. 100 and 70 Ma and have the northern part of the study area. S2 in the ported elsewhere (apatite fission track: subsequently resided continuously at tem- northern part of the study area formed at 40 39 Dumitru et al., in press; Ar/ Ar: Calvert, peratures below 85 ЊC (and thus in the up- lower temperatures than did S1. Whereas S1 1992; Hannula and McWilliams, 1995; per 3–4 km of the crust, assuming a thermal is defined by metamorphic mineral growth U-Pb: Amato et al., 1994). Here we summa- gradient of 25 ЊC/km) (e.g., Naeser, 1979; (increasing in grade downward through the rize the relevant conclusions of each study. Green et al., 1989). The apatite ages are con- section), S2 in the northern part of the study U-Pb and 40Ar/39Ar ages from the Kig- sistent with two interpretations. They may area formed mainly by the processes of so- luaik gneiss dome can be used to constrain indicate that the rocks were reheated as a lution and reprecipitation, implying that de- the age of D2 deformation and high-grade result of increased heat input and cooled be- formation occurred at temperatures prob- (M2) metamorphism and cooling in the Kig- low Ϸ120–95 ЊC following peak metamor- ably significantly Ͻ400 ЊC (e.g., Gray and luaik gneiss dome. Monazites from amphib- phism in the Kigluaik gneiss dome. Alter- Durney, 1979). olite-facies pelites yield 91 Ϯ 1 Ma ages, natively, they may indicate cooling during Metamorphic conditions at deeper struc- which are interpreted to date the tempera- the unroofing of the upper part of the sec- tural levels, particularly in the Kigluaik ture peak of high-grade metamorphism tion sometime between 100 and 70 Ma. gneiss dome, were not high P/low T during

(Amato et al., 1994). Deformation (D2)in Apatite fission-track dating from the Kig- the formation of S2. In the Kigluaik gneiss the core of the gneiss dome occurred later luaik gneiss dome (Dumitru et al., in press) dome, the dominant (presumably S2) folia- than 105 Ϯ 3 Ma (the U-Pb zircon intrusive indicates that the deepest structural levels of tion formed during amphibolite- to granu- age of a garnet-bearing orthogneiss) and the southwestern Seward Peninsula under- lite-facies metamorphism, as documented was probably waning by 92 Ϯ 2 Ma (U-Pb went low-temperature cooling and unroof- by Miller et al. (1992), Calvert (1992), and zircon intrusive age of the syn- to late-tec- ing during the Eocene. The western Kig- Amato et al. (1994). At the intermediate tonic Kigluaik granite) (Amato et al., 1994). luaik Mountains underwent rapid cooling structural levels of the Nome Group, the dif- Rapid cooling of staurolite to sillimanite ϩ from ϳ125 to 60 ЊC at ca. 53 Ma. The core ference between metamorphic conditions

K-feldspar–grade rocks from 500 ЊCto of the range underwent protracted cooling when S1 and S2 formed is less obvious. Both 300 ЊC between 87 Ma and 83.5 Ma is doc- through 125–60 ЊC and cooled through S1 and S2 are mostly defined by mineral as- umented by 40Ar/39Ar dating of hornblende, ϳ100 ЊC between 42 and 33 Ma. semblages that can be stable under either biotite, and muscovite (Calvert et al., 1991; blueschist- or greenschist-facies conditions. Calvert, 1992). DISCUSSION The occurrence of rare syntectonic biotite in

Phengites, which define the S1 foliation in S2 (Figs. 5F, 7E, and 7F) indicates that S2 the Teller area, yield 40Ar/39Ar plateau ages Structural History most likely formed under greenschist-facies ranging from 116 to 125 Ma (Fig. 3) (Han- conditions. Furthermore, as others have ar- nula and McWilliams, 1995). Depending on The interpretation of the timing and tec- gued (Thurston, 1985; Patrick and Lieber- the maximum temperature achieved by the tonic origin of the S2 foliation is controver- man, 1988), the orientations and style of rocks, the ages may represent either the age sial. Others (Thurston, 1985; Patrick, 1988) structures we associate with D2 (S2 and Ls) of highest temperature metamorphism have described the gently dipping dominant are quite similar in the Nome Group and

(probably M1b), or the cooling of the rocks foliation in the Nome Group as a transpo- along the flanks of the Kigluaik gneiss dome through ϳ350 ЊC (Purdy and Ja¨ger, 1976). sition foliation, formed within the same tec- (Fig. 2). S2 appears to have formed during a Phengites from south and west of the Kig- tonic cycle and at the same high-P/low-T met- separate deformational event from S1 at all luaik Mountains yield disturbed ages, some amorphic conditions as S1. We disagree with structural levels. of which are unreasonably old (up to 383 their interpretation. We interpret the S1 and Ma) (Hannula and McWilliams, 1995). The S2 foliations as having formed at different Metamorphic P-T Path and Relative Timing disturbed spectra probably reflect the pres- times, under different metamorphic and tec- of Metamorphism and Deformation ence of excess argon, and these ages are tonic circumstances. therefore not useful in defining the meta- Metamorphic conditions were clearly dif- Our interpretation of the metamorphic morphic and deformational history of the ferent during the formation of S1 and S2 in P-T history and the relative timing of D1 and rocks. Apatite fission-track ages from the Teller area (Dumitru et al., in press) (Fig. 3) are ã derived from samples with low apatite yields and poor grain quality. As a result, the un- Figure 7. (A) Crossite oriented perpendicular to the dominant foliation (S1?) with chlo- certainties on the ages are substantial and rite pressure shadows. Long direction of photo is 1.44 mm across. From sample locality ؍ chlorite; Ab ؍ crossite; Chl ؍ poorly known. Despite this, all ages but one LD-13 in Figure 3. (B) Sketch of photomicrograph A. Cr mixture of titanite and ilmenite. (C) Garnet containing inclusions of ؍ fall within the range 106 Ϯ 11 to 75 Ϯ 4 albite; Ttn/ilm ,Ϯ1␴), with a weighted mean age of 83 Ma. blue-purple amphibole and epidote and surrounded by a matrix of chlorite ؎ biotite, albite) These ages overlap the 91 Ma age of peak green amphibole, epidote, and titanite. Long direction of photo is 1.44 mm across. From ;garnet ؍ metamorphism and pluton intrusion and the sample locality LM-12b in Figure 3. (D) Sketch of photomicrograph C. Grt ;chlorite ؍ blue-green amphibole; Chl ؍ epidote; Act ؍ blue-purple amphibole; Ep ؍ Ma 40Ar/39Ar 500–300 ЊC cooling Gln 87–83 -albite. (E) Garnets partially replaced by biotite ؎ chlorite in a chlor ؍ biotite; Ab ؍ ages in the Kigluaik gneiss dome (Calvert, Bt 1992; Amato et al., 1994). The apatite ages ite-albite schist. Biotite forms asymmetric pressure shadows around garnet. Sample lo- indicate that rocks in the northern part of cality is shown as 7E in Figure 3. Long direction of photo is 1.44 mm. (F) Sketch of garnet the area cooled below ϳ120–85 ЊC some- porphyroblasts and asymmetric biotite pressure shadows from photomicrograph E.

546 Geological Society of America Bulletin, May 1995 A B

C D

E F

Geological Society of America Bulletin, May 1995 547 HANNULA ET AL.

schist-facies (M1b) metamorphic conditions by a reaction such as

pumpellyite ϩ chlorite ϩ quartz ϭ

clinozoisite ϩ tremolite ϩ H2O

(Liou et al., 1985). The fine-grained chlorite

and white mica, which define the S1 foliation in the northern Teller area, are compatible with deformation under a wide range of relatively low-temperature conditions. The prevalence of greenschist-facies assem- blages in surrounding metabasites, however,

Figure 8. Classification of sodic amphiboles (after Leake, 1978). Electron microprobe suggests that the minerals defining the S1 3؉ data are reported in full in Appendix B. Fe content was calculated as suggested by foliation grew during M1b. The S2 foliation, Robinson et al. (1982), normalizing to 15 cations exclusive of K. See Figure 3 for sample on the other hand, formed primarily by low- localities. temperature dissolution of quartz and cal- cite (Ͻ400 ЊC, e.g., Gray and Durney, 1979).

The formation of S2 at low temperatures D2 deformation within that history are illus- mations, one (D1) compatible with could have occurred during the ϳ70–100 trated by the schematic P-T paths shown in greenschist-facies metamorphism, and one Ma cooling of these rocks below ϳ100 ЊC Figure 10. that occurred at lower temperatures. North (Dumitru et al., in press). In the northern part of the study area (up-section) of the pumpellyite-out isograd In the central and southern parts of the (‘‘North Teller’’ curve, Fig. 10), metabasites (Figs. 3 and 4), the metamorphic mineral Teller area (‘‘South Teller’’ curve, Fig. 10), preserve evidence of an early high-pressure assemblage (pumpellyite ϩ chlorite ϩ epi- mafic rocks provide evidence for M1a and metamorphism (M1a) and a possible green- dote ϩ actinolite, with epidote replacing M1b metamorphism, whereas metasedi- schist-facies overprint (M1b), whereas meta- pumpellyite) suggests an evolution from ments contain an S1 foliation compatible sediments contain evidence for two defor- pumpellyite-actinolite- (M1a) to green- with greenschist-facies metamorphism and an S2 foliation that formed under lower temperature conditions. Just south of the pumpellyite-out isograd, crossite replaced by foliated chlorite (LD-13, Figs. 3, 7A, and 7B) indicates a P-T path from transitional

blueschist facies (M1a) to greenschist facies (M1b), with a foliation (S1) formed during M1b. Evidence for a similar history can be found farther south (LM-12b, Figs. 3, 7C, and 7D), where one metabasite contains in- clusions of glaucophane ϩ epidote in garnet surrounded by a matrix of weakly foliated chlorite, actinolite rimmed with hornblende, epidote, titanite, apatite, opaque oxides, and minor amounts of biotite and calcite. In metasediments, most mineral growth took

place during the formation of S1 and is com- patible with deformation during green- schist-facies metamorphism. The mineral assemblages in phyllites and mica schists vary little from north to south (quartz ϩ al- bite ϩ white mica ϩ chlorite ϩ epidote Ϯ stilpnomelane), although the grain size of metamorphic minerals coarsens significantly from north to south. Metamorphism during

D2 was probably subgreenschist to lower greenschist facies, as evidenced by the dis- Figure 9. Classification of (A) calcic and (B) sodic-calcic amphiboles (after Leake, 1978). crete crenulation cleavage and the lack of 3؉ Electron microprobe data are reported in full in Appendix B. Fe content was calculated new growth of chlorite ϩ white mica in S2 as suggested by Robinson et al. (1982), normalizing to 13 cations exclusive of K, Na, and until ϳ4 km (2.5 miles) north of the south- ferro-actinolitic ern quadrangle boundary (Fig. 3). White ؍ actinolitic hornblende; FAH ؍ Ca. See Figure 3 for sample localities. AH hornblende. mica 40Ar/39Ar plateau ages indicate cooling

548 Geological Society of America Bulletin, May 1995 STRUCTURAL AND METAMORPHIC RELATIONS, SEWARD PENINSULA

net, biotite, white mica, quartz, and calcite), implying a later evolution to epidote-amphib-

olite-facies conditions (M1b?). In schists south and west of the Kigluaik Mountains (‘‘Nome Group S of Kigluaiks,’’ Fig. 10), abundant new growth of chlorite, white mica, albite, and biotite (Figs. 5D, 5E,

and 5F) occurs syntectonic with S2. The presence of biotite, both in pressure shad- ows around albite (Fig. 5F) and as a syntec- tonic replacement of garnet (Figs. 7E and

7F) in chloritic schists, indicates that S2 formed under greenschist-facies conditions

(M2), not under blueschist-facies conditions as argued by Thurston (1985) and Patrick and Evans (1989). Schematic P-T paths from the Kigluaik gneiss dome (‘‘Kigluaik Mts. sillimanite zone’’ and ‘‘Kigluaik Mts. granulites’’ in Fig. 10) are derived from data collected by Lieberman (1988) and Amato et al. (1992,

1994). Although direct evidence for M1a and M1b metamorphism is lacking in the Kig- Figure 10. Schematic pressure-temperature-deformation diagram. All pressure- luaik gneiss dome, we agree with Thurston temperature paths are highly schematic and are intended only as an illustration of our (1985) and Patrick and Lieberman (1988) ideas about the metamorphic and deformational evolution of the southwestern Seward that the high-grade rocks of the Kigluaik Peninsula, not as the representation of a set of quantitative P-T data. P-T paths of North Mountains probably underwent high-pres- Teller, South Teller, and Nome Group south of Kigluaiks are based on qualitative estimates sure metamorphism along with the overlying of pressure and temperature from mineral assemblages and textures in metabasites. Argon Nome Group. Lieberman (1988) calculated white mica ages from the Teller area are from Hannula and McWilliams (1995). Apatite P-T conditions of 700–750 ЊC, 5–6 kbar for fission-track ages from the Teller area are from Dumitru et al. (in press). P-T conditions the sillimanite ϩ K-feldspar zone and 800– at peak temperatures for rocks from the Kigluaik Mountains (sillimanite zone and gran- 850 ЊC, 8–10 kbar for granulites in the east- ulites) are from Lieberman (1988). The age and pressure of intrusion of the Kigluaik pluton ern Kigluaik Mountains. The presence of are from Amato et al. (1992, 1994). Lower temperature cooling ages of rocks from the rare early kyanite (Thurston, 1985; A. Cal- Kigluaik Mountains are from Calvert (1992). ‘‘North Teller’’ refers to rocks north of the vert, 1991, personal commun.) and late an- pumpellyite-out isograd; ‘‘South Teller’’ refers to rocks south of the pumpellyite-out iso- dalusite-bearing pegmatite dikes (Thurston, grad but north of the first occurrence of new mineral growth synchronous with D .Lw 2 1985) within the isograds flanking the Kig- luaiks and hercynite ϩ cordierite within the ؍ epidote blueschist facies (Evans, 1990); GS ؍ lawsonite blueschist facies; Ep BS ؍ BS (pumpellyite-actinolite facies; core of the gneiss dome (Amato et al., 1994 ؍ epidote amphibolite facies; PA ؍ greenschist facies; EA -suggests that the rocks underwent signifi ؍ prehnite-pumpellyite facies; KFS ؍ zeolite facies; PP ؍ amphibolite facies; Zeo ؍ Am -apatite fission track. cant decompression at elevated tempera ؍ K-feldspar; ApFT tures. High-grade metamorphism and plu- through ϳ350 ЊC after D1 deformation and 1990). The pressure and temperature esti- ton intrusion occurred syntectonically at ca. M1b metamorphism took place at ca. 120 mates for M1a in the Nome Group south of 91 Ma (Amato et al., 1994) and were fol- Ma (Hannula and McWilliams, 1995). The the Kigluaik Mountains (ϳ460 ЊC, 12 kbar) lowed by rapid cooling to ϳ300 ЊCby83Ma formation of the S2 foliation at subgreen- are taken from Patrick and Evans (1989). (Calvert, 1992). schist-facies conditions is compatible with Although metabasites in the southern part its formation during ca. 70–100 Ma cooling of our study area do not contain glauco- Structural Section through ϳ100 ЊC (Dumitru et al., in press). phane, they do contain actinolite similar in

Information about M1a is less well pre- composition to that which Thurston (1985) A schematic and highly generalized served west and south of the Kigluaik argued coexisted stably with glaucophane in crustal section for the area north of the Kig- Mountains (‘‘Nome Group S of Kigluaiks,’’ the Salmon Lake area. It is unclear whether luaik Mountains was constructed utilizing Fig. 10). Although we found no diagnostic the actinolite in the southern part of our unit thicknesses obtained from the cross sec- blueschist-facies assemblages in our study study area grew at blueschist-facies condi- tion shown in Figure 4 (Fig. 11). The sche- area south of the Kigluaik Mountains, such tions (as Thurston [1985] and Patrick and matic crustal column shows the approximate assemblages have been found at presumably Evans [1989] argued occurred elsewhere on inferred depths in the crust of rocks from similar structural levels elsewhere on the the Seward Peninsula), or if it grew under the southwestern Seward Peninsula meas-

Seward Peninsula (south of eastern Kigluaik greenschist-facies conditions. The actinolite ured in relation to the attitude of D2 struc- Mountains, Thurston, 1985; Patrick and is rimmed by hornblende (coexisting with tures. We do not have independent control Evans, 1989; Solomon quadrangle, Buxton, epidote, albite, chlorite, and titanite Ϯ gar- on paleohorizontal; metamorphic grade ap-

Geological Society of America Bulletin, May 1995 549 HANNULA ET AL.

free metabasites locally preserve blueschist- facies mineral assemblages, and 2 km of rocks with rare pumpellyite-actinolite-facies assemblages. At the top of the column we added the approximate thickness of faulted Ordovician-Silurian and younger limestones found in the York Mountains mapped by Sainsbury (1972). We used the present-day structural thickness of limestone as meas- ured from the top of the highest peak to the low-angle faults bounding these carbonates below, which is much less than the origi- nal stratigraphic thickness of the section (ϳ5 km). The schematic restoration of the crustal column places the Kigluaik pluton at a depth of ϳ15 km (ϳ3.9 kbar). If a significant amount of material has been eroded from the top of the crustal column, this would be a minimum depth. Alternatively, if the dom- inant foliation was dome shaped during in- trusion of the Kigluaik pluton, the intrusion depth should be much less than this esti- mate. Amato et al. (1992) calculated that the Kigluaik granite intruded at a pressure of 3–4 kbar based on aluminum-in-horn- blende geobarometry. This pressure is sim- ilar to that estimated from the crustal col- umn. The fission-track data presented earlier suggest that the rocks shown in the upper 6 km of this structural column were not far from the surface after 70–100 Ma. The pressures and temperatures esti-

mated for the M1a metamorphic isograds do not make sense given their location within the current structural section. The pressures implied by the pumpellyite-actinolite and blueschist-facies mineral assemblages are clearly not accounted for by the structural Figure 11. Schematic structural section through the western Seward Peninsula based on overburden that can be reconstructed from cross-section thicknesses shown in Figure 4. Thickness of carbonate section in the York the geology (Fig. 11). Furthermore, the iso- Mountains is estimated from Sainsbury (1972). Sodic amphibole localities are shown by grads may be more closely spaced than one black diamonds (glaucophane) and gray diamonds (crossite). D temperatures and pres- 2 would expect for a blueschist-facies thermal sures of mineral zones within the Kigluaik gneiss dome and location of garnet lherzolite gradient. The temperatures at which these are based on the work of Lieberman (1988) in the eastern Kigluaik Mountains. Cooling isograds form are not well known and de- ages of high-grade rocks are from Calvert (1992). The age of the Kigluaik pluton is from pend strongly on bulk-rock composition Amato et al. (1994). Maximum temperatures in the York Mountains are from Anita Harris (Liou et al., 1985; Evans, 1990; Oh et al., (personal commun.). Locations of geochronology samples from the Teller region were 1991). Experiments suggest that pumpelly- projected onto cross section A–A؅ (Fig. 4) parallel to strike of the foliation (generally ite-actinolite-facies assemblages give way parallel or subparallel to lithologic layering in these locations). Dark gray units represent to blueschist- or greenschist-facies assem- metabasites from the Teller region. blages at temperatures Ͻ375 ЊC (Liou et al., 1985). Thermodynamic calculations suggest pears to increase perpendicular to S2 luaik gneiss dome; ϳ3.5 km of chloritic that garnet appears in blueschist-facies met- (Figs. 3 and 4), so we believe that the current schists (the Nome Group of Sainsbury abasites at temperatures Ͼ475 ЊC, although map view represents an oblique crustal sec- [1972]); and 5 km of lower-grade rocks (our field evidence suggests that a temperature tion. The depths on this section are only units Pzmvu, Pzqms, Pzb, and Pzsl, included between 400 and 460 ЊC may be more real- crudely approximated. in Sainsbury’s [1972] slate of the York re- istic (Evans, 1990). Oh et al. (1991) showed The crustal section we have constructed gion). Within the York slate we can differ- that the temperature at which garnet ap- consists of, from bottom to top (Fig. 11): 5 entiate ϳ1 km of rocks in which metabasites pears in blueschists and eclogites is strongly km of rocks above biotite grade in the Kig- contain rare garnet, 2 km in which garnet- dependent on the amount of Mn in the

550 Geological Society of America Bulletin, May 1995 STRUCTURAL AND METAMORPHIC RELATIONS, SEWARD PENINSULA rocks. The composition of garnet from the to be continuous and decreases in metamor- level tin granites that cut the faults (Hudson garnet-in isograd in our study area is phic grade up-section. (2) The S2 foliation is and Arth, 1983). Thus it is possible that Al39Sp15Py1Gr45. Caron and Pe´quignot gently dipping and axial planar to recum- movement along these faults took place dur- (1986) reported garnet with a similar Mn bent isoclinal folds of an earlier foliation ev- ing or soon after pluton emplacement and content (Sp11–21) from 420 ЊC eclogites. erywhere in the section (Fig. 2). (3) The peak metamorphism in the Kigluaik gneiss Thus the lower temperature estimates (400– deformationoccurredduringincreasingmet- dome and may be responsible for the cool- 420 ЊC) probably are the best approxima- amorphic temperatures (and pluton intru- ing recorded by the 70–100 Ma apatite fis- tions for garnet-in temperatures in our study sion) deep in the section, but at very low sion-track ages on rocks near Teller. area. The minimum thermal gradient im- temperatures at higher structural levels. (4) The progression of structural fabrics fur- plied by these temperature estimates is Structural fabrics progress from more or less ther suggests that the cooling of the upper 12.5 ЊC/km. If the pumpellyite-out isograd penetrative (depending on position in the part of the structural section occurred dur- represents a lower temperature (as would be structural section) S2 and Ls to calcite-filled ing D2 deformation. Ductile thinning of the expected with the addition of Fe to the ex- tension gashes, implying similar stretching crust also provides a possible explanation perimental system of Liou et al. [1985]) or if directions during decreasing temperature. for both the somewhat high M1a thermal the garnet-in isograd formed at a higher (5) The age of metamorphism synchronous gradient suggested by the structural distance temperature, the implied thermal gradient with deformation at the deepest structural between the M1a pumpellyite-out and gar- would be even higher. Combined pressure levels (91 Ma, Amato et al., 1994) is within net-in isograds, and for the jumps in M2 and temperature estimates from eclogites error of apatite fission-track cooling ages pressure recorded by the staurolite-grade to elsewhere on the Seward Peninsula (Patrick from shallower structural levels (Figs. 10 granulite-facies rocks of the Kigluaik gneiss and Evans, 1989) suggest a lower thermal and 11). (6) Both M1a (in the Teller area) dome (Figs. 10 and 11). Finally, there is no gradient, ϳ10.7 ЊC/km. It is possible that the and M2 isograds (in the Kigluaik Moun- evidence for the duplication of section or pumpellyite-out and garnet-in isograds may tains) are unusually closely spaced (Figs. 3, emplacement of high-grade over low-grade have formed at different times (and thus 4, and 11). (7) There is no evidence for ei- rocks that one might expect to find in an represent a ‘‘metamorphic field gradient’’ ther duplication of section or high-grade area that had not undergone deformation [Spear et al., 1984] rather than a true geo- rocks structurally above low-grade rocks since it had been subducted. thermal gradient). Alternatively, the unusu- anywhere on the southwestern Seward Crustal extension may also explain some ally close spacing between the isograds may Peninsula. other geologic observations on the Seward be the result of D2 deformation, or a com- All seven of these observations can be ex- Peninsula. Patrick (1988) described quartz bination of these two effects. plained if D2 deformation took place during petrofabrics indicating a top-to-the-north M2 isograds are also unusually closely regional extension. The metamorphic core shear sense from a number of localities spaced (Fig. 11) (Miller et al., 1992). Fur- complexes of the western United States, within the Nome Group. Although Patrick thermore, the structural thicknesses be- where the deeper levels of Basin and Range (1988) ascribed these fabrics to northward- tween the various mineral zones are too extension are exposed, are typified by sub- directed overthrusting of an arc terrane over small to account for the pressure differences horizontal metamorphic foliations (Coney, the Seward Peninsula, this sense of shear recorded between adjacent zones (staurolite 1980), and recumbent isoclinal folds in the could as easily have resulted from top-to- zone: 3.5–4.5 kbar; sillimanite zone: 4–5 Alps similar to those on the Seward Penin- the-north extensional removal of the York kbar; sillimanite ϩ K-feldspar zone: 5–6 sula have been ascribed to late-stage exten- Mountain limestones from above the Nome kbar, Lieberman, 1988). These pressures sion and termed ‘‘collapse folding’’ by Group. may also possibly have been recorded at dif- Froitzheim (1992). Any scenario that com- Finally, the Seward Peninsula is currently ferent times corresponding to the time the bines plutonism at depth with some kind of an area with ϳ30- to 35-km-thick continen- particular isograd experienced its maximum unroofing (erosional or tectonic) can pro- tal crust (Barnes, 1977), surrounded on temperature. The cooling of the Kigluaik duce synchronous heating at depth, cooling three sides by a region of even thinner crust gneiss dome from peak metamorphic con- at shallower levels, and, in general, closely (the Bering Strait, Norton Sound, and ditions occurred very rapidly from 725 ЊCto spaced isotherms within the crust. Chukchi Sea). The current crustal thick- ϳ300 ЊC between 91 and 83 Ma at all met- A close spatial and temporal association nesses are clearly much less than those that amorphic grades, however (Amato et al., between magmatism and normal faulting must have existed during the blueschist-fa- 1992; Calvert, 1992; Amato et al., 1994), so has been demonstrated in many places in cies metamorphism of the Nome Group. it is unlikely that the mineral assemblages the Basin and Range (Gans et al., 1989). On The deepest basins in the Bering Sea are recorded drastically different equilibrium the Seward Peninsula, numerous faults that Tertiary in age and thought to be related to pressure conditions at different times during place shallower on deeper structural levels strike-slip or transtensional faulting (Tol- uplift. Thus the close spacing of these iso- are found in the York Mountains, at the son, 1987; Worrall, 1991), but this faulting grads and the pressure gaps across them shallowest crustal levels exposed (Fig. 1). does not involve significant deformation far may also be the result of D2 deformation. These faults were identified as thrust faults from the faults and thus is unlikely to be the by Sainsbury (1972), but because they in- cause of the loss of crustal thickness across

Significance of D2 Deformation variably place less metamorphosed on more the rest of the Bering Sea region. A seismi- metamorphosed rocks or younger strata on cally active normal fault that cuts Holocene Any tectonic scenario to explain the older strata, we suspect they may be normal glacial deposits (Hudson and Plafker, 1978) events that occurred synchronous with D2 faults. The timing of movement along these bounds the north side of the Kigluaik Moun- must take the following points into consid- faults is poorly defined, but it must be earlier tains (Fig. 1). Eocene apatite fission-track eration: (1) The structural section appears than the ca. 70–80 Ma age of the shallow- ages from within the Kigluaik Mountains,

Geological Society of America Bulletin, May 1995 551 HANNULA ET AL. however, indicate that unroofing due to late cies in the southern parts of the slate of the schist collection, and for pointing us in the

Cenozoic movement on that normal fault York region and the Nome Group) to a M1b direction of the Teller quadrangle in the first must be Ͻϳ2–3 km (Dumitru et al., in higher-temperature overprint (greenschist place. We would also like to thank Andy press). Although the age of the loss of facies in the slate of the York region; green- Calvert and Liz Symchych for their assist- crustal thickness in the Bering Strait region schist to epidote-amphibolite facies in the ance in the field and Jeff Amato for taking is imprecisely known, it is likely that much of Nome Group). last minute photomicrographs. This manu- it occurred in the Cretaceous. Both groups of rocks also were subjected script has been improved by the comments Numerous recent studies ascribe some of to two distinct deformational events. The of J. G. Liou, Cynthia Dusel-Bacon, Alison the exhumation of various high-pressure ter- first accompanied the metamorphic pro- Till, and an anonymous reviewer. ranes to extension (Lister et al., 1984; Platt gression from M1a high-pressure metamor- and Vissers, 1989; Anderson and Jamtveit, phism to M1b greenschist-facies metamor- REFERENCES CITED 1990; Ave´ Lallement and Guth, 1990; Little phism and took place at or before 125 Ma. Amato, J. M., Wright, J. E., and Gans, P. B., 1992, The nature and age of Cretaceous magmatism and metamorphism on the et al., 1992). Various observations from the The second formed the prominent, gently Seward Peninsula, Alaska: Geological Society of America Seward Peninsula, however, suggest that ex- dipping foliation visible throughout both Abstracts with Programs, v. 24, no. 5, p. 2. Amato, J. M., Wright, J. E., Gans, P. B., and Miller, E. L., 1994, tension was not the only mechanism to play groups of rocks and the structurally deeper Magmatically induced metamorphism and deformation in the Kigluaik gneiss dome, Seward Peninsula, Alaska: Tec- a role in the uplift of the Nome Group. As Kigluaik gneiss dome and formed at tem- tonics, v. 13, p. 515–527. discussed earlier, the rocks of the southwest- peratures varying from very low (probably Anderson, T. B., and Jamtveit, B., 1990, Uplift of deep crust during orogenic extensional collapse: A model based on field stud- ern Seward Peninsula experienced a pro- ϽϽ 400 ЊC) in the slate of the York region, ies in the Sogn-Sunnfjord region of western Norway: Tec- tonics, v. 9, p. 1097–1111. gression of metamorphic conditions from through greenschist facies (ϳ300–500 ЊC) in Armstrong, R. L., Harakal, J. E., Forbes, R. B., Evans, B. W., and M blueschist- to M greenschist-facies the Nome Group, to upper amphibolite and Thurston, S. P., 1986, Rb-Sr and K-Ar study of metamorphic 1a 1b rocks of the Seward Peninsula and Southern Brooks Range, conditions (or M1a pumpellyite-actinolite- granulite facies at the core of the Kigluaik Alaska: Geological Society of America Memoir 164, p. 184–203. to M1b greenschist-facies conditions at shal- gneiss dome, where peak metamorphism is Ave´ Lallement, H. G., and Guth, L. R., 1990, Role of extensional lower levels) prior to D deformation dated at ca. 91 Ma (Amato et al., 1994). tectonics in exhumation of eclogites and blueschists in an 2 oblique subduction setting, northeastern Venezuela: Geol- (Fig. 10). This suggests that some exhuma- Strain associated with the second deforma- ogy, v. 18, p. 950–953. Barnes, D. F., 1977, Bouguer gravity map of Alaska: U.S. Geo- tion and heating (probably due to relaxation tional event also increases down-section. logical Survey Geophysical Investigations Map GP-913, scale 1:2 500 000. of a depressed subduction-zone thermal This deformation appears to have produced Bell, T. H., and Brothers, R. N., 1985, Development of P-T pro- gradient, as suggested by Patrick and significant vertical attenuation of the struc- grade and P-retrograde, T-prograde isogradic surfaces dur- ing blueschist to eclogite regional deformation/metamor- Lieberman [1988]) occurred before the ex- tural section, resulting in close spacing of phism in New Caledonia, as indicated by progressively developed porphyroblast microstructures: Journal of Met- tensional D2 deformation began. Further- both M1a and M2 isograds, and is partially amorphic Geology, v. 3, p. 59–78. more, even the original stratigraphic thick- responsible for the current 30–35 km thick- Brown, E. H., 1977, Phase equilibria among pumpellyite, law- sonite, epidote and associated minerals in low grade meta- ness of ‘‘upper plate’’ rocks now exposed in ness of the previously thickened crust. Cool- morphic rocks: Contributions to Mineralogy and Petrology, v. 64, p. 123–136. the York Mountains (ϳ5.5 km) is insuffi- ing and unroofing of the lower-grade rocks Buxton, C. L., 1990, Geology and pre-metamorphic evolution of cient to explain the metamorphic pressures occurred during or soon after this second the Nome Group blueschist terrane, Horton Creek area, Seward Peninsula, Alaska [Master’s thesis]: Seattle, Univer- experienced by even the pumpellyite-actin- deformational event, as evidenced by the sity of Washington, 115 p. Calvert, A. T., 1992, Structural evolution and thermochronology of olite-facies rocks (Fig. 11). Together these 70–100 Ma apatite fission-track ages from the Kigluaik Mountains, Seward Peninsula, Alaska [Mas- observations suggest that some of the orig- these rocks, and may have taken place dur- ter’s thesis]: Stanford, California, Stanford University, 86 p. Calvert, A. T., Gans, P. B., Amato, J. M., Miller, E. L., and inal overburden of the Seward Peninsula ing normal faulting within the York Moun- O’Sullivan, P., 1991, Uplift and cooling history of the Kig- luaik metamorphic complex, Seward Peninsula, Alaska: Ge- blueschists (possibly including an ophiolite tains carbonates. This extensional deforma- ological Society of America Abstracts with Programs, v. 23, no. 5, p. A435. equivalent to the Angayucham terrane, tion is clearly at least in part responsible for Caron, J.-M., and Pe´quignot, G., 1986, The transition between which structurally overlies similar blue- the exhumation of the blueschists but does blueschists and lawsonite-bearing eclogites based on obser- vations from Corsican metabasalts: Lithos, v. 19, p. 205–218. schist-facies rocks in the southern Brooks not preclude earlier erosional or extensional Cloos, M., 1982, Flow melanges: Numerical modeling and geologic constraints on their origin in the Franciscan subduction Range) was removed, most likely by erosion unroofing. complex, California: Geological Society of America Bulle- or faulting, prior to the ductile extensional tin, v. 93, p. 330–345. Coney, P. J., 1980, Cordilleran metamorphic core complexes: An deformation recorded in our map area. ACKNOWLEDGMENTS overview, in Crittenden, M. D., Coney, P. J., and Davis, G. H., eds., Cordilleran metamorphic core complexes: Ge- ological Society of America Memoir 153, p. 7–31. CONCLUSIONS This paper represents one chapter of a Dewey, J. F., 1988, Extensional collapse of orogens: Tectonics, v. 7, p. 1123–1139. Ph.D. dissertation by the first author at Dumitru, T. A., Miller, E. L., O’Sullivan, P. B., Amato, J. M., Hannula, K. A., Calvert, A. T., and Gans, P. B., in press, This study ties together the metamorphic Stanford University, based on research car- Cretaceous to Recent extension in the Bering Strait region, and structural histories of rocks from a va- ried out during tenure of a National Science Alaska: Tectonics. Ernst, W. G., 1988, Tectonic history of subduction zones inferred riety of crustal levels now exposed on the Foundation Graduate Fellowship. Support from retrograde blueschist P-T paths: Geology, v. 16, p. 1081–1084. southwestern Seward Peninsula. Rocks pre- for field and laboratory expenses came from Evans, B. W., 1990, Phase relations of epidote-blueschists: Lithos, v. 25, p. 3–23. viously mapped as the ‘‘slate of the York the GSA J. Dillon Alaska Research Award, Forbes, R. B., Evans, B. W., and Thurston, S. P., 1984, Regional region’’ (Fig. 1) have undergone a metamor- a GSA Penrose grant, the McGee Fund progressive metamorphism, Seward Peninsula, Alaska: Journal of Metamorphic Geology, v. 2, p. 43–54. phic and deformational history related to (Stanford University), and Chevron Field Froitzheim, N., 1992, Formation of recumbent folds during synor- ogenic crustal extension (Austroalpine nappes, Switzer- rocks assigned to the Nome Group. Both Thesis Awards to K. Hannula and from land): Geology, v. 20, p. 923–926. groups of rocks have undergone a progres- National Science Foundation grant EAR- Gans, P. B., Mahood, G. A., and Schermer, E., 1989, Synexten- sional magmatism in the Basin and Range Province: A case sive metamorphism from M1a high-pressure/ 9018922 to E. Miller. study from the Eastern Great Basin: Geological Society of America Special Paper 233, 53 p. low-temperature metamorphism (pumpelly- We would like to thank Bernard Evans Gray, D. R., and Durney, D. W., 1979, Crenulation cleavage dif- ite-actinolite facies in the northern parts of and Brian Patrick for allowing us to examine ferentiation: Implications of solution-deposition processes: Journal of Structural Geology, v. 1, p. 73–80. the slate of the York region, blueschist fa- and discuss samples from their Alaska blue- Green, P. F., Duddy, I. R., Gleadow, A. J. W., and Lovering, J. F.,

552 Geological Society of America Bulletin, May 1995 STRUCTURAL AND METAMORPHIC RELATIONS, SEWARD PENINSULA

1989, Apatite fission track analysis as a paleotemperature synthesis of northern Alaska: U.S. Geological Survey metamorphism and deformation in the Sesia Zone, Western indicator for hydrocarbon exploration, in Naeser, N. D., and Open-File Report 92-0330, 183 p. Alps: Journal of Geology, v. 92, p. 21–36. McCullogh, T. H., ed., Thermal history of sedimentary ba- Mull, C. G., 1985, Cretaceous tectonics, depositional cycles, and Sainsbury, C. L., 1972, Geologic map of the Teller quadran- sins: Methods and case studies: New York, Springer-Verlag, the Nanushuk Group, Brooks Range and Arctic Slope, gle, western Seward Peninsula, Alaska: U.S. Geological Sur- p. 181–195. Alaska, in Huffman, A. C., ed., Geology of the Nanushuk vey Miscellaneous Geologic Investigations Map I-685, Hannula, K. A., and McWilliams, M. O., 1995, Reconsideration of Group and related rocks, North Slope, Alaska: U.S. Geo- scale 1:250 000. the age of blueschist-facies metamorphism on the Seward logical Survey Bulletin, v. B1614, p. 7–36. Sainsbury, C. L., Coleman, R. G., and Kachadoorian, R., 1970, Peninsula, Alaska, based on phengite 40Ar/39Ar results: Naeser, C. W., 1979, Fission-track dating and geologic annealing Blueschist and related greenschist facies rocks of the Seward Journal of Metamorphic Geology, v. 13, p. 125–139. of fission tracks, in Ja¨ger, E., and Hunziker, J. C., ed., Lec- Peninsula, Alaska: U.S. Geological Survey Professional Harris, L., 1985, Progressive and polyphase deformation of the tures in isotope geology: New York, Springer-Verlag, Paper 700-B, p. B33–B42. Schistes Lustres in Cap Corse, Alpine Corsica: Journal of p. 154–169. Sainsbury, C. L., Hummel, C. L., and Hudson, T., 1972, Recon- Structural Geology, v. 7, p. 637–650. Oh, C., Liou, J. G., and Maruyama, S., 1991, Low-temperature naissance geologic map of the Nome quadrangle, Seward Helper, M. A., 1986, Deformation and high P/T metamorphism in eclogites and eclogitic schists in Mn-rich metabasites in Peninsula, Alaska: U.S. Geological Survey Open-File Re- the central part of the Condrey Mountain window, north- Ward Creek, California; Mn and Fe effects on the transition port 72-326, scale 1:250 000. central Klamath Mountains, California and Oregon, in between blueschist and eclogite: Journal of Petrology, v. 32, Sanderson, D. J., 1973, The development of fold axes oblique to Evans, B. W., ed., Blueschists and eclogites: Geological So- p. 275–301. the regional trend: Tectonophysics, v. 16, p. 55–70. ciety of America Memoir 164, p. 125–141. Patrick, B. E., 1988, Synmetamorphic structural evolution of the Spear, F. S., Selverstone, J., Hickmott, D., Crowley, P., and Hudson, T., and Arth, J. G., 1983, Tin granites of the Seward Seward Peninsula blueschist terrane, Alaska: Journal of Hodges, K. V., 1984, P-T paths from garnet zoning: A new Peninsula, Alaska: Geological Society of America Bulletin, Structural Geology, v. 10, p. 555–565. technique for deciphering tectonic processes in crystalline v. 94, p. 768–790. Patrick, B. E., and Evans, B. W., 1989, Metamorphic evolution of terranes: Geology, v. 12, p. 87–90. Hudson, T., and Plafker, G., 1978, Kigluaik and Bendeleben faults, the Seward Peninsula blueschist terrane: Journal of Petrol- Suppe, J., 1979, Structural interpretation of the southern part of Seward Peninsula: U.S. Geological Survey Circular C ogy, v. 30, p. 531–555. the northern Coast Ranges and Sacramento Valley, Cali- 0772-B, p. B47–B50. Patrick, B. E., and Lieberman, J. E., 1988, Thermal overprint on fornia: Geological Society of America Bulletin, v. 90, Jayko, A. S., and Blake, M. C., 1989, Deformation of the eastern the Seward Peninsula blueschist terrane: The Lepontine in p. 327–330. Franciscan Belt, northern California: Journal of Structural Alaska: Geology, v. 16, p. 1100–1103. Thurston, S. P., 1985, Structure, petrology, and metamorphic his- Geology, v. 11, p. 375–390. Philippot, P., 1990, Opposite vergence of nappes and crustal ex- tory of the Nome Group blueschist terrane, Salmon Lake Joyce, J., 1992, Blueschist metamorphism and deformation on the tension in the French-Italian western Alps: Tectonics, v. 9, area, Seward Peninsula, Alaska: Geological Society of Samana Peninsula; a record of subduction and collision in p. 1143–1164. America Bulletin, v. 96, p. 600–617. the Greater Antilles, in Mann, P., ed., Geological Society of Platt, J. P., 1986, Dynamics of orogenic wedges and the uplift of Till, A. B., Dumoulin, J. A., Gamble, B. M., Kaufman, D. S., and America Special Paper 262, p. 47–76. high-pressure metamorphic rocks: Geological Society of Carroll, P. I., 1986, Preliminary geologic map and fossil data, Leake, B. E., compiler, 1978, Nomenclature of amphiboles: Min- America Bulletin, v. 97, p. 1037–1053. Solomon, Bendeleben, and southern Kotzebue quadrangles, eralogical Magazine, v. 42, p. 533–563. Platt, J. P., and Vissers, R. L. M., 1989, Extensional collapse of Seward Peninsula, Alaska: U.S. Geological Survey Open- Lieberman, J. E., 1988, Metamorphic and structural studies of the thickened continental lithosphere: A working hypothesis for File Report 86-276, scale 1:250 000. Kigluaik Mountains, western Alaska [Ph.D. thesis]: Seattle, the Alboran Sea and Gibraltar arc: Geology, v. 17, Tolson, R. B., 1987, Structure and stratigraphy of the Hope Basin, University of Washington, 192 p. p. 540–543. southern Chukchi Sea, Alaska, in Scholl, D. W., Grantz, A., Liou, J. G., Maruyama, S., and Cho, M., 1985, Phase equilibria and Purdy, J. W., and Ja¨ger, E., 1976, K-Ar ages on rock-forming min- and Vedder, J. G., eds., Geology and resource potential of mineral parageneses of metabasites in low-grade metamor- erals from the Central Alps: Memorie degli Instituti di Geo- the continental margin of western North America and ad- phism: Mineralogical Magazine, v. 49, p. 321–333. logia e Mineralogia dell’ Universita´ di Padova, v. 30, p. 1–31. jacent ocean basins: Houston, Texas, Circum-Pacific Coun- Lister, G., Banga, G., and Feenstra, A., 1984, Metamorphic core Ridley, J., 1984, The significance of deformation associated with cil for Energy and Mineral Resources, p. 59–72. complexes of the Cordilleran type in the Cyclades, Aegean blueschist facies metamorphism on the Aegean island of Worrall, D. M., 1991, Tectonic history of the Bering Sea and the Sea, Greece: Geology, v. 12, p. 221–225. Syros, in Dixon, J. E., ed., The geological evolution of the evolution of Tertiary strike-slip basins of the Bering shelf: Little, T. A., Holcombe, R. J., Gibson, G. M., Offler, R., Gans, eastern Mediterranean: Geological Society [London] Spe- Geological Society of America Special Paper 257, 120 p. P. B., and McWilliams, M. O., 1992, Exhumation of late cial Publications, v. 17, p. 545–550. Paleozoic blueschists in Queensland, Australia, by exten- Robinson, P., Spear, F. S., Schumacher, J. C., Laird, J., Klein, C., sional faulting: Geology, v. 20, p. 231–234. Evans, B. W., and Doolen, B. L., 1982, Phase relations of Miller, E. L., Calvert, A. T., and Little, T. A., 1992, Strain-col- metamorphic amphiboles: Natural occurrence and theory, lapsed metamorphic isograds in a sillimanite gneiss dome, in Veblen, D. R., and Ribbe, P. H., ed., Amphiboles: Pe- Seward Peninsula, Alaska: Geology, v. 20, p. 487–490. trology and experimental phase relations: Mineralogical So- MANUSCRIPT RECEIVED BY THE SOCIETY JULY 9, 1993 Moore, T. E., Wallace, W. K., Bird, K. J., Karl, S. M., Mull, C. G., ciety of America, v. 9B, p. 1–227. REVISED MANUSCRIPT RECEIVED OCTOBER 12, 1994 and Dillon, J. T., 1992, Stratigraphy, structure, and geologic Rubie, D. C., 1984, A thermal tectonic model for high-pressure MANUSCRIPT ACCEPTED OCTOBER 18, 1994

Printed in U.S.A.

Geological Society of America Bulletin, May 1995 553