JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 99, NO. B8, PAGES 15,115-15,139, AUGUST 10, 1994

Mid-Cretaceous paleomagnetic results from Marie Byrd Land, West : A test of post-100 Ma relative motion between East and West Antarctica

V. J. DiVenere and D. V. Kent Lamont-Doherty Earth Observatory and Department of Geological Sciences, Columbia University, Palisades New York

I. W. D. Dalziel Institutefor Geophysics,University of Texasat Austin

Abstract. As part of the tripartite,United States- United Kingdom- New Zealand, 1990-1991 SouthPacific Rim InternationalTectonics Expedition, oriented samples were collectedfor palco- magneticanalysis from mid-Cretaceous(circa 100 Ma) intrusiverocks at samplinglocalities across350 km of the Ruppertand Hobbs Coast area of Marie Byrd Land, West Antarctica. Paleomagneticresults are presentedalong with severallines of evidence,including a positivetilt test based on the attitude of circa 117 Ma volcanic rocks that the circa 100 Ma rocks intrude, whichargue that these results are a representativeestimate of the mid-Cretaceousmagnetic field in MarieByrd Land (MBL). Thenew circa 100 Ma meansouth pole (224.1øE / 75.7øS,A95 = 3.8ø, N = 19 sitemeans) is concordantwith otherWest Antarcticresults of similar age implying that at leastMarie Byrd Land, ThurstonIsland and the AntarcticPeninsula have not experiencedany paleomagneticallyresolvable relative motion since the mid-Cretaceous.However, the polesfrom thesePacific-bordering blocks of West Antarcticaare significantlyoffset from a syntheticapparent polar wanderpath that was producedfor the East Antarcticcraton, implying relativemovement betweenEast Antarcticaand Pacific West Antarcticasince about 100 Ma. Thoughthe palcomag- netic estimatefor east-westAntarctic relative motion may be reconciledwith geologicestimates for extensionin the RossSea at the extremesof the errorenvelope, the bestpalcomagnetic esti- mate of relativemotion suggestsa largeramount of total extensionbetween East and West Antarctica(MBL) thanpreviously suspected. Both estimates call for severalhundreds of kilo- metersof post-100Ma displacementbetween East Antarcticaand the Pacific-borderingblocks of West Antarctica.

Introduction to one another(assuming no East-WestAntarctic relative mo-

Relative movements between the East Antarctic craton and tion) [Molnar and Stock, 1987] or that there must be a Cenozoicplate boundarysomewhere between the northPacific West Antarctica have been discussedby many authors. It has and East Antarctica(assuming that the hotspotsare fixed with long been known that East Antarcticais built on Precambrian respect to one another) [Gordon and Cox, 1980; Duncan, basement,and more recently it has becomeapparent that West Antarctica is composedalmost entirely of Phanerozoiccrust 1981]. Marie Byrd Land is the crucial link connecting the which is divided into four major crustalblocks [Dalziel and Pacific plate with East Antarctica and the Atlantic bordering Elliot, 1982]: Marie Byrd Land (MBL), ThurstonIsland (TI), plates. This is becausethe spreadingon the Pacific-Antarctic the Antarctic Peninsula (AP), and the Ellsworth-Whitmore Ridge, which began in the Late Cretaceous,documents relative Mountains (EWM) (Figure 1). The overlap of AP with the motion between the Pacific plate and MBL [Mayes et al., Falklands Plateau in Gondwana reconstructions[e.g., Norton 1990], and any MBL-East Antarctic motion must be accounted and Sclater, 1979] suggeststhat the peninsulamust have expe- for in order to compare Pacific plate motions with the Atlantic rienced some relative motion with respectto East Antarctica bordering plates. sincethe inceptionof Gondwanabreakup in the Jurassic.The Palcomagnetic studies from AP [Longshaw and Griffiths, pattern of thin crust in West Antarctica adjacent to the 1983; Wattseta/., 1984; Kellogg and Rowley, 1989; Grunow, Transantarctic Mountains [Bentley, 1991] also suggestsex- 1993], EWM [Watts and Bramall, 1981; Grunow et al., tension between East and West Antarctica. Finally, studies 1987a], and TI [Grunow et al., 1987b; Grunow eta/., 1991] comparingPacific plate hotspotswith Atlantic hotspotshave have placed constraintson the position and relative motions alternatively concluded that the hotspotsmoved with respect of these blocks with respect to the East Antarctic craton. Middle Jurassic(--175 Ma) palcomagneticpoles from AP and Copyright1994 by the AmericanGeophysical Union. EWM are offset from the well-establishedMiddle Jurassicpole from East Antarctica,supporting the idea of post-175 Ma rela- Papernumber 94JB00807. tive motion between East and West Antarctica (at least AP and 0148-0227/94/94JB-00807505.00 EWM). Grunow et al. [1991] proposedthat this offset was due

15,115 15,116 DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND

Weddell Sea

East

270 E Antarctica

Sea,) ;

180 E

Figure 1. Antarctic location map: AP, Antarctic Peninsula;BC, Borchgrevink Coast; BSB, Byrd subglacial basin; BST, Bentley subglacial trough; C-I, CIROS-1 drill site; CS, Cape Spring; CT, Central Trough; EB, Eastern Basin; EP, Edward VII Peninsula; EWM, Ellsworth-Whitmore Mountains; LC, Lassiter Coast; LI, LivingstonIsland; MBL, Made Byrd Land; MM, Mount Murphy; NVL, north Victoria Land; OC, Orville Coast; PIB, Pine Island Bay; PS, Penola Strait; RHC, Ruppert/Hobbs Coast study area; TI, Thurston Island; VLB, Victoria Land Basin; 270, DSDP Site 270.

to Jurassicopening in the Weddell Sea basin. It has been fur- New paleomagneticresults are presentedfrom west central ther suggested[Grunow,. 1993] that the West Antarctic blocks MBL which confirm a high-latitude mid-Cretaceousposition rotated independentlyof each other between 175 and 110 Ma for MBL adjacent to the East Antarctic craton. These results but thatthere has been little or no motionwith respectto East are concordant with mid-Cretaceous paleomagnetic results Antarcticasince the mid-Cretaceous(-110 Ma). from TI and AP, implying no paleomagnetically resolvable The formerposition of MBL with respectto TI, AP, and relative motions between the Pacific-borderingblocks of West EWM and alsoto EastAntarctica is problematicalon the basis Antarctica(i.e., all of West Antarcticaexcept EWM; see Figure of the two previously publishedpaleomagnetic studies of 1) since the mid-Cretaceous. The West Antarctic mid- Cretaceousrocks from MBL. Scharnbergerand Scharon Cretaceouspaleomagnetic poles are compared with a newly [ 1972]reported results from a smallnumber of sitesin igneous constructedsynthetic apparentpolar wander path (APWP) for rocks with poor age control. They interpretedtheir resultsto EastAntarctica. The polesfor MBL, TI, and AP areoffset from indicate a low paleolatitudefor MBL, discordantwith the rest the East Antarctic APWP, from which it may be concludedthat of Antarcticaduring the Cretaceous,suggesting that MBL was the Pacific West Antarctic blocks have moved with respect to an exoticterrane. This contrastswith resultsfrom Grindley East Antarctica since about 100 Ma. and Oliver [1983], who studiedrhyolitic volcanicsand mafic dikeswarms, several of whichwere radiometrically dated at 90- Geologic Background 110 Ma. Theirmean paleomagnetic pole implied a highpale- olatitudefor MBL, consistentwith a positionadjacent, but The geologichistory of westernMBL beginswith the depo- with some displacementrelative to East Antarctica since the sition of a lower Paleozoic turbidite sequence,the Swanson mid-Cretaceous. Grunow et al. [1991] noted that this result Formation, which was folded and mildly metamorphosedin the would also imply relative motion between MBL and the re- Late Ordovician [Adams, 1986]. The SwansonFormation may mainder of West Antarctica. be correlative with the Group of western South DIVENERE ET AL.: PALEOMAGNETICRESULTS FROM MARIE BYRD LAND 15,117

rioDonald Ruppert Coast c.,. •.• .•o- .•?MILeMn•uri• v,,•o ß amp - ' MtMcCoy •o•. • M• .MtMatikonis • Heights • Wilkbt.= BaseCamp • Mt Pearson• o •MtPrince ß > MtSteinfeld • Cam Fosdick •s ' 0 50 • •o" ' I I •- km

East

Antarctica

76øS

Figure 2. Map of studyarea, west central Marie Byrd Land.

Island, New Zealand, and similar turbiditc sequenceson the younger,mid-Cretaceous Byrd Coast Granites of the Ford Campbell Plateau [Adams, 1986]. Bradshaw et al. [1983], Rangesand mid-Cretaceous granites of EdwardVII Peninsula. while noting that the SwansonFormation had sedimentologi- Previouslypublished K-At ages for granitic rocks on the cal similarities with the Robertson Bay Group of North RuppertCoast [SpOrli and Craddock,1981] anddikes on the Victoria Land bordering on the East Antarctic craton, do not Ruppert-HobbsCoast [GrindIcy and Oliver, 1983] cluster consider them to necessarilybe the same unit. In the Ford around 100 Ma. Rangesof western MBL (Figure 2), the SwansonFormation Alkali granitesand syenitesare well representedin the was intruded by the subduction-related,I-type, Devonian- Ruppert-HobbsCoast (Figure 2). S7cuitesare foundthrough- Carboniferous Ford Granodiorite, which may be correlative out the Ickes Mountains and the McDonald Heights area. with similar Devonian igneousrocks in North Victoria Land Alkali gabbros are found at Cape Burks and McDonald (East Antarctica), Tasmania and Victoria (), and South Heights. The units have been dated by the Rb-Sr (R. J. Island (New Zealand) [Adams, 1987; GrindIcy and Davey, Pankhurst,personal communication, 1993) and zircon U-Pb 1982; Mason and Taylor, 1987]. Both the Swanson and 4øAr/39Ar[Palais et al., 1993; D. G. Palaisand S. B. Formation and Ford Granodiorite are intruded by Upper Mukasa, personal communication, 1993] methods. Jurassic/LowerCretaceous, I-type, Byrd Coast Granite in the PreliminaryRb-Sr resultsfor the McDonaldHeights and Ickes Ford Ranges,where a younger,mid-Cretaceous pulse of mag- Mountainssyenites yield ages ranging from 95 to 99 Ma, matism is also detected[Adams, 1987]. Mid-Cretaceousgran- while zircon U-Pb agesrange from 100 to 105 Ma, clustering ites from Edward VII Peninsula of westernmost MBL and mid- at 102 _+2 Ma. PreliminaryRb-Sr resultsand zirconU-Pb re- Cretaceous alkali granites and syenites of the Ruppert and sultsfor the peralkalinegranite at Wilkins Nunatakalso give Hobbs Coast of west central Marie Byrd Land record A-type, anage of about100 Ma. Two4øAr/39Ar results from biotite and rift related magmatism associatedwith the rifting of New hornblendeseparates from the alkalineg abbro at CapeBurks Zealand/CampbellPlateau from the MBL margin [SpOrli and yield nearlyidentical plateau ages of 99 + 1 Ma. Preliminary Craddock, 1981; Weaver et al., 1992; Palais et al., 1993]. An Rb-Sr agesfor dike rocksat Mount Petrasare about100 Ma. erosional surface is overlain by upper Tertiary alkaline vol- The circa 100 Ma syenitesand alkali granitesin the Ickes canic rocks [LeMasurier and Rex, 1983; LeMasurier and Rex, Mountains intrude bedded volcanic rocks with a preliminary 1991]. The Tertiary volcanic rocks are interpreted to be re- Rb-Sr age of about 117 Ma. The syenitesat McDonald lated to a mantle plume [Hole and LeMasurier, 1994]. Heightsintrude granodiorite with Rb-Sr and zircon U-Pb ages Most previous geochronologicstudies in the have also of about 117 Ma. been concentratedin the Ford Rangesof westernMBL, includ- ing dating of the SwansonFormation, Ford Granodiorite, and Samplingand Techniques CretaceousByrd CoastGranite [Boudetteet al., 1966; Halpern, 1968; Wade and Wilbanks, 1972; Halpern, 1972; Adams, As part of the palcomagneticsampling during the 1990- 1986; Adams, 1987]. Sparsegeochronological work in cen- 1991 South Pacific Rim International Tectonics Expedition tral MBL (Ruppert-Hobbs Coast area) has suggestedthat the (SPRITE)field seasonin Marie Byrd Land,204 orientedcores granitic rocks there may be correlative in age with the and 11 orientedhand samples were collectedfrom 35 sitesin •5,• •8 DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND circa 100 Ma syenite,peralkaline granite, gabbro,and mafic very strongly magnetized sites in gabbro at Cape Burks were dikes in the Ickes Mountains,McDonald Heights, and Mount significantdifferences observed between the magnetic and Sun Petrasareas (Figure 2). compassdirections. At thesesites the Sun compassreadings The cores were oriented with a magnetic compassand were used. checkedwith a Sun compassat all siteswhere the outcropwas All sampleswere subjectedto either progressivethermal not in shadow. Magnetic declination was calculatedfor each demagnetizationup to 675 øC, or progressivealternating field site from the International Geomagnetic Reference Field (AF) demagnetization to maximum fields of 100 mT. [Langel, 1992] and appropriatecorrections were appliedto the Proceduresand samplestorage were performedin a magneti- field data. Differencesbetween the magneticand Suncompass cally shieldedroom with a nominal ambient field of -250 nT. were lessthan 2ø to 3ø for samplesfrom most sites. In these Thermal and alternatingfield demagnetizersare housedin their casesthe magnetic compassreadings were used for the core own mu-metal shieldsreducing the field in these to approxi- orientations,since they were always available. Only at three mately 5 nT. Measurements were performed in a

N/Up W/Up N/Up NRM NRM 3OO 340 400

5mT (e) Sample2.7

N/Up 15 0.1Nmf (•300 • 500 40 25 / ½•oø 565

WF--

575 ,;0 NRM •

s N 100 øC 0.02Nm 2 3 S/Dn E/Dn

(a) Sample1.1 (]•) Sample3.6 W I I E O.OlNm S/Dn N N Mt Sinha Syenite MB 161

1

w J/Jo •.8 Thermal

o o Treatment 600øC (c•) 100mT (e) S S

Figure 3. (a, b) Orthogonalplots of demagnetizationdata from Mount Sinha syenite,with open symbols representingthe verticaltrace and solid symbolsthe horizontaltrace of the magnetizationvector. (c) Sample which was rejected with inset showing high temperaturesegment enlarged. (d) Remanent intensity (normalizedto initial intensity)versus sample treatment. (e) Equal-areaprojection of sampleNRM directions and ChRM componentdirections, with open symbolsplotting on the upper hemisphere;triangle represents the time-averageddipole field directionfor this location. DIVENERE ET AL.: PALEOMAGNETIC RESLILTS FROM MARIE BYRD LAND 15,119

SuperconductingTechnologies two-axis cryogenic magne- orthogonalplots. The ChRM directionsuniformly have steep tometer,or a Digicospinner magnetometer for verystrongly negativeinclination, indicating normal polarity. Maximum magnetizedsamples. blockingtemperatures around 580øC and moderate coercivities Componentdirections were determined via principalcom- indicatemagnetite as the principalremanence carrying min- ponentanalysis [Kirschvink, 1980] of linearsegments cho- eral. This is consistentwith reflected light microscopicin- sen by eye from orthogonalplots [Zijderveld, 1967]. spectionof polishedsections of representativesyenite and Componentsfor approximatelyone-third of the samples, gabbrosamples from McDonaldHeights which showed ig- whichhad a greaterdegree of noise,were anchored to theori- neousmagnetite crystals with [111] ilmenite exsolution gin. The principalcomponent directions were plotted on lamellae or coarse magnetite-ilmeniteintergrowths, with equal-areanets and mean directions were determined by stan- lessermartitc. Thermal and AF demagnetizationyielded simi- dard Fisher statistics[Fisher, 1953]. Virtual geomagnetic lar directions,but thermal demagnetizationgave cleanerdecay poles(VGPs) were calculated for each site and mean palcomag- trajectories and thereforethe majorityof samplesfrom each neticpoles and their associated confidence limits were calcu- sitewere demagnetized thermally. Bulk magneticsusceptibil- latedfrom these. All testsof significancebetween poles were ity was monitoredduring thermal demagnetization experi- madeusing the formal method of McFaddenand Lowes [1981] mentsand the absenceof major changesin susceptibilityindi- at the 95% confidencelevel (p = 0.05). All confidenceinter- catesthat no significantalteration of magneticmineralogy vals cited in the text are 95% unlessotherwise noted. For the occurredduring laboratory heating. Figures3-8 illustratede- graphicalcomparison of meanpoles from different crustal magnetizationbehavior and detailsare given for each sam- blockswe plot63% circles of confidence(A63) because they pling locale in the followingsections. providea goodvisual estimate of thesimilarity (simple over- lap)or d•fference (no overlap) of meanpoles [e.g., Irving and Results From McDonald Heights Irving, 1982]. Mount Sinha (west) syenite. Nine oriented cores from each of three sites were collected in variably weathered PaleomagneticResults syenitefrom a nunatakjust westof MountSinha (Figure 3). The circa 100 Ma rocksin this studygenerally display simi- The averagenatural remanent magnetization (NRM) intensity lar demagnetizationbehavior. Variable, low-stability magne- is 0.1 Aim and the directionsare scatteredwith negativeincli- tizationsare removed during AF andthermal demagnetization nations (Figure 3e). A stableChRM was isolatedin four of by 15 to 30 mT or 250ø to 500øC.Above these treatment lev- ninesamples from site 1, only one samplefrom site2, andall els, mostsamples have a stablecharacteristic remanent mag- ninesamples from site3 (Figures3a and3b). The stablesam- netization(ChRM) evident by lineardecay to the originon the ples have a discreteunblocking temperature range between

al

w/up w/up NRM NRM J/Jo 200 øC 5 Nm 4OO ß 2.7AF , - • 515 600 øC 0 Treatment 100rnT (c) 10mT 0.1 Nm 55O 20

35 55 565 75 w •oMB 158 575 100 585 N 600øC t N E/Dn E/Dn (a) Sample1.5t (]•) Sample2.7 (a) S Figure4. (a, b) Orthogonalplots of demagnetizationdata from Mount Sinha gabbro. (c) Intensityversus sampletreatment. (d) Equal-areaprojection of ChRMcomponent directions; all symbolsdefined as in Figure3. Table1. SiteMean Directions and South Paleomagnetic Poles

Directions Pole

In Situ Tilt Corrected Site n/N k %5 D I D I PlonøE PlatøN A95

Mount SinhaSyenite, McDonald Heights, Longitude/Latitude: 223.80/-75.07 In Situ Pole MB161.1 4/9 268 5.6 236.3 -85.0 246.2 -68.0 MB161.2 a 1/9 23.2 -86.2 203.3 -82.4 MB161.3 9/9 108 5.0 327.7 -84.9 264.8 -81.7 Mean 2/3 282.6 -86.5 251.3 -75.0

MountSinha Gabbro, McDonald Heights, Longitude/Latitude: 223.85/-75.07 In Situ Pole MB158.1 7/7 560 2.6 329.2 -85.1 261.7 -81.8 MB158.2 6/7 241 4.3 304.7 -86.9 247.8 -77.5 MB 158.3 7/7 409 3.0 206.8 -85.1 234.7 -65.9 Mean 3/3 336 6.7 282.4 -87.5 243.4 -75.3 13.5

PedenCliffs Syenite, McDonald Heights, Longitude/Latitude: 223.48/-74.95 Tilt Corrected Pole MB160.1 3/6 253 7.8 108.9 -86.4 201.7 -71.3 MB 160.2 4/7 89 9.7 92.6 -83.9 184.9 -70.3 MB160.3 4/7 55 12.5 75.5 -82.5 170.9 -71.8 MB 160.4 5/7 78 8.7 6.4 -86.2 217.2 -82.3 MB160.5 9/11 323 2.9 56.3 -87.8 207.4 -76.9 MB160.6 5/5 125 6.9 76.8 -84.3 181.6 -73.5 Mean 6/6 656 2.6 74.7 -85.8 190.7 -74.8 5.2 Acomp 38 30 4.3 29.5 -66.0 85.4 -60.6

Mount VanceSyenite, Ickes Mountains, Longitude/Latitude: 220.60/-75.40 Tilt Corrected Pole MB223.1 7/7 68 7.4 1.3 -73.3 8.6 -88.2 217.8 -78.9 MB223.2 7/7 210 4.2 348.5 -74.0 283.4 -86.7 247.2 -75.5 MB223.3 7/7 224 4.0 348.0 -72.0 306.1 -85.4 259.8 -78.2 Mean 3/3 1133 3.7 352.5 -73.2 308.3 -87.2 242.6 -78.1 7.2

Mount LangwaySyenite, Ickes Mountains, Longitude/Latitude: 220.22/-75.48 Tilt Corrected Pole MB305.1 3/3 283 7.3 336.4 -76.0 250.9 -84.0 252.3 -68.5 MB305.2 7/7 312 3.4 353.2 -76.4 230.6 -87.8 231.4 -72.4 MB305.3 9/9 302 3.0 344.8 -78.1 223.6 -85.4 237.3 -67.9 Mean 3/3 1299 3.4 344.7 -76.9 237.6 -85.8 240.8 -69.8 6.8

Mount Petras Intrusions,Longitude/Latitude: 231.40/-75.91 In Situ Pole MB224 t' 3/3a 325 6.8 149.4 -84.3 217.5 -65.5 MB 225" 3/3 a 183 9.1 57.2 -86.5 202.4 -78.2 Mean 2/2 117.2 -86.7 212.5 -72.0

WilkinsNunatak Granite, Longitude/Latitude: 220.05/-75.65 In Situ Pole MB302.123 3/3`/ 47 18.2 342.5 -81.9 317.5 -85.1 MB302.4 5/6 584 3.2 349.3 -80.6 358.0 -85.0 MB302.5 5/7 356 4.! 336.9 -79.1 342.5 -80.4 MB302.6 5/6 379 3.9 329.6 -81.1 315.1 -81.3 MB302.7 5/6 175 5.8 343.3 -81.8 324.1 -85.3 MB302.8 2/2`/ 343.3 -80.4 344.7 -83.8 Mean 6/6 2939 1.2 340.7 -80.9 333.4 -83.7 2.3

Cape Burks Gabbro,Longitude/Latitude: 223.13/-74.77 In Situ Pole MB202.1 6/6 82 7.5 142.3 -76.0 234.0 -33.3 197.8 -50.4 MB202.2 7/7 244 3.9 171.6 -74.8 233.3 -25.9 217.3 -46.4 MB202.3 7/7 152 4.9 152.2 -73.9 231.3 -30.9 203.5 -46.0 MB202.4 5/5 298 4.4 168.7 -75.1 233.5 -26.7 215.4 -47.0 MB202.5 6/6 948 2.2 174.9 -73.9 232.9 -24.7 219.5 -44.8 MB202.6 5/5 712 2.9 171.3 -76.2 234.8 -26.4 217.3 -48.7 MB202.7 5/5 209 5.3 194.3 -61.7 173.1 -31.7 235.0 -28.0 MB202.9 7/7 386 3.1 174.6 -55.5 164.9 -22.3 218.5 -20.9 Mean 8/9 76 6.4 171.7 -71.5 216.5 -42.1 9.4 Circa100 Ma MeanPole (Vance, Langway, Sinha syenite and gabbro, Peden Cliffs, and MountPetras) 19/20 79 224.1 -75.7 3.8 Abbreviations:n/N, numberof samplesaccepted / number collected for sitemeans or number of sitesaccepted over number of sitescollected for groupmeans; k, Fisher'sprecision parame- ter;a95 , radius of 95% coneof confidenceabout mean direction; directions, D, meandeclination; L meaninclination; pole, Plon, mean paleomagnetic pole east longitude; Plat, mean paleomag- neticpole north latitude (negative is southlatitude); A95 , radius of 95%cone of confidenceabout mean pole. a Onlyone stable sample; site not used in furthercalculations. c Porphyry. b MaficDike. a Handsamples. DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND 15,121 about 550ø and 580øC over which the remaining magnetiza- bility component, with a mean direction calculated at D = tion decaysto zero (samples 1.8 and 3.6 in Figure 3d). The 29.5,I = -66.0,and a95 = 4.3ø, is significantlydifferent from remainderof the samplesdid not yield stabledecay trajectories the high-stability component, axial dipole field direction and and only a small portion of the NRM remainedby 300øC present-dayfield direction. Given its low stability and dissim- (sample2.7 in Figures3c and 3d). The stableChRMs are well ilarity to other directions, the low-stability component is grouped(Figure 3e). likely to have been acquiredduring samplecollection or prepa- Mount Sinha gabbro. Seven oriented cores were col- ration. The groupingis probably due to the fact that all sam- lected from each of three sites in a gabbroicmember near the ples were collected along a single cliff face and therefore have contact with the intruded Lower Cretaceousgranodiorites on very similar core orientations.The ChRM remainingafter the the north flank of Mount Sinha (Figure 4). NRM intensity is removal of the low-stability magnetizationis well grouped strong,with an averageof 4 A/m. NRM directionsare nearly about an easterly and nearly vertical, direction, similar to the vertical upwardand well grouped. Alternatingfield demagneti- Mount Sinha ChRM directions(Figure 5d). The mean direc- zation shows a hard magnetization which is usually not com- tion calculatedfrom the six site mean directions(D = 74.7, I = pletely demagnetizedby 100 mT, while thermal demagnetiza- -85.8,a95 = 2.6ø) (Table1) is significantlydifferent from both tion shows that the NRM is little affected below about 500 øC the axial dipole field (D = 0, I =-82.3) and the present-day (Figures 4a, 4b, and 4c). The ChRM directions are well field (D = 71.8, I = -75.2). grouped(Figure 4d). The mean direction calculated from the two stable Mount Paleomagnetic Results From the Ickes Mountains Sinha syenite site means (declination(D) = 282.6, inclination (/) =-86.5, N=2) is nearly identical to the mean directioncal- Mount Vance syenite. Seven oriented cores from each culatedfrom the threegabbro site means (D =282.4, I = -87.5, of three sites were collected from the syenite at the base of a95- 6.7ø). Both.,aresignificantly different from the time-av- Mount Vance in the Ickes Mountains(Figure 6). The syenite eragedaxial dipole field (D = 0, I =-82.4) and the present-day intrudes gently dipping, Lower Cretaceousbedded volcanic field (D = 71.9, D = -75.1). See Table 1. rocks. The averageNRM intensityis 0.3 Adm. The ChRM is Peden Cliffs syenite. A total of 43 oriented cores was usually not completelydemagnetized by 100 mT (Figures 6a collected from six sites in syenite and microdiorite inclusions and 6c). Maximum unblockingtemperatures (Figure 6b and at PedenCliffs (Figure 5). The averageNRM intensityis 0.1 6c) and Curie temperaturesappear to be slightlyelevated (585 ø A/m. A consistentbreak is seenin the demagnetizationtrajec- to 600øC)above those expected for puremagnetite (about 575 ø tory of most samplesat around10 to 20 mT during AF treat- to 580øC),suggesting a non stoichiometricmagnetite or pos- ment and around 300øC during thermal demagnetization sibly a low titanium ilmenohematiterather than pure mag- (Figures5a and 5b). The lower stabilitycomponent is north- netite as the remanencecarrying mineral. easterlywith intermediateto steep inclination. The low-sta- Mount Langway syenite. Nineteen oriented cores

1

W_/U NRM W_/_Up NRM • J/Jo

100 øC 0.1Nm-- 0.1 Nm - ß o 0 Treatment 600 øC 100 mT (c) N

200 15 20 35 70 45 4001•300 60 55 550 IN w

575 600 N

E/Dn E/Dn (a) Sample 2.4 (b) Sample 5.9 (d) S Figure 5. (a, b) Orthogonalplots of demagnetizationdata from PedenCliffs syenite. (c) Intensityversus treatment.(d) Equal-areaprojection of ChRM directions;all symbolsdefined as in Figure3. 15,122 DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND

w/up W/Up NRM 1 NRM -

0.2 Nm J/Jo

3.1AF• • 0 , •

(c) 0 Treatment 100600 mTøC 200 øC N 0.02 Nm

10 mT 300

2O 4OO 35 500 525 55 550 W E 75 565 100 575 585 580 595 Sl• N N E/Dn E/Dn

(a) Sample 3.1 (b) Sample 3.3t (d) S Figure 6. (a, b) Orthogonal plots of demagnetizationdata from Mount Vance syenite. (c) Intensity versus sample treatment. (d) Equal-area projection of in situ ChRM componentdirections; all symbols defined as in Figure 3.

1

w/up W/Up -- j• NRMJ/Jo .•100øC

mT .•200 0 0 Treatment 600 øC 15 (c) 100mT .• 400 N 30 0.2Nm 540 0.2Nrn 50

75

W E 590• 580 s N E/Dn E/Dn

(a) Sample2.3 (b) Sample2.6 (d) S Figure7. (a, b) Orthogonalplots of demagnetizationdatafrom Mount Langway syenite. (c) Intensityver- sussample treatment (d) Equal-areaprojection of in situChRM component directions; all symbolsdefined as in Figure 3. DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND 15,123

were collected at three sites from the syenite at Mount fromMount Vance and Mount Langway which have northwest- Langway, 5 km from Mount Vance (Figure 7). The average erlydeclinations and steep negative inclinations (a mean di- NRM intensity is 0.2 A/m. Demagnetizationtrajectories are rection calculated from their results is D = 327.7, i = -73.0, N very similar to thosefrom Mount Vance but maximum block- = 2). ing temperaturesare about 580 øC (Figures 7a, 7b, and 7c) suggestingmagnetite as the remanence carrying mineral. Other Circa 100 Ma Palcomagnetic Results The in situ mean direction calculated from the three Mount Mount Petras intrusions. At Mount Petras, 250 km Vancesyenite site mean directions (D = 352.5,I = -73.2,a95 = SE of McDonald Heights, three oriented hand samples were 3.7ø) is not significantlydifferent from the in situ mean direc- taken from a mafic dike intrudingan intrusiveporphyry on the tion from the three Mount Langway syenite site means (D = southwest spur (MB224), and another three were collected 344.7,I = -76.9,a95 = 3.4ø). Theyare significantly different from the porphyry on the southeastspur (MB225). Two or from both the axial dipole (D = 0, I = -82.6) and the present- three specimensper block were subjectedto AF demagnetiza- day field (D = 75.3, I = -75.7). Theseresults are broadly simi- tion and one specimenper block to thermal demagnetization lar to palcomagneticresults of GrindIcy and Oliver [1983] (Figures 8a- 8d). All specimensyielded a stable ChRM

W/Up N/Up W/Up W/Up NRM ,-, NRM 6- 7.5 mT

5mT NRM/ 475 I 10 5A/m

o 0.02 A/m 20

• 35

/ ( 55

25 75 100

s 45 _ 575a,5__•.• S• N E/Dn E/Dn S/Dn E/Dn

(a) MB 224.1.4 (]3) MB 225.2.2 (e) MB 302.6.3 (h) MB 202.5.4

1' _ •_ _- 202.3.3Thermal 12• al

o Treatment 6oooc 0 Treatment 600 oc 0 Treatment 6oooc 100 mT 100 mT 100 mT (c) (f) 0) N N N

W E W E W E

(a) S (g) S (5) S

Figure 8. Demagnetizationdata from (a) - (d) Mount Petras intrusions,(e) - (g) Wilkins Nunatak granite, and (h) - (j) Cape Burks gabbro;all symbolsdefined as in Figure 3. 15,124 DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND

(Figures 8a and 8b). The mean direction is D = 117.2, I = 9: strike/dip= 240ø / 35ø W) definesa plungingsynform; the -86.7,for N = 2 sitemeans (D = 117.2,I = -86.7,a95 = 5.0ø for palcomagneticdirections from the two limbs are in reasonably N = 6 block means). Grindley and Oliver [1983] reportedre- good agreementbefore correctingfor this apparentstructure sults from six sites (one or two blocks per site) from rhyo- and divergeas a resultof returningthe limbs to the horizontal. dacite flows at Mount Petrasciting K-At plagioclaseand horn- Given the very strongmagnetization and high stabilityin this blendeages ranging from 81 to 104 Ma. A meandirection cal- gabbroit would seemthat the magnetizationis not likely a re- culated from their 6 site mean directions(D = 214.0, I = -75.0, cent overprint. Therefore the synform may be the original a95= 8ø) is significantlydifferent from our Mount Petras intru- layering attitude due to cooling on the walls of the magma sions result. chamber or an early collapse feature due to withdrawal of Wilkins Nunatak granite. Twenty-five oriented cores magma from an underlyingmagma chamberbefore cooling and 5 orientedhand sampleswere taken from six sitesin per- throughthe Curie temperature(~580øC). The in situmean di- alkaline granite at Wilkins nunatak, approximately 15 km rectionis D = 171.7,I = -71.5,a9s = 6.4ø, for N = 8 sitemeans. southof the Ickes Mountains(Figures 8e - 8g). The granitein- The presentresult is similar to, but groupsbetter than, a mean trudes folded and metamorphosedLower Cretaceousvolcanic direction calculatedfrom the Cape Burks resultsof Grindley rocks. The meandirection (D = 340.7,I = -80.9,a95 = 1.2ø) andOliver [1983] (D = 161.5,I = -79.8,a95 = 18ø, N = 3 sites). (Table 1) is similar to the in situ mean direction from the Ickes Mountains syenite. Mean 100 Ma PalcomagneticPole Position Cape Burks gabbro. Sixty-one oriented cores were col- for Marie Byrd Land lected from nine sites in a layered gabbroiccomplex at Cape Burks, on the coast approximately25 km north of McDonald The palcomagneticresults presentedhere are expectedto Heights (Figures 8h - 8j). The sampleswere quite strongly representcirca 100 Ma magnetizationsbecause of (1) the high magnetized,with NRMs rangingfrom 2 to 144 A/m, and yield stability of the characteristicmagnetization to both alternat- very stableChRMs. Sun compassreadings were takenfor each ing field and thermal treatmentin most samples;(2) the con- sampleat eight of the sites. Moderate to large deviationswere cordanceof K-Ar [SpOrli and Craddock,1981; GrindIcy and seen in the magnetic compassreadings so the Sun compass Oliver,1983] and 4øAr/39Ar [Palais et al., 1993]ages to Rb-Sr core azimuths were used for all samples. The resultsfrom one [R. J. Pankhurst, personal communication,1993] and zircon very stronglymagnetized site had to be discardedbecause no U-Pb [Palais et al., 1993; D. G. Palais and S.B. Mukasa,per- Sun compassreadings were taken due to samplingin shadow. sonal communication, 1993] ages, implying a short em- The layering(sites 1 to 6: strike/dip= 340ø / 60øE,sites 7 to placementhistory at about 100 Ma and no younger thermal

LS

PC PC

180 E 180 E (a) (b)

Figure 9. Circa 100 Ma mean poles from Mount Sinha gabbro(SG), Sinha west nunataksyenite (SS), Peden Cliffs syenite(PC), Ickes Mountainssyenites at Mount Vance and Mount Langway(VS and LS), Mount Petras (MP), WilkinsNunatak granite (WG), andCape Burks gabbro (CB) plottedwith A9s circles of confidence.(a) Before and (b) after tilt correctionof Mount Vance and Langway syenites. This and following figures are or- thographic projections. DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND 15,125 events;and (3) the uniformnormal polarity consistent with IckesMountains syenite results using the averagebedding at- magnetizationacquisition and emplacement ages during the tiradein the overlyingvolcanics (strike/dip = 270ø / 15ø N) CretaceousNormal Superchron(118 to 84 Ma) [Kent and (assumingthe volcanicswere flat-lying when the syenitesac- Gradstein, 1986]. quiredtheir magnetization) yields tilt-corrected mean poles for Meanpaleomagnetic poles were calculated for eachof the Mount Vance (VS) and Mount Langway(LS) that fall in the eightcirca 100 Ma locationsby averagingthe site mean vir- clusterwith the McDonald Heightsand Mount Petrasresults tualgeomagnetic poles (VGP). A clusterof polesfalls near the (Figure9b). This agreementgives us confidencethat the as- coastof MBL includingpoles from PedenCliffs (PC), Mount sumptionof originallyhorizontal volcanic bedding is reason- Sinhasyenite (SS), Sinha gabbro (SG), and Mount Petras (MP) able in this case. The consistentpaleomagnetic directions (Figure9a). Thepoles from Cape Burks gabbro (CB), Wilkins observedacross the relativelylarge areaof McDonaldHeights Nunatakgranite (WG), and the in situpoles from the syenite at (PedenCliffs, Mount Sinha,Sinha west Nunatak) implies no MountVance (VS) andMount Langway (LS), however,fall discernibledifferential block tilting within this area. The some 30 ø to either side of the cluster. consistencyof the meanpole from Mount Petras with the tilt- Thisspread of circa100 Ma polesdemands some explana- corrected Ickes Mountains results and the McDonald Heights tion. Sincethe sampledunits are all plutonicand therefore resultsimplies no relativetilts overan areaof approximately cooledslowly, the patternis unlikelyto representpaleosecu- 300 krn parallelto the coastand 150 krn inland,other than lar variationof the Earth'smagnetic field. Also, the angular that observedin the Ickes Mountains. The tilt test, performed distance(-65 ø) betweenthe CapeBurks and Mount Vance by combiningthe locationmean poles from MountVance, syenitein situpoles is muchtoo great to be accountedfor by MountLangway, Peden Cliffs, Sinhagabbro, Sinha syenite, apparentpolar wander given the very similar ages of therock and Mount Petrasbefore and after correctingfor tilt at Vance units. Thereforegiven that New Zealand/CampbellPlateau andLangway, is statisticallysignificant at the 95% (as well as riftedfrom this margin shortly before 85 Ma [Mayeset aL, 99%) confidencelevel. 1990], a structuralexplanation appears likely. The similarity of the mean pole from Wilkins Nunatak to Structural control is difficult to achieve in the the meanpoles from Mount Vance and Langwayimplies tilt- Ruppert/HobbsCoast area, which is characterizedby isolated ing of Wilkins nunatak in the same sense as the Ickes igneousoutcrops separated by expansesof snowand ice. Mountains,and the close proximity of Wilkins Nunatak (15 However,in the Ickes Mountainsthe syeniteintrudes Lower km to the south)to the demonstrablytilted Mount Vance and Cretaceous bedded volcanics which outcrop at Mount Mount Langwaymakes this a reasonableassumption. The po- Vance/MountLeMasurier. Applyinga tilt correctionto the sitionof the mean pole from Cape Burks, on the coastapprox-

N=19MeanPole ß

Peden Cliffs Syenite Sinha Syenite Sinha Gabbro Vance Syenite LangwaySyenite Mt Petras

60 S

30 S I 180 E

Figure 10. The 100Ma meanpole for MarieByrd Land with A95 circle of confidence,plotted with N--19 site means. 15,126 DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND imately 25 km to the northof McDonaldHeights, implies 10ø son for the difference is somewhat puzzling. The intrusive to 15ø of tilting in the oppositesense as that in the Ickes rocks sampled at Mount Petras in the present study yield a Mountains, but again, about an east-weststrike. crystallization age, by the Rb-Sr method, of about 100 Ma, Given the consistencyof the majorityof the poles(PC, SS, and bear an alkaline geochemistrysimilar to the alkaline ig- SG, MP, VS tc, LS tc) overa largearea of MBL, with a positive neous rocks on the Ruppert/Hobbs Coast (S. D. Weaver, per- tilt test,we believethat theseresults should give a goodesti- sonal communication, 1993). The "1ow-dipping...viscous mate of the 100-Ma pole position for MBL. A circa 100 Ma rhyodaciteflows" from Mount Petrasstudied by Grindley and mean southpaleomagnetic pole for MBL was calculatedby Oliver [1983, p. 573] gave discordantK-At agesranging from combiningthe 19 site mean poles from McDonald Heights 81 to 104 Ma, but thesemay be minimum ages not reflecting (Peden Cliffs syenite, Sinha syenite, Sinha west gabbro), the eraplacementand magnetizationage. They proposedcalc- Mount Petras,and the tilt correctedIckes Mountainssyenites alkaline geochemical affinities for these rocks which would (Mount Vance syenite, Mount Langway syenite) (Figure 10 suggesta relationshipto Lower Cretaceousor older rock suites andTable 1). The CapeBurks andWilkins Nunatakresults are and not the circa 100 Ma mid-Cretaceous alkaline suite. We not includedbecause their deviationfrom the main groupof note that the exclusion of our Mount Petras results would VGPs is inferredto be the resultof tiltingbut thereare no pa- hardly affect our mean pole position, while the removal of leohorizontal markers with which to make tilt corrections for Grindley and Oliver's Mount Petras results would make their these. It should be noted, however, that inclusion of both the mean pole not significantly different from our mean. The Cape Burks and Wilkins NunatakVGPs would not signifi- Cape Burks, Ickes Mountains, and Mount Petras site means of canfly alter the mean pole sincethese results fall on either side Grindley and Oliver [1983] accountedfor 10 of the 26 site of the mean of the other 19 sites,and would carry approxi- meansthat went into their overall mean pole. On the basisof mately equal weight (eight sites versus six sites, respec- the good age control for our sampled units, positive tilt test, tively). and good agreement across a large area of MBL, we believe The circa 100 Ma pole for MBL determinedin the present that our resultsrepresent an improvedestimate of the circa 100 study,224.1øE / 75.7øS,A05 = 3.8ø, is significantlydifferent Ma paleomagneticpole position for MBL located at 224.1øE/ from the MBL pole of Grindley and Oliver [1983] at 242øE/ 75.7øS,A0• = 3.8ø. 65øS,A05 = 8.8ø (Table 2, GO MBL 100in Figure11). Thean- gular separationis 12.2ø. Their mean pole was basedon re- sults from volcanic flows from Mount Petras, as well as a vari- Testing the Integrity of Mid-Cretaceous West Antarctica ety of dikesover muchof the samearea as the presentstudy. They cited K-At ages from flows from the Mount Petrasarea Marie Byrd Land's immediateWest Antarctic neighborsin and individualdikes from half of theirremaining sampling lo- the mid-Cretaceous included Thurston Island (TI) to the east cationsranging from approximately90 to 110 Ma. Tilting of andNew Zealand/CampbellPlateau to the north. The boundary coastalfault blockshas been identified in the presentstudy in between MBL and the TI block is a deep glacially eroded the IckesMountains area and inferred for CapeBurks. Tilting trough in Pine Island Bay, which may also be related to the presumably also affected the dikes sampled in the Ickes Udintsev Fracture Zone [Sandwell and McAdoo, 1988; SPRITE Mountains and at Cape Burks by Grindley and Oliver, even Group and Boyer, 1992; Lawver et al., 1993]. South New though they made no tilt corrections. The present Mount Zealand (SNZ) (including the Campbell Plateau and Chatham Petras results are consistent with the other circa 100 Ma re- Rise) is separatedfrom MBL by crust generatedduring Late sults that make up the new 100 Ma mean pole for MBL but Cretaceous to Recent seafloor spreading on the thereis a significantdiscrepancy between these results and the Pacific/Antarctic ridge. What follows is an evaluation of the Mount Petrasresults of Grindley and Oliver [1983]. The tea- new circa 100 Ma pole for MBL with respectto existingpale-

Table 2. WestAntarctic Paleomagnetic Poles

Block Age N A95 PlonøE PlatøN Source

MBL 100 19 3.8 224.1 -75.7 this study MBL 100 26 8.8 242 -65 Grindleyand Oliver [1983] TI 110 7 7.6 210 -73 Grunow et al. [ 1991] AP 107 13 6.9 199 -74 Grunow [ 1993] AP 106 16 10.3 208.5 -74.8 recalculatedin this study from data of Kelloggand Rowley [ 1989] andKellogg [ 1980] AP 106M 29 6.2 203.8 -74.3 calculatedin thisstudy from site means of Kellogg[1980], Kelloggand Rowley [1989], and Grunow [1993] AP 104 7 12.3 132 -77 Watts et al. [1984] AP 102 18 11.5 229 -87 Kellogg and Reynolds[1978] AP 98 5 8.8 117 -86 Valencioeta/. [1979] NZ 94 46 7.0 208 -75 Oliveret al. [1979](174/-52 in NZ coordinates) CrustalBlocks: MBL, Made Byrd Land; TI, ThurstonIsland; AP, Antarctic Peninsula; NZ, New Zealand(reconstructed backto Made Byrd Land after Mayes et al. [1990]). Age in mega annum; N, num- berof sitesused to calculate mean; A05, radius of the 95% cone of confidence about the mean pole; Plon andPlat are mean paleomagnetic poleeast longitude and north latitude (negative is south latitude). DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND 15,127

Grindleyand Oliver (1983) site means Mt Petras ß x Ickes Mts Cape Burks • -I- Others

-!-

60 S

30 S I 180 E

Figure 11. Comparisonof the presentresult (MBL 100) with the previousresult of Grindleyand Oliver [1983](GO MBL). Bothmean poles plotted with A9s circles of confidence.Additional poles are Grindley and Oliver's site means. omagneticdata of similar agefrom TI and SouthNew Zealand Fromthe southern AP, Kellogg [1980] reported paleomag- as well as the remainder of West Antarctica in order to evaluate neticresults from plutonit rocks from the Orville Coast region the tectonic integrity of West Antarctica since the mid- (Figure1), with Rb-Srages ranging from 103 to 109 Ma. The Cretaceous(Figure 12). meanpole (195øE/ 71øS)is in goodagreement with MBL 100, The presentresult (MBL 100 in Figure 12) is within 5' and TI 110, and NZ 94. Kellogg and Rowley [1989] addedthree is not significantlydifferent from the circa 110 Ma polefrom more site meansand new Rb-Sr and K-Ar ages,ranging from ThurstonIsland (TI 110 in Figure12 andTable 2) of Grunowet 100 to 113 Ma. Combiningall of the sitemeans of Kellogg al. [1991]. Grunowet al. [1991]noted that the MBL poleof [1980] and Kellogg and Rowley[1989] givesa meanpaleo- Grindleyand Oliver [1983] requiredrelative motion between magneticpole for thesouthern AP at 208.5øE/ 74.8øS, A9s- MBL and TI since about 100 Ma. However, our resultsshow 10.3' (AP 106 in Figure12 andTable 2), whichgroups very thatMBL andTI havenot experienced any palcomagnetically well with, and is not significantlydifferent from, MBL 100, TI resolvable relative motion since about 100 Ma. 110 and NZ 94. Kellogg and Rowley [1989] had separated Oliveret al. [1979] reporteda paleomagneticpole from theseresults into two groupsbased on comparisonof the de- volcanicflows southeast of the AlpineFault on SouthIsland, clinationsonly, suggestingthat one group was offset from the NewZealand with K-Ar ages ranging from 92 to 98 Ma. Barley other as a result of oroclinal rotation of the southern AP. eta/. [1988]reported an Rb-Srage of 89 Ma fromthese rocks. However,the nearly25 ø paleolatitudinaldifference implied by This pole (NZ 94 in Figure 12 and Table 2), reconstructedto their two groupmeans can not be reconciledby simplevertical MBL usingthe rotationparameters of Mayeset al. [1990],is axis rotation. Bearingin mind that at high latitudes,1 ø of tilt not significantlydifferent from MBL 100 and TI 110 and lends equalsnearly 2 ø of pole shift,it seemsmuch more likely that supportfor a mid-Cretaceouspole positionfor MBL and TI, small tilts, comparable to those identified on the locatedadjacent to the coastof MBL. Ruppert/Hobbs Coast, contribute to the differences observed For the remainderof West Antarctica,there are no post betweenthe two groups. The combinedmean pole calculated Middle Jurassicpaleomagnetic results from EWM. However, above may therefore provide a more valid estimate for the severalrather scattered mid-Cretaceous poles have been pub- southernAP than eitherof the two groupmean poles, since it lished from various AP locations. may more completelyaverage out the effectsof possibletilts. 15,128 DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND

AP 102 (Lassiter) AP98 Cape Spring)

-l- GO AP 104 (Livingston) AP 106 (Orville)

TI 11 (Livingston& Penola)

180 E

Figure 12. Marie Byrd Land and other West Antarctic circa 100 Ma poles: MBL 100, the presentresult; GO MBL, previousresult for MBL of Grindley and Oliver [1983]; TI 110 [Grunow et al., 1991]; AP 106 from the Orville Coast of the southernAntarctic Peninsula [Kellogg and Rowley, 1989] recalculated in this paper; AP 102 from the Lassiter Coast of the south-centralAntarctic Peninsula[Kellogg and Reynolds, 1978]; AP 104 from Livingston Island, northernAntarctic Peninsula[Watts et al., 1984]; AP 107 from the northern Antarctic Peninsula [Grunow, 1993]; AP 98 from Cape Spring of the northern Antarctic Peninsula [Valencio et al., 1979]; NZ 94 from southNew Zealand [Oliver et al., 1979] rotated back to Marie Byrd Land using the pole of rotationof Mayeset al. [1990];selected poles shown with A63 circlesof confidence.

A recent study of mid-Cretaceous intrusive and extrusive for solely by apparentpolar wanderduring this shorttime in- rocks from Livingston Island and Penoh Strait of the northern terval (about4 to 9 m.y.). Also, the AP 102 and AP 98 poles AP [Grunow, 1993] (Figure 1) yields a circa 107 Ma pole are not consistentwith MBL 100, which would imply a com- (199øE/ 74øS,A95 = 6.9ø) (AP 107,Figure 12, Table 2) which plicatedhistory of relativemotions between AP and MBL con- agrees well with MBL 100, TI 110, AP 106, and NZ 94. As trary to the more consistentpicture evidencedby the more re- implied by Grunow [1993], the AP 107 result supersedesthe centlypublished results (AP 107 and AP 106). previous result of Watts et al. [1984] (AP 104) from It is presentlyunclear whether the differencesamong the Livingston Island because 8 of 13 site means are from units circa 100 to 110 Ma paleomagneticpoles from AP discussed that constituted(yet give different directionsthan) 4 of 7 site above are due to oroclinal rotation, localized rotations and/or means of Watts et al. [1984]. tilts, significantdifferences in the n'tagnetizationages of the There are two earlier resultsfor circa 100 Ma rocks from AP, units sampled(greater than thoseapparent from the cited ra- mostly front dikes, front the LassiterCoast (Figure 1) of south- diometric ages), incon'tpleteremoval of overprints,or other centralAP [Kelloggand Reynolds,1978] (AP 102 in Figure12 causes. On the assumptionthat the two most recent studies and Table 2) and Cape Spring (Figure 1) of northern AP [Kellogg and Rowley, 1989; Grunow, 1993] incorporatethe [Valencio et al., 1979] (AP 98). Both results show uniform best data, we calculate a circa 106 Ma mean pole for AP by normal polarity and mean poles coincidentwith the dipole con'tbiningthe site meansfrom Grunow [1993] andKellogg field (geographicpole), althoughthe partial den'tagnetization and Rowley [1989] (which includessite n'teansfrom Kellogg proceduresused should have beenadequate to removerecently [1980]).The aLean pole is 203.8øE/ 74.3øS, A95 = 6.2ø for N = acquiredviscous components. Watts et al. [1984] and Grunow 29 sites (AP 106M), which fits very well with MBL 100, TI [1993] suggestedapparent polar wander as the causefor the 110, and NZ 94. difference between AP 107/AP 106 and AP 102/AP 98, but the It may be concludedfront the correspondenceof MBL 100 angulardistance (14 ø) seemsto be ratherlarge to be accounted with TI 110 that the two blockshave not experiencedany sig- DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND 15,129 nificant motion with respect to one another (or no more than Also, Besse and Courtillot's plate reconstructionsdid not ac- that allowed by the paleomagneticerror) since about 100 Ma. count for deformation within the African plate due to Based on the assumptionsabove regarding the AP poles, it Cretaceousextension in the Benue Trough betweenabout 136 may also be concludedfrom the correspondenceof AP 106M and 84 Ma [Burk and Dewey, 1974;Pindell and Dewey, 1982; with MBL 100 and TI 110 that AP has not experienced any Pindellet al., 1988]. For the purposeof producinga synthetic significant relative motion with respectto MBL and TI since APWP it is important to consider the errors involved in the about 100 Ma. plate tectonicreconstructions used and to transferdata through The agreement of the circa 175 Ma poles from AP independentplate circuits,if possible,in order to avoid a po- [Longshaw and Griffiths, 1983] and EWM [Grunow et al., tentially systematicbias due to errors in a single reconstruc- 1987a] was cited as evidence that AP and EWM may have tion. In this study we constructa Cretaceousto Recentsyn- moved as a single entity since the Middle Jurassic[Grunow et thetic APWP for East Antarcticawith an emphasison the mid- a/., 1987a]. More recently, Grunow [1993] proposedindepen- Cretaceousin order to facilitate a comparisonof the West dent motionsof AP, TI, and EWM during the Late Jurassicand Antarctic mid-Cretaceous results with an East Antarctic refer- Early Cretaceous, based on dissimilar paleomagnetic poles ence. The new path utilizessome recently published poles and from AP and TI, even though there are no younger paleomag- recent plate tectonic reconstructions,incorporates the Benue netic results from EWM or even exposures of post-Middle Trough opening into the reconstruction,and transferspoles Jurassicrocks with which to test either hypothesis. In either through two independentplate circuits (-Antarctica and case, our analysis shows that most of the elements of West Australia-Antarctica).Details of the pole selectionand plate Antarctica(at least MBL, TI and AP) have not experiencedpa- kinematicmodel are given in the Appendix. leomagnetically resolvable differential motion since about The selectedpaleomagnetic reference poles, rotated into the 100 Ma. The mid-Cretaceous mean paleomagnetic pole for East Antarcticreference frame (Table A1, Appendix), lie in a MBL, TI, andAP is locatedat 212.2øE/ 74.5øS,A95 = 4.6ø for fight swath (Figure 13) which suggeststhat the individual N = 3 (MBL 100, AP 106M, TI 110) with an approximatemean poles are each reasonableestimates of the magneticfield and age of 105 Ma. that there are no major problems with the reconstructions. The 95 Ma pole from Australiagroups well with 95 Ma poles Comparison With East Antarctica: Synthetic from Africa and the 100 Ma pole from , and the 112 Ma pole from Australialies betweenthe 112 Ma pole from Apparent Polar Wander Path for East Antarctica Africa and the 116 Ma pole from India, supportingthe validity The mid-Cretaceous paleomagnetic poles from West of the reconstructions.Mean poleswere calculatedat 10 m.y. Antarctica may be compared with an apparent polar wander intervals from 125 to 75 Ma using a 20-m.y.-wide sliding path (APWP) for East Antarctica in order to test the mid- window. The meanpoles (also listedin Table A2, Appendix) Cretaceousto Recent integrity of Antarctica. However, the are labeledaccording to the mean age of the combinedpoles only reliable Mesozoic reference pole available for East rather than the center of the window. Antarctica is the circa 175 Ma pole basedon many studiesof The syntheticAPWP for East Antarctica is comparedin the Ferrar Dolerites, Dufek Intrusion, and Kirkpatrick Basalts Figure 14 with Besseand Courtillot's[1991] masterpath ro- (220.4øE/ 52.7øS,A05 = 4.4ø) [Kellogg,1988]. Thereare no tatedfrom Africa into the EastAntarctic reference frame using paleomagnetic results from rocks of Cretaceous age on the rotation poles interpolated from their plate kinematic model East Antarctic craton although several overprints of suspected [Besse and Courtillot, 1988] (based on work by Segoufin Cretaceous age have been reported from Jurassic Ferrar [1980], Segoufin and Patriat [1980], Patriat [1983], and Dolerite and Kirkpatrick Basalt as well as the Cambro- Patriat et al. [1985]) and usingtheir chosenmagnetic polarity OrdovicianBowers Group in the area of the Rennick Graben in timescale[Harland et al., 1982]. Their meanpoles are also la- North Victoria Land [Mcintosh et al., 1982; Delisle, 1983; beled accordingto mean age rather than window center. Even Delisle and Fromm, 1984; Schmierer and Burroester, 1986; though the data sets and reconstructionmodels are somewhat Delisle and Fromm, 1989]. Inclination-onlydata were - different,the resultantpaths are similar. The principaldiffer- tained from the Kerguelen Plateau on the Antarctic plate for encesare (1) our Early Cretaceouspoles (125 and 117 Ma) are 70-90 Ma [Sakai and Keating, 1991] and 100-115 Ma more westerly than the Early Cretaceouspoles (127 and 121 [lnokuchi and Heider, 1992], but this type of data yields mean Ma) of Besseand Courtillot, partly due to our compensation inclinations which tend to be biased toward shallower inclina- for extension in the Benue Trough in our plate kinematic tions at high site paleolatitudes[McFadden and Reid, 1982], model;(2) their 121, 112, and 98 Ma meanpoles approach the such as in these studies. Therefore, in order to compare the southpole more quickly than our 117, 102, and 93 Ma poles; West Antarctic mid-Cretaceouspaleomagnetic results with an and (3) their Late Cretaceouspoles and Cenozoic poles remain East Antarctic reference,a syntheticAPWP must be produced nearer to the presentgeographic south pole than do ours. by transferring poles from other . The mid-Cretaceousmean poles from MBL, TI, and AP are Irving and Irving [1982] compiledpolar wanderpaths for plotted againstthe East Antarctic APWP in Figure 15. MBL the majorcontinental blocks, excluding Antarctica. Besse and 100, AP 106M, and TI 110 are all offset to the east of the East Courtillot [1991] constructeda synthetic apparentpolar wan- Antarctic APWP and are significantly different from it (MBL der masterpath for the Atlantic-b•rderingcontinents by com- 100 and TI 110 are significantly different from both the 102 biningwhat they regardedas the mostreliable data from North and 117 Ma East Antarctic mean poles; AP 106M is signifi- America, , India, and Africa. Although Besse and cantly different from the 102 Ma East Antarctic pole but not Courtillot had the benefit of severalnewer poles and usedmore significantly different from the 117 Ma pole, though the latter stringentselection criteria than did Irving and Irving, some test is marginal and inconclusive [McFadden and Lowes, additional paleomagneticpoles are now available for North 1981]). The mean pole for the Pacific-bordering blocks of America,and an updatedage is availablefor a pole from India. West Antarctica (MBL, TI, AP), calculatedfrom MBL 100, TI 15,130 DIVENERE ET AL,: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND

Africa -+- North America (• Australia •* India •

+ 112 112

116

129 130

180 E

Figure 13. Cretaceous poles transferred from Africa, North America, India, and Australia into the East Antarctic reference frame. Poles are listed in Table A1.

110, andAP 106M (212.2øE/ 74.5øS,A95 = 4.6ø),is signifi- addition of a full 300 km of reconstruction related error; the cantly different from the East Antarctic APWP (compared to angularseparation from the East Antarctic 102 Ma pole then both the East Antarctic 102 and 117 Ma poles). The angular becomes10.2 ø + 8.9 ø (8.6ø+ 8.1ø to East Antarctica 117 Ma distance separatingthe Pacific West Antarctic circa 105 Ma pole). mean pole and the 102 Ma mean pole for East Antarctica is The circa 98 and 112 Ma mean poles from the APWP of 10.2ø + 6.2ø (8.6 + 5.4ø to East Antarctica117 Ma pole). Besseand Courtillot [1991] rotatedinto an East Antarcticref- Our preferred plate kinematic model used in constructingthe erenceframe (Figure 14) are fartherremoved from the West APWP assumesa two plate configurationfor Africa, with Early Antarcticpoles and supportthese observations. It can there- to mid-Cretaceous extension in the Benue Trough (see fore be concludedthat there has been a paleornagneticallyre- Appendix). As a check, we recalculatedthe APWP assuminga solvable movement of Pacific West Antarctica (MBL, TI and one plate Africa. The East and West Antarctic mean poles re- AP) with respectto EastAntarctica since about 100 Ma. main significantly different in this model as well. The error assignedto the APWP is only basedon the distri- Tectonic Implications bution of poles and does not include uncertaintiesin the rota- tion parameters for transferring poles to East Antarctica. The paleomagneticevidence for post-100Ma relative mo- Duncan and Richards [1991] noted that typical uncertainties tion between East Antarctica and Pacific West Antarctica is for relative plate motion additions, as calculated by Molnar consistentwith mounting geological evidence that the area and Stock [1987], are in the 200 to 300 km range. Given the betweenthe TransantarcticMountains (TAM) and the coastal tightnessof the swath of rotated poles (Figure 13), we believe areas of West Antarctica representsa Mesozoic-Cenozoic rift that 200 to 300 km should represent a maximum reconstruc- system. This area was termed the West AntarcticRift System tion uncertainty (i.e., some of the reconstructionuncertainty by LeMasurier [1978]. Fitzgerald [1992] concludesthat the is averagedout by combining independentplate circuits). The TAM are probably a rift-margin uplift. Stem and ten Brink Pacific West Antarctic (MBL, TI, AP) mean pole is signifi- [1989] modeledthe TAM in terms of flexural uplift along the canfly different from the East Antarctic APWP even with the boundarybetween thermally old East Antarcticlithosphere and DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND l$,131

East Antarctic APWP This study ß Besse and Courtillot(1991 ) ß

73

270 E

121

127

180 E

Figure 14. Comparisonof the newly constructedAPWP for East Antarcticawith the masterAPWP of Besse and Courtillot [1991] rotated into the East Antarctic reference frame.

thermally young West Antarctic lithosphere,but ten Brink et in the south central Ross Sea bottomed in marble and calc-sili- al. [1993] consider this inadequate and propose lateral heat cate gneiss basement rocks [Ford and Barrett, 1975] similar to flow from the rift to accountfor TAM uplift. North-south lin- calcareous metasedimentary rocks of the lower Paleozoic ear basinsin the Ross Sea are parallel to the TAM and appear Skelton Group of southernVictoria Land [Blank et al., 1963] to continue beneath the Ross Ice Shelf. From west to east, the and calc-silicategneiss and marble at Mount Murphy in east- Victoria Land Basin, Central Trough, and Eastern Basin are ern MBL [DiVenere et al., 1993]. The gneissat Site 270 is separatedby interbasinhighs (Figure 1). Up to 14 km of sed- overlain by 25 m of breccia capped by a subaerealpaleosol iments have been seismically imaged in the Victoria Land [Ford and Barrett, 1975]. Unconformablyoverlying the pale- Basin [Cooper et al., 1987] and a maximum of 7 km of strata osol is a mid-upper to Recent sedimentarysequence, in the Central Trough and Eastern Basin [Davey et al., 1983]. including 1 m of shallow water carbonaceoussandstone, ! m The interbasin highs swing to a NNW-SSE trend southwards of shallow marine glauconitic sandstone(K-Ar age of 26 Ma), beneath the Ross ice shelf, following the trend of the TAM. and 384 m of glaciogenic sediments,which subsidedfrom sea Fitzgerald [1992] and Wilson [1993] identify a component level to 500 m depth (current water depth is 630 m) by the end of dextral motion from the pattern of Cenozoic normal fault- of the early Miocene [Hayes and Frakes, et al., 1975]. The ing in the TAM in addition to the dominant extensional com- deepening was ascribed to either tectonic subsidenceor iso- ponentoriented perpendicular to the trend of the TAM. Storey static compensation due to glacial loading of the [1991] suggests that the deep lozenge-shaped subglacial [Hayes et al., 1975; Hayes and Frakes, 1975]. basins between MBL, TI, and EWM, the Bentley Subglacial Cooper et al. [1991] summarizedmultichannel seismic re- Trench and the Byrd SubglacialBasin (Figure 1), may be pull flection data of Hinz and Block [1984] and Cooper et al. apart basins formed at releasing bends of a dextral strike-slip [1987] in a generalizedprofile acrossthe Ross Sea that shows fault system. fault bounded sedimentarybasins in the Victoria Land basin The oldest sediments thus far drilled in the Ross Sea are and sedimentsdraping faulted basementto the west in the re- Lower Oligocene, 33-38 Ma, from the CIROS-1 drill hole in mainder of the Ross Sea. Based on the profile (combined with McMurdo Sound [Barrett, 1987], though basement was not the CIROS-1 and Site 270 data), faulting in the eastern two- reached. Drilling at Deep Sea Drilling Project (DSDP) Site 270 thirds of the Ross Sea ended before the mid-upper Oligocene 15,132 DIVENERE ET AL.: PALEOMAGNETIC RESULTSFROM MARIE BYRD LAND

,5

TI 110 AP 106M

180 E Figure15. The100 to 110 Ma poles from MBL, TI, andthe southern AP,with Ao3 circles of confidence, plottedagainst the East Antarctic APWP with A63 error envelope. The envelope for the APWP was constructed bydrawing a smooth curve tangent to theA63 circles of themean poles. West Antarctic poles are listed in Table2. EastAntarctic poles are listed in TableA2.

unconformityseen at the bottom of Site 270, but apparently this,Behrendt and Cooper[1991] inferred350 lcmof exten- continuedto Recent times in the Victoria Land basin adjacent sionin theRoss Sea, or about50% of thenominal present-day to the TAM (Figure 1). width. However, the Phanerozoiccrust of West Antarctic Fitzgerald [ 1992] and Stumpand Fitzgerald [ 1992] infer four (excluding EWM) may not have been as thick as the phasesof uplift in the TAM from apatitefission-track age pro- Precambriancratonic crust of EastAntarctica. The average files. The episodesare at approximately110 Ma, which they crustalthickness of coastalWest Antarcticaappears to be associatewith initial rifting of Australia, 80 Ma, which they about30 km [Bentleyet al., 1960;Adams, 1972; Bentley, associatewith the separation of New Zealand and Australia 1991], which wouldimply only about275 km of extension. from Antarctica,55 Ma, the time of cessationof spreadingin Thesecrude estimates then suggest about 300 lcm (275 to 350 the TasmanSea, and Plio- times, during which pe- km) of extensionacross the Ross Sea betweenPacific West riod there is a renewal in alkaline volcanism in the western Antarctica and cratonic East Antarctica. Ross Sea and in MBL, possiblyincluding subglacial volca- ThePacific West Antarctic circa 105 Ma meanpole is offset noesand high heatflow within the West Antarcticrift system fromthe East Antarctic 102 Ma poleby 10.2q- 6.2 ø (or q-8.9 ø [Blankenshipeta/., 1993]. However, Wilch et al. [1993] and withthe addition of themaximum reconstruction related error). ten Brink et al. [1993] presentresults that minimizeany late Sincethe West Antarctic pole is locatedover the coast of MBL Plioceneto Recent uplift. it impliesa similarmovement of MBL (plusTI andthe south- Estimates of the amount of extension in the Ross Sea have ernAP) fromEast Antarctica. Translating the angularoffset been made by comparingthe thicknessof the crust on either intokilometers by assuminga simple extension history about side of the boundary between East and West Antarctica. a distanteuler pole, implies about 1130 q- 690 km of relative Crustalthickness in the West Antarcticrift systemis gener- motionbetween East and West Antarctica(1130 q- 990 km ally in the range 17-23 km with thicknessesof 27 km under withthe maximum reconstruction error). The range of paleo- Ross Sea basementgravity highs, whereasthe thicknessof magnetically acceptablemotions of Pacific West Antarctica the adjacentTAM rift shoulderis about40 km [Behrendtand with respectto East Antarctica is thus about 440 to 1820 lcm Cooper, 1991; Bentley, 1991; Cooper et al., 1991]. From (140to 2120km withmaximum reconstruction error). DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND 15,133

1

outh • America

~ 100 Ma ß'

- /

East

Antarctica

CP

••1•MBL 1øøl i110{• •EAnt1ø2 / Australia M•EAntj.• Figure 16. A circa100 Ma reconstructionwith southpole centered on theEast Antarctic 102 Ma meanpole. SouthNew Zealand(SNZ), includingCampbell Plateau (CP) andChatham Rise (CR), arerestored to MBL. Inset showsWest Antarctic poles (MBL 100, TI 110, andAP 106M) in theirreconstructed positions with respectto thecirca 117 and102 Ma EastAntarctic mean poles (EAnt 117 and 102), all withA63 circlesof confidence.

The best palcomagneticestimate (1130 kin) for separation The timing of this extension is constrainedby the palco- between East Antarctica and Pacific West Antarctica is close to magnetic and geochronologic data to postdate 100 Ma. the largest dimension across the Ross Sea (about 1030 km According to the geologicevidence, the largestportion likely from Edward VII Peninsulaof MBL to the BorchgrevinkCoast occurredsometime prior to the mid-Late Oligoceneafter which of North Victoria Land, East Antarctica) (Figure 1) and is con- time all extension appearsto be confined to only the western siderably larger than the nominal geological estimate of ex- third of the Ross Sea. The Ross Sea could have experienceda tension(about 300 km) from crustal thicknessarguments. We major episodeof extensionduring Late Cretaceousrifting of regard the palcomagneticanalysis to be reasonableand con- the New Zealand continental block preceding its separation servativebecause (1) our East AntarcticAPWP yields a smaller from Antarctica at about 85 Ma. If all significant East-West offset with respect to the mid-Cretaceous Pacific West Antarcticmotion ceasedby the time of New Zealand separation Antarctic palcomagnetic poles than the alternate APWP of it would imply, in conjunctionwith the palcomagneticdata, a Besse and Courtillot [1991], and (2) the new palcomagnetic separationrate equivalent to 7.5 + 6.5 cm/yr between East pole for MBL, which is consistentwith polesfrom TI and AP, Antarctica and Pacific West Antarctica over the interval 100 to greafiyreduces the amountof extensionthat would be required 85 Ma. Alternatively, if East-West Antarctic relative motion in the Ross Sea compared to the previous MBL result of continued into the Cenozoic, lower separation rates are im- Grindley and Oliver [1983], which was inconsistentwith the plied. In that caseMBL-East Antarcticrelative motionswould other West Antarctic results. The amount of extension ex- affect the plate circuit linking the Pacific and African hotspots pectedfrom the comparisonof crustal thicknessesmay under- [Molnar and Stock, 1987; Duncan and Richards, 1991]. estimatethe true separationbetween East Antarctica and MBL We next consider the reconstructionof the 100 Ma position because,for example, the crustal thicknessesmay not be accu- of the West Antarctic blocks with respect to East Antarctica. rately known, there could be oceanic crust flooring the rift Given the lack of a circa 100 Ma pole for EWM, it is left in its basins, and extension could have occurred over a larger area presentposition with respectto East Antarctica. The simplest includingintervening portions of MBL. In any case,the geo- model is to rotate MBL, TI, and AP about a common rotation logical estimateis within the maximum error of the palco- pole. However, it is difficult to reconcilethe circa 100 Ma pa- magneticestimate and the two estimatesare thereforenot in- lcomagneticpoles and the geologic constraintsby a simple compatible.Both supportrelative motion between MBL (plus rigid rotation of Pacific West Antarctica back to East TI andAP) andEast Antarctica of somehundreds of kilometers. Antarcticadue to problemsof overlap of AP with EWM, and 15,134 DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND creation of gaps (implying later subduction) in the Weddell There presentlyis no completelysatisfying solution for the Sea and in the Drake Passage. We note that at face value 100 Ma reconstruction,but the new palcomagneticconstraints (disregardingthe circles of confidenceabout the palcomag- presentedhere suggestsa larger amountof total extensionbe- netic poles) MBL 100 is offset the farthest from the East tween East and West Antarctica (MBL) than previously sus- Antarctic APWP and AP 106M is offset the least. We therefore pected from geological arguments. choose as a second possibility to allow some differential mo- tion between the Pacific West Antarctic blocks within the er- Appendix: Synthetic Apparent Polar Wander rors of the West Antarctic palcomagneticpoles. A possible Path for East Antarctica reconstruction is shown in Figure 16 in which South New Zealand and Campbell Plateau are restoredback to MBL [after The most comprehensivepalcomagnetic data sets for the Mayes et al., 1990], MBL is restoredto East Antarcticaby ap- Cretaceousare from North America and Africa (Table A1). We proximately50% closureof the RossEmbayment, TI is rotated also use palcopoles from India, as did Besse and Courtillot moderately and AP only slightly counterclockwise. This [1991], but unlike them we do not use the poles from Eurasia model is at the limits of both the palcomagnetic error con- because:(1) their selected data set contains no mid-Cretaceous straints as well as the geologic spatial constraints. The results, which are of greatest interest here, (2) the Early Pacific West Antarctic palcomagneticpoles are restorednearer Cretaceousresults from China and Korea require the assump- to the East Antarctic APWP: TI 110 is reconciled with the East tion of no internal deformation of Eurasia since that time, Antarctic APWP, AP 106M is at the limit of the paleomagnetic even thoughthis may not be specificallytrue consideringthe error, and MBL 100 may only be reconciledwith the inclusion complex tectonics associatedwith the collision of India with of reconstruction related error. At the same time, the model Eurasia,and (3) use of the Eurasiapoles requiresa very long requires a large amount of extension (approaching 100%) in plate circuit with two additional relative motion pairs the northern Ross Sea. However, the Campbell Plateau is (Eurasia/Greenlandand Greenland/NorthAmerica) to transfer shown with its present outline, though it is also likely to con- the poles to Antarctica. The African, Indian, and North sist of extendedcontinental crust (Bradshaw [ 1991] suggested Americanpoles must ultimately all be transferredthrough the about 30% extension). Accounting for Campbell Plateau ex- Africa-Antarctica plate circuit, and therefore their reconstruc- tension would reduce the amount of extensionrequired for the tion associatederrors are not entirely independent. The northern Ross Sea crust. This model implies a post 100 Ma Australian poles then become an important check on the re- shearzone betweenMBL and TI, which Grunoweta/. [1991] constructions. We rely largely on the pole selection criteria had previously suggestedbased on the comparisonof TI 110 of other authors in recent reviews of Cretaceous results from with the MBL result of Grindley and Oliver [1983]. Africa [Hargraves, 1989], North America [Globerman and

Table AI. PaleomagneticPoles Transferred to East An•tarctic Reference Frame

Original Transferred

Pole Age A95 Plon Plat PLon Plat Continent Source DeccanTraps 65 2.4 101.3 -36.9 54.4 -80.0 India 1 Madagascardolerite and volcanics 75 4.1 39.6 -63.5 30.8 -77.1 Africa 2 Fuerteventuralavas 80 4.8•' 40.8 -68.8 18.8 -84.0 Africa 3 Niobrara Formation 88 4 188.0 64.0 47.3 -78.0 North America 4 Kimberlitepole 1 90 5.2 46.1 -64.1 17.9 -85.1 Africa 5 Madagascarvolcanics 90 4.9 60.0 -69.1 166.0 -87.2 Africa 2 WadiNatash 95 5.8 78.1 -69.3 177.7 -80.3 Africa 6 Wadi Natash* 95 3.6 70.9 -64.7 144.8 -82.9 Africa 7 MountDromedary 96 9.9 138.0 -56.0 173.5 -82.4 Australia 8 MagnetCove 100 4.6 197.3 72.6 137.9 -82.9 NorthAmerica 9 LupataLava 112 4.0 79.0 -61.8 186.7 -77.0 Africa 10 OtwayGroup 112 3.6 148.7 -48.9 181.3 -73.3 Australia 11 RajmahalTraps 116 6 117.0 -7.0 185.6 -71.8 India 12 White Mountains 122 6.9 187.4 71.9 168.6 -77.1 North America 13 MonteregianHills 124 2 191.0 73.0 171.8 -74.9 NorthAmerica 14 Newfoundland dikes 129 4 207.0 71.0 164.7 -68.4 North America 15 Kaoka lavas, Namibia 130 3.9•' 86.6 -48.3 177.0 -70.4 Africa 16 Kimberlitepole 2 130 9.7 89.9 -47.6 175.0 -68.1 Africa 5 Agesin megaannum. Continent is thatfrom whichpole was transferred. See Table A3 for Eulerpoles for transferringpoles. See Tables1 and 2 for explanationsof otherlisted parameters. Source: 1, Vandammeet al. [1991]; 2, calculatedby McElhinnyand Cowley[1978] from dataof Andriamirado [1971]; 3, Storetvedtet al. [1979]; 4, Shiveand Frerichs [1974]; 5, Hargraves [1989]; 6, Schultet al. [1981]; 7, Ressetaret al. [1981]; 8, Robertson[1963]; 9, Globermanand Irving [1988]; 10, Goughand Opdyke[1963]; 11, Idnurm[1985]; 12, Klootwijk[1971] with age of Baksi[1986]; 13, VanFossen and Kent [1992]; 14, Foster and Symons[1979]; 15, LaPointe [1979]; 16, Gideskehauget al. [1975]. * Calculated from their 31 site mean VGPs. • A95Conservatively estimated as the major axis of theconfidence ellipse. DIVENERE ET AL.' PALEOMAGNETIC RESLILTS FROM MARIE BYRD LAND 15,135

Irving, 1988; Van Fossen and Kent, 1992], and India [Besse CretaceousMesaverde Group [Kilbourne, 1969] and Maudlow and Courtillot, 1988, 1991]. The Cretaceouspaleomagnetic Formation [Swensonand McWilliams, 1989] poles, whereas data set for Australia is reviewed here. Only poles with well- we reject the Late Cretaceouspoles from North America be- definedmeans (A95 less than 10 ø) are accepted. causeof the above mentionedambiguity apparentfrom the re- The Cretaceous APWP for Africa has been analyzed by cently available studiesof rocks of this age. Hargraves [1989]. In his choice of 10 poles, ranging in age Only two Cretaceouspoles from India, the Rajmahal Traps from about 130 to 75 Ma, which met his selection criteria, two pole of Klootwijk [1971] and the Deccan Traps pole of results(Lupata Lavas, Mlanje Massif) were deemedmarginally Vandammeet al. [1991], have well-constrainedages and char- acceptable but retained because they filled a temporal gap acteristic magnetizations. Both were acceptedby Besse and (110-120 Ma). The Mlanje Massif pole is rejected for this Courtillot [1991] for inclusion in their master APWP and are studybecause it is basedon only eight samples,its A95ex- usedin the presentstudy. The Rajmahal Traps have an updated ceeds10 ø, and polesof this age are availablefrom otherconti- ageof 116Ma basedon nøAr/39Ar [Baksi, 1986] (Table A1). nents. The remaining nine poles are accepted (Table A1). Four paleomagneticpoles from Cretaceousunits have been Besse and Courtillot [1991] used the same African data for this reported for Australia. In an early study, Robertson [1963] time period as Hargraves [1989], but the data were combined gave a thoroughreport of paleomagneticresults based on AF somewhatdifferently. Besse and Courtillot used six separate and thermal demagnetizationand rock magnetic studiesfrom poles from Madagascar from the original work by the Mount Dromedaryigneous complex with an age of circa 96 Andriamirado [1971] while Hargraves [1989] (and the present Ma [McDougall and Wellman, 1976] (with new decay con- study) used the two mean poles of McElhinny and Cowley stantsof Steigerand Jiiger [ 1977]). ldnurm [ 1985] reportedre- [1978] recalculatedfrom the two age groupsof the data. Besse sults based on progressiveAF and thermal demagnetization and Courtillot also combined the Mlanje Massif and Lupata from volcaniclastic sedimentary rocks of the Otway Basin lavas poles into a single pole. with an age of approximately 112 Ma. We accept these two The North American Cretaceousstillstand, as identified by results for inclusion in the APWP (Table A1). Globermanand Irving [1988] and updatedby Van Fossenand We reject poles from two other studiesof Cretaceousrocks Kent [1992], is based on five poles which range in age from from Australia. Robertsonand Hastie [1962] reportedresults approximately129 Ma to 88 Ma. These poles form a cluster from the circa 98 Ma Cygnetalkaline complex but theseearly in North American coordinates, implying no significant ap- resultsare basedon batchdemagnetization at only 7.5 mT and parent polar wander of North America during this interval. yielda poorlydefined mean (A95 = 10ø).Schmidt [1976] re- One of these results, the Arkansas intrusions pole of portedpaleomagnetic results from the Bunburybasalts, listing Globermanand Irving [1988], is a combinedaverage based on an age of 90 Ma, but the dual polaritiesobserved are unlikely five sites from the circa 88 Ma Granite Mountain intrusion and to have been acquired during the Cretaceous Normal 19 sites from the circa 100 Ma Magnet Cove plus Potash Superchron(118 to 84 Ma). The age of the Bunburybasalts Sulfur Springs intrusions. Given the difference in their age, was basedon discordantK-At resultsfrom McDougall and the two group means represent distinct readings of the mag- Wellman [1976] with agesranging from 88 to 105 Ma; these netic field and are separatedfor the present study. The circa authorsconcluded that their resultswere affectedby argonloss 100 Ma Magnet Cove (+Potash Sulfur Springs) pole is ac- cepted. The ca 88 Ma Granite Mountain pole groupswell with the other Cretaceousstillstand poles but is rejected becausethe Table A2. East AntarcticSynthetic Apparent Polar Wander Path A95exceeds 100. The post-88 Ma Cretaceouspoles from North America are more scattered and, at face value, can either be used Age Mean to supporta continuationof the stillstand(based on the results Range Age of Vugteveenet al. [1981], Swensonand McWilliams [1989], N k A95 PlonøE PlatøN Source andJolly and Sheriff[1992]) or a suddenjump to the Paleocene Pliocene 3.5 5 884 2.6 320.5 -86.4 1 pole position (based on the results of Diehi [1991] and Miocene 15 5 190 5.6 320.9 -81.9 1 Gundersonand Sheriff[1991]). Moreover, theseresults are all Oligocene* 35 1 4.4'{' 345.5 -77.6 2 from units within or closely associatedwith the Cretaceous Paleocene 62 1 4.4'{' 14.0 -79.9 2 foreland thrust belt, and some or all are likely to have been ro- 65-84 73 3 331 6.8 36.4 -80.6 3 tated. The post-88 Ma North American Cretaceouspoles are 75-94 85 5 167 5.9 36.0 -83.4 3 thereforenot included due to this ambiguity. In summary,the 85-104 93 7 125 5.4 130.9 -86.0 3 five Cretaceous stillstand poles from Van Fossen and Kent 95-114 102 6 255 4.2 171.8 -80.3 3 [1992], with the above-mentioned treatment of the results of 105-124 117 4 681 2.9 179.2 -74.9 3 Globerman and Irving [1988], are accepted from North 115-134 125 6 360 3.5 173.9 -71.9 3 America (Table A1). The differences between Besse and Courtillot's [1991] North American poles and those accepted Ages in mega annum. N, numberof studies. See Tables 1 here are that Besseand Courtillot used:(1) the Mount Ascutney and 2 for explanationsof otherlisted parameters. Source: pole of Opdyke and Wensink[1966], which is here superseded 1, Africanmean poles of Tauxeet al. [1983], Pliocenepole = 329øE/ 86øS,Miocene pole = 334øE/ 81øSin African co- by the White Mountainspole of Van Fossenand Kent [1992]; ordinates;2, African mean polesof Schneiderand Kent (2) the Isachsendiabase pole of Larochelie and Black [1963] [1990],Oligocene pole = 4.2øE/ 74.6øS,Paleocene pole = which wasrejected by Globermanand Irving [ 1988] (and in the 33.0øE/ 70.1øEin African coordinates;3, mean of trans- presentstudy) because A95 is greaterthan 10ø; (3) theMagnet ferredpoles in TableA1. SeeTable A3 for Eulerpoles used Cove pole of Scharon and Hsu [1969] in addition to the to transferAfrican mean poles into East Antarctic coordinates. MagnetCove pole of Globermanand Irving [1988] thoughthe * their pole basedon geocentricaxial dipole model. latter pole supersedesScharon and Hsu's pole, which is other- '{'A95 estimated as the majoraxis of the confidenceel- wiserejected because A95 is greaterthan 10ø; and (4) theLate lipse. 15,136 DIVENERE ET AL.: PALEOMAGNETIC RESLILTS FROM MARIE BYRD LAND

Table A3. Euler Poles Used for Transferring cluded for completenessand is basedon the African Paleocene PaleomagneticPoles to EastAntarctica and Oligocene poles of Schneider and Kent [1990] and the African Pliocene and Miocene mean poles of Tauxe et al. [1983] (Table A2). Continent Age Latitude Longitude Angle, Ma øN øE deg The North American poles were transferredto northwestern Africa using the reconstructions of Pindell et al. [1988]. New Zealand 95 64.03 -56.96 57.65 Rotationparameters for openingin the Benue Trough between India 65 9.62 !6.39 -46.72 136 Ma and 84 Ma (northwestern Africa to ) India 115 -1.39 9.90 -80.85 were derived from the equatorialfit of northwesternAfrica and Australia 96 1.00 38.00 -28.30 of Pindell et al. [1988] and the South Atlantic Australia 1 12 0.19 38.24 -29.09 fit and opening of Rabinowitz and LaBrecque [1979] Africa 3.5 12.90 -39.77 -0.52 (following the model of Pindell et al. [1988]). The 15 11.48 -53.79 -2.04 American poles, as well as the African poles, were then trans- Africa 35 9.76 -39.52 -5.38 ferred from southernAfrica to East Antarctica using the recon- Africa 62 2.66 -39.52 -10.86 Africa 75 -2.50 -40.99 -14.01 structionsof Royer et al. [1988] and Royer and Coffin [1992]. Africa 80 -4.63 -39.74 -15.96 Australianpoles were transferredusing poles of rotation inter- Africa 90 -3.54 -36.58 -21.83 polated from the reconstructions of Royer and Sandwell Africa 95 -4.48 -34.96 -24.27 [1989], and the Indian poles were transferredusing the recon- Africa 112 -6.49 -31.50 -34.31 structionsof Royer and Coffin [1992]. The interpolationsas- Africa 130 -7.18 -28.25 -46.27 sume a constantspreading rate through the CretaceousQuiet North America 88 66.11 108.12 35.33 Zone. The timescale used was that of Kent and Gradstein North America 100 65.99 104.08 44.34 [1986]. The interpolatedrotation poles are listed in Table A3. North America 122 62.98 94.68 58.24 Most of the African poles are from southernAfrica. The North America 124 62.56 93.85 58.84 northernAfrican poles are either youngerthan the openingof North America 129 61.48 92.00 60.30 the BenueTrough or in the caseof the two Wadi Natash, (circa 95 Ma) poles there is less than 1ø of rotationin the Ages accordingto timescaleof Kent and Gradstein model. It is unclear how the model appliesto Egypt, at the [1986]. Rotation parametersare interpolatedbetween the reconstructionsfrom the following sources. New Zealand pivotof the openingwedge. Thereforeno correctionfor ex- to MBL, Mayes et al. [1990]; India to Antarctica,Royer tensionin the Benue Trough has been appliedto the Wadi and Sandwell [1989] 65 Ma; Royer and Coffin [1992] 117 Natash poles. Ma; Australia to Antarctica,Royer and Sandwell [ 1989]; Africa to Antarctica,Mayes et al. [1990] 3.5 - 35 Ma; Acknowledgments.We wouldlike to thankSPRITE membersSam Royer et al. [1988] 62 - 80 Ma; Royer and Coffin [1992] 90 Mukasa and Dave Palais of the University of Michigan, Bob Pankhurst - 130 Ma; North America to NW Africa, Pindell et al. and Bryan Storey of the British AntarcticSurvey (BAS), and John [1988]; NW Africa to southAfrica derivedfrom NW Africa- Bradshawand Steve Weaver of the University of Canterburyfor their SouthAmerica equatorialfit of Pindell et al. [ 1988] and contributionsto this researchand permissionto use their unpublished South Atlantic fit and openingof Rabinowitz and data. We also thank mountaineersPete Cleary and Andy Harris of the LaBrecque [1979]; Africa to Antarctica,Royer et al. [1988] New ZealandAntarctic Research Program (NZARP; now NZAP) and 88 Ma; Royer and Coffin [1992] 100 - 129 Ma. Damo Carrollof BAS for their expertguiding in the field andassistance Interpolationsassume constant spreading rate in K quiet in samplecollection; Twin Otterpilot Paul Robertsonand air mechanic zone; negative rotation angles are clockwise. Alan Hopkinsof BAS for safelydelivering us to distantoutcrops; U.S. Navy AntarcticDevelopment Squadron VXE6 for LC-130 support;the U.S. Office of Polar Programsand AntarcticSupport Associates for lo- and only supportedan age of at least 90 Ma. The Bunbury gisticalsupport; and NZARP and the personnelof ScottBase for field basaltsare underlainby Upper Jurassicstrata and overlainby supportand baseaccommodations. The first authorwould also like to Lower Cretaceous strata [Veevers, 1984; Playford et al., thankWalter Pitmanand SteveCande for helpful discussionsregarding 1976]. Veevers [1984] considersthe emplacementof the the globalreconstructions. We thankLisa Gahaganat the Institutefor Geophysics,University of Texasat Austinfor help with the Antarctic Bunbury basalts to be related to the initiation of seafloor reconstructionin Figure16. Thanksto Myrl Beck,Karl Kellogg,Russell spreadingbetween Greater India and Australia at about 128 Ma Burmester,and an anonymousreviewer for thoughtfulreviews and (-132 Ma using timescale of Kent and Gradstein, 1986). commentsthat helped improve this paper. This researchwas supported Davies et al. [1989] and Storey et al. [1992] consider the by NSF Office of PolarPrograms grants DPP 8916470(D. V. K.) and Bunbury basaltsto be related to the inceptionof the mantle DPP 8917127 (I. W. D. D.). Lamont-Doherty contribution number plume responsiblefor the Early CretaceousNaturaliste Plateau, 5204. RajmahalTraps, and KerguelenPlateau, beginning at around 120-130Ma. It thereforeseems likely that the mixedpolarity References magnetizationsreported for the Bunburybasalts were acquired during emplacementin the Early Cretaceous,before the onset Adams,C. J., Geochronologicalstudies of the SwansonFormation of of the CretaceousNormal Superchron(118 Ma). Also, a num- Marie Byrd Land, West Antarcticaand correlationwith northern ber of palcomagneticresults from Australiahave been reported VictoriaLand, East Antarctica, and South Island, New Zealand, N. Z. J. Geol.Geophys., 29, 345-358,1986. in older rocks, apparently remagnetizedduring the mid- Cretaceous[Schmidt and Embleton,1981; Schmidt,1976], but Adams,C. J., Geochronologyof granite terranes in theFord Ranges, MarieByrd Land, West Antarctica, N. Z. J. Geol.Geophys., 30, 51- the magnetizationage is only inferredby the generalsimilar- 72, 1987. ity with the Cretaceouspoles. These remagnetizationpoles Adams,R. D., Dispersion-wavestudies in Antarctica,in Antarctic are not suitable for inclusion in the APWP. Geologyand Geophysics,edited by R. J. Adie,pp. 473-480, The Cenozoic portion of the East Antarctic APWP is in- Universitetsforlaget,Oslo, 1972. DIVENERE ET AL.: PALEOMAGNETIC RESLILTS FROM MARIE BYRD LAND 15,137

Andriamirado,C.R.A., Recherchespa16omagn6fiques sur Madagascar. the trail of a Dupal plume, Contrib. Mineral. Petrol., 103, 457-469, Resultatset interpr6tafionsdans le cadrede la dislocationde la partie 1989. occidentaledu Gondwana,Ph.D. thesis,350 pp., Univ. of Strasbourg, Delisle, G., Resultsof palcomagneticinvestigations in northernVictoria Strasbourg,France, 1971. Land, Antarctica,in Antarctic Earth Science,edited by R. L. Oliver, Baksi,A. K.,4øAr/39Ar incremental heating study of whole-rocksam- P. R. James,and J. B. Jago, pp. 146-149, AustralianAcademy of ples from the Rajmahaland Bengaltraps, eastern India (abstract), Science, Canberra, New South Wales, 1983. Terra Cognita,6, 161, 1986. Delisle, G., and K. Fromm, Palcomagnetic investigation of Ferrar Barley, M. E., S. D. Weaver, and J. R. de Laeter, Strontiumisotope Supergrouprocks, North Victoria Land, Geol. Jahrb., B 41, 41-55, compositionand geochronologyof intermediate-silicicvoltanits, 1984. Mount Somers and Banks Peninsula, New Zealand, N. Z. J. Geol. Delisle, G., and K. Fromm, Further evidence for a Cretaceousthermal Geophys.,31, 197-206, 1988. event in North Victoria Land, Geol Jahrb., E 38, 143-151, 1989. Barrett,P. J., Oligocenesequence cored at CIROS-1, westernMcMurdo Diehl, J. F., The Elkhorn Mountains revisited: New data for the Late Sound,N.Z Antarct. Rec., 7 (3), 1-7, 1987. Cretaceouspalcomagnetic field of North America,J. Geophys.Res., Behrendt,J. C., and A. Cooper,Evidence of rapidCenozoic uplift of the 96, 9887-9894, 1991. shoulderescarpment of the CenozoicWest Antarcticrift systemand DiVertere, V. J., J. D. Bradshaw, S. D. Weaver, D. G. Palais, R. J. a speculationon possibledimate forcing, Geology, 19, 315-319, Pankhurst,and B. C. Storey, Geological investigationsin eastern 1991. Marie Byrd Land,West Antarctica,Antarct. J. U.S., 28, 5-6, 1993. Bentley, C. R., Configurationand structureof the subglacialcrust, in Duncan,R. A., Hotspotsin the southernoceans - An absoluteframe of The Geology of Antarctica, edited by R. J. Tingey, pp. 335-364, referencefor motion of the Gondwanacontinents, Tectonophysics, Clarendon, Oxford, 1991. 74, 29-42, 1981. Bentley,C. R., A. P. Crary,N. A. Ostenso,and E. C. Thiel, Structureof Duncan,R. A., and M. A. Richards,Hotspots, mantle plumes, flood West Antarctica, Science, 131, 131-136, 1960. basalts,and true polar wander,Rev. Geophys.,29, 31-50, 1991. Besse,J., and V. Courtillot,Palcogeographic maps of the continentsbor- Fisher,R. A., Dispersionon a sphere,Proc. R. Soc.London Ser. A, 217, deringthe Indian Oceansince the Early Jurassic,J. Geophys.Res., 295-305, 1953. 93, 11791-11808. 1988. Fitzgerald, P. G., The TransantarcticMountains of SouthernVictoria Besse,J., and V. Courtillot,Revised and syntheticapparent polar wan- Land:The applicationof apafitefission track analysis to a rift shoul- der paths of the African, Eurasian, North American and Indian der uplift, Tectonics,II, 634-662, 1992. plates,and true polar wandersince 200 Ma, J. Geophys.Res., 96, Ford, A. B., and P. J. Barreu, Basementrocks of the south-centralRoss 4029-4050, 1991. Sea,Site 270, DSDP Leg 28, Initial Rep.Deep SeaDrill. Proj., 28, Blank, H. R., R. A. Cooper,R. H. Wheeler,and I. A. G. Willis, Geology 861-868, 1975. of the Koettlitz-Blue Glacier region, southern Victoria Land, Foster,J., andD. T. A. Symons,Defining a palcomagneticpolarity pat- Antarctica,Trans. R. Soe.N. Z., 2, p. 791, 1963. tern in the Monteregianintrusives, Can. J. Earth Sci., 16, 1716-1725, Blankenship,D. D., R. E. Bell, S. M. Hodge, J. M. Brozena, J. C. 1979. Behrendt, and C. A. Finn, Active volcanism beneath the West Gideskehaug,A., K. M. Creer,and J. G. Mitchell,Palcomagnetism and Antarcticice sheetand implicationsfor ice-sheetstability, Nature, K-Ar agesof the South-WestAfrican basalts and their bearing on 361,526-529, 1993. thetime of initial riftingof the SouthAtlantic Ocean, Geophys. J. R. Boudette,E. L., R. F. Marvin, and C. E. Hedge, Biofitie, potassium- Astron. Soc., 42, 1-20, 1975. feldsparand whole-rockages of adamellite,Clark Mountains,west Globerman,B. R., and E. Irving, Mid-Cretaceouspalcomagnetic field Antarctica,U.S. Geol. Surv.Prof. Pap.,550-D,D-190-94, 1966. for North America: restudy of 100 Ma intrusive rocks from Bradshaw,J. D., Cretaceousdispersion of Gondwana:Continental and Arkansas,J. Geophys.Res., 93, 11721-11733, 1988. oceanic spreadingin the south-westPacific-Antarctic sector, in Gordon,R. G., and A. Cox, Palcomagnetictest of the Early Tertiary GeologicalEvolution of Antarctica,edited by M. R. A. Thomson,J. platecircuit between the Pacificbasin plates and the Indianplate, J. A. Crame,and J. W. Thomson,pp. 581-585, CambridgeUniversity Geophys.Res., 85, 6534-6546, 1980. Press,New York, 1991. Gough,D. I., and N. D. Opdyke,The palcomagnetismof the Lupataal- Bradshaw, J. D., P. B. Andrews, and B. D. Field, Swanson Formation kaline voltanits, Geophys.J. R. Astron.Soc., 7, 457-468, 1963. and related rocks of Marie Byrd Land and a comparisonwith the GrindIcy,G. W., and F. J. Davey, The Reconstructionof New Zealand, RobertsonBay Group of northernVictoria Land, in Antarctic Earth Australia, and Antarctica,in Antarctic Geoscience,edited by C. Science,edited by R. L. Oliver, P. R. James,and J. B. Jago,pp. 274- Craddock,pp. 15-29,University of WisconsinPress, Madison, 1982. 279, AustralianAcademy of Science,Canberra, New SouthWales, GrindIcy,G. W., and P. J. Oliver, Palcomagnetismof Cretaceousvol- 1983. canic rocks from Marie Byrd Land, in Antarctic Earth Science, Burk, K., and J. F. Dewey, Two platesin Africa duringthe Cretaceous?, editedby R. L. Oliver, P. R. James,and J. B. Jago,pp. 573-578, Nature, 249, 313-316, 1974. AustralianAcademy of Science,Canberra, New SouthWales, 1983. Cooper,A. K., F. J. Davey, and J. C. Behrendt,Seismic stratigraphy and Grunow, A.M., Palcomagneticdata from the AntarcticPeninsula and structure of the Victoria Land Basin, western Ross Sea, Antarctica, their implicationfor openingof the WeddellSea, J. Geophys.Res., in TheAntarctic Continental Margin: Geologyand Geophysicsof the 98, 13815-13833, 1993. WesternRoss Sea, Earth Sci. Ser., vol. 5B, editedby A. K. Cooper Grunow, A.M., I. W. D. Dalziel, and D. V. Kent, Ellsworth-Whitmore and F. J. Davey, pp. 27-65, Circum-PacificCouncil for Energyand Mountainscrustal block, western Antarctica: New palcomagneticre- Mineral Resources,Houston, Tex., 1987. sults and their tectonicsignificance, in Gondwana Six: Structure, Cooper,A.K., F.J. Davey, and K. Hinz, Crustalextension and origin of Tectonics,and Geophysics,Geophys. Monogr. Ser., vol. 40, editedby sedimentarybasins beneath the Ross Sea and Ross Ice Shelf, in G. D. McKenzie,pp. 161-171,AGU, Washington,D.C., 1987a. GeologicalEvolution of Antarctica,edited by M. R. A. Thomson,J. Grunow, A.M., D. V. Kent, and I. W. D. Dalziel, Mesozoic evolution A. Crame,and J. W. Thomson,pp. 285-291, CambridgeUniversity of West Antarcticaand Weddell Sea Basin: New palcomagnetic Press,New York, 1991. constraints,Earth and Planet. Sci. Lett., 86, 16-26, 1987b. Dalziel, I. W. D., and D. H. Elliot, West Antarctica: Problem child of Grunow,A.M., D. V. Kent, and I. W. D. Dalziel, New palcomagnetic Gondwanaland,Tectonics, I, 3-19, 1982. data from ThurstonIsland: Implicationsfor the tectonicsof West Davey, F. J., K. Hinz, and H. Schroeder,Sedimentary basins of the Ross Antarcticaand Weddell Sea Opening,J. Geophys.Res., 96, 17935- Sea region,in Antarctic Earth Science,edited by R. L. Oliver, P. R. 17954, 1991. James,and J. B. Jago,pp. 533-538, AustralianAcademy of Science, Gunderson,J. A., and S. D. Sheriff,A new Late Cretaceouspalcomag- Canberra, New South Wales, 1983. netic pole from the Adel Mountains, west central Montana, J. Davies, H. L., et al., Basaltbasement from the KerguelenPlateau and Geophys.Res., 96, 317-326, 1991. 15,138 DIVENERE ET AL.: PALEOMAGNETIC RESLILTS FROM MARIE BYRD LAND

Halpern,M., Agesof Antarcticand Argentine rocks bearing on conti- Tectonicsfrom a gravityanomaly and infraredsatellite image, GSA nentaldrift, Earth Planet.Sci. Lett., 5, 159-167, 1968. Today,3, 117-122, 1993. Halpern,M., Rb-Srtotal-rock and mineralages from the Marguerite LeMasurier,W. E., The CenozoicWest Antarcticrift systemand its as- Bay area, Kohler Range and Fosdick Mountains,in Antarctic sociated volcanic and structural features, Geol. Soc. Am. Abstr. Geology and Geophysics,edited by R. J. Adie, pp. 197-204, Programs,10, 443, 1978. Universitetsforlaget,Oslo, 1972. LeMasurier, W. E., and D.C. Rex, Rates of uplift and the scaleof ice Hargraves,R. B., Palcomagnetismof Mesozoickimberlites in Southern level instabilitiesrecorded by volcanic rocks in Marie Byrd Land, Africa and the Cretaceousapparent polar wandercurve for Africa, West Antarctica,in Antarctic Earth Science,edited by R. L. Oliver, J. Geophys.Res., 94, 1851-1866, 1989. P. R. James,and J. B. Jago, pp. 663-670, AustralianAcademy of Hadand,W. B., A. V. Cox, P. G. Llewellyn,C. A. G. Pickton,A. G. Science, Canberra, New South Wales, 1983. Smith,and R. Walters,A GeologicTime Scale, 131 pp., Cambridge LeMasurier,W. E., and D.C. Rex, The Marie Byrd Land volcanic UniversityPress, New York, 1982. provinceand its relationto the CainozoicWest Antarcticrift system, Hayes,D. E., andL. A. Frakes,General synthesis, Deep Sea Drilling in The Geologyof Antarctica,edited by R. J. Tingey,pp. 249-284, ProjectLeg 28, Initial Rep.Deep Sea Drill. Proj.,28, 919-942,1975. Clarendon, Oxford, 1991. Hayes,D. E., et al., Sites270, 271,272, Initial Rep.Deep SeaDrill. Longshaw,S. K., and D. H. Griffiths, A palaeomagneticstudy of Proj., 28, Washington(U.S. GovernmentPrinting Office), 211-334, Jurassicrocks from the AntarcticPeninsula and its implications,J. 1975. Geol. Soc.London, 140, 945-954, 1983. Hinz, K., and M. Block, Resultsof geophysicalinvestigations in the Mason,B., and S. R. Taylor, High-gradebasement gneisses and grani- WeddellSea and in the RossSea, Antarctica, Proceedings of 11th toids in Westland, New Zealand, J. R. Soc. N. Z., 17, 115-138, 1987. Petroleum Congress,(London) 1983, vol. 2,_Geology Mayes,C. L., L. A. Lawvet, and D. T. Sandwell,Tectonic history and ExplorationReserves, pp. 79-81,John Wiley, New York, 1984. new isochronchart of the SouthPacific, J. Geophys.Res., 95, 8543- Hole,M. J., andW. E. LeMasurier,Tectonic controls on thegeochemi- 8567, 1990. cal compositionof Cenozoic marie alkaline volcanic rocks from McDougall, I., and P. Wellman, Potassium-argonages for some WestAntarctica, Contrib. Mineral Petrol.,in press,1994. AustralianMesozoic igneous rocks, Aust. J. Earth Sci., 23, 1-9, 1976. Idnurm, M., Late Mesozoic and Cenozoic palaeomagnetismof McElhinny,M. W., andJ. A. Cowley,Palcomagnetic directions and pole Australia,I, A redeterminedapparent polar wanderpath, Geophys. positions,XV Pole numbers15/1 to 15/232, Geophys.J. R. Astron. J. R. Astron. Soc., 83, 399-418, 1985. Soc., 52, 259-276, 1978. Irving, E., and G. A. Irving, Apparentpolar wanderpaths from the McFadden, P. L., and F. J. Lowes, The discriminationof mean directions Carboniferousthrough Cenozoic and the assemblyof Gondwana, drawnfrom Fisherdistributions, Geophys. J. R. Astron.Soc., 67, 19- Geophys.Surv., 5, 141-188, 1982. 33, 1981. Inokuchi,H., and F. Heider,Palcolatitude of the southernKerguelen McFadden,P. L., and A. B. Reid, Analysisof palcomagneticinclination Plateauinferred from the palcomagneticstudy of UpperCretaceous data,Geophys. J. R. Astron.Soc., 69, 307-319, 1982. basalts,Ocean Drill. ProgramSci Results, 120, 89-96, 1992. Mcintosh, W. C., P. R. Kyle, E. M. Cherry, and H. C. Noltimier, Jolly,A.D., andS. D. Sheriff,Palcomagnetic study of thrust-sheetmo- Palcomagneticresults from the KirkpatrickBasalt Group, Victoria tion along the Rocky Mountain front in Montana,Geol. Soc. Am Land, Antarct. J. U.S., 17, 20-22, 1982. Bull., 104, 779-785, 1992. Molnar, P., and J. Stock, Relative motionsof hotspotsin the Pacific, Kellogg,K. S., Palcomagneticevidence for oroclinalbending of the Atlantic and Indian Oceans since late Cretaceoustime, Nature, 327, southernAntarctic Peninsula, Geol. Soc. Am. Bull., 91, 414-420, 587-591, 1987. 1980. Norton, I. O., and J. G. Sclater, A model for the evolution of the Indian Kellogg,K. S., A palcomagneticinvestigation of rocksfrom the Ohio Ocean and the breakup of Gondwanaland,J. Geophys.Res., 84, Rangeand the Dry Valleys, TransantarcticMountains, Antarctica, N. 6803-6830, 1979. Z J. Geol, Geophys.,31, 77-85, 1988. Oliver, P. J., T. C. Mumrot, G. W. GrindIcy, and P. Vella, Kellogg, K. S., and R. L. Reynolds,Paleomagnetic results from the Palaeomagnetismof the Upper CretaceousMt SomersVolcanics, LassiterCoast, Antarctica, and a test for oroclinalbending of the Canterbury,New Zealand, N. Z. J. Geol. Geophys.,22, 199-212, AntarcticPeninsula, J. Geophys.Res., 83, 2293-2298,1978. 1979. Kellogg,K. S., andP. D. Rowley,Structural geology and tectonics of the Opdyke, N. D., and H. Wensink, Palcomagnetismof rocks from the Orville Coastregion, southern Antarctic Peninsula, Antarctica, U.S. White Mountain plutonic-volcanicseries in New Hampshireand Geol.Surv. Prof. Pap., 1498, 25 pp., 1989. Vermont,J. Geophys.Res., 71, 3045-3051, 1966. Kent, D. V., andF. M. Gradstein,A Jurassicto recentchronology, in Palais,D. G., S. B. Mukasa,and S. D. Weaver,U-Pb and 4øAr/39Ar The Geologyof NorthAmerica, vol. M, TheWestern Atlantic Region, geochronologyfor plutons along the Ruppert and Hobbs Coasts, editedby P. R. Bogt and B. E. Tucholke,pp. 45-50, Geological Marie Byrd Land, West Antarctica: Evidence for rapid transition Societyof America, Boulder,Col., 1986. from arc to rift-related magmatism(abstract), Eos, Trans. AGU, 74 Kilbourne,D. E., Palcomagnetismof somerocks from the Mesaverde (16), SpringMeeting suppl., 123, 1993. Group,south-western Wyoming and north-easternUtah, Geol. Soc. Patriat, P., Reconstructionde l'tvolution du systb.me de dorsalesde Am. Bull., 80, 2069-2074, 1969. l'Octanlndien par les mithodesde la cinimatiquedes plaques, thb•se Kirschvink,J. L., The least-squaresline andplane and the analysisof d'ttat, 308 pp., Univ. ParisVII, 1983. palcomagneticdata, Geophys.J. R. Astron.Soc., 62, 699-718, 1980. Patfiat,P., J Stgoufin,J. Goslin,and P. Beauzart,Relative position of Klootwijk,C. T., Palaeomagnetismof the upperGondwana Rajmahal Africa and Antarcticain the Upper Cretaceous:Evidence for non Traps,Northeast India, Tectonophysics,12, 449-467,1971. stationarybehaviour of fracture zones,Earth Planet. Sci. Lett., 75, Langel,R. A., InternationalGeomagnetic Reference Field, 1991 revi- 204-214, 1985. sion: InternationalAssociation of Geomagnetismand Aeronomy Pindell, J., and J. F. Dewey, Pertoo-Triassicreconstruction of Western (IAGA) DivisionV, WorkingGroup 8: Analysisof the mainfield Pangeaand the evolutionof the Gulf of Mexico/Caribbeanregion, andsecular variation, Phys. Earth Planet.Inter., 70, 1-6, 1992. Tectonics,l, 179-211, 1982. LaPointe,P. L., Palcomagnetismof the Notre Dame lamprophyre dikes, Pindell,J. L., S.C. Cande,W. C. PitmanIII, D. B. Rowley,J. F. Dewey, Newfoundland,and the openingof the NorthAtlantic Ocean, Can. J. J. LaBrecque, and W. Haxby, A plate-kinematicframework for Earth Sci., 16, 1823-1831, 1979. modelsof Caribbeanevolution, Tectonophysics, 155, 121-138,1988. Larochelie,A., andR. R. Black,An applicationof palcomagnetismin Playford,P. E., A. E. Cockbain,and G. H. Low, Geologyof the Perth estimatingthe ageof rocks,Nature, 198, 1260-1262,1963. Basin,Western Australia, Bull. Geol.Surv. West Aust., 124, 311 pp., Lawvet, L. A., I. W. D. Dalziel, and D. T. Sandwell,Antarctic Plate: 1976. DIVENERE ET AL.: PALEOMAGNETIC RESULTS FROM MARIE BYRD LAND 15,139

Rabinowitz, P. D., and J. LaBrecque, The Mesozoic south Atlantic and the early magmatichistory of Fuerteventura(Canary Island), J. Oceanand evolution of its continentalmargins, J. Geophys.Res., 84, Geophys.,46, 319-334, 1979. 5973-6002, 1979. Storey, B.C., The crustalblocks of West Antarctica within Gondwana: Ressetar,R., A. E. M. Nairn, and J. R. Monrad, Two phasesof reconstructionand break-up model, in Geological Evolution of Cretaceous-Tertiarymagmatism in the eastern of Egypt: Antarctica, edited by M. R. A. Thomson, J. A. Crame, and J. W. Palcomagnetic,chemical and K-Ar evidence, Tectonophysics,73, Thomson,pp. 587-592, Cambridge University Press, New York, 169-193, 1981. 1991. Robertson,W. A., Palcomagnetismof some Mesozoicintrusives and Storey, M., et al., Lower Cretaceous volcanic rocks on continental tuffsfrom eastern,Australia, J. Geophys.Res., 68, 2299-2312, 1963. marginsand their relationshipto the KerguelenPlateau, Proc. Ocean Robertson,W. A., and L. Hastie,A palaeomagneticstudy of the Cygnet Drill. ProgramSci. Results, vol. 120, 33-47, 1992. AlkalineComplex of Tasmania,Aust. J. EarthSci., 8, 259-268,1962. Stump,E. S., and P. G. Fitzgerald,Episodic uplift of the Transantarctic Royer,J.-Y., and M. F. Coffin, Jurassicto Eoceneplate tectonic recon- Mountains,Geology, 20, 161-164, 1992. structionsin the KerguelenPlateau region, Ocean Drill. Program, Swenson,P., and M. McWilliams, Paleomagneticresults from the Upper 120, 917-928, 1992. CretaceousMaudlow and Livingstonformations, southwest Montana, Royer,J.-Y., andD. T. Sandwell,Evolution of the EasternIndian Ocean Geophys.Res. Lett., 16, 669-672, 1989. since the Late Cretaceous:Constraints from Geosat altimetry, J. Tauxe, L., J. Besse,and J. LaBrecque,Palaeolatitudes from DSDP Leg Geophys.Res., 94, 13755-13782,1989. 73 sedimentcores: implications for the apparentpolar wander path Royer,J.-Y., P. Patriat,H. Bergh,and C. R. Scotese,Evolution of the for Africa during the late Mesozoic and Cenozoic, Geophys.J. R. SouthwestIndian Ridge from the Late Cretaceous(anomaly 34) to Astron. Soc., 73, 315-324, 1983. the middle Eocene(anomaly 20), Tectonophysics,155, 235-260, ten Brink, U.S., S. Bannister, B.C. Beaudoin, and T. A. Stern, 1988. Geophysicalinvestigations of the tectonic boundarybetween East Sakai, H., and B. Keating,Palcomagnetism of Leg 119 - holes737A, and West Antarctica, Science,261, 45-50, 1993. 738C, 745B, and746A, OceanDrill. Program,119, 751-770, 1991. Valencio, D. A., J. E. Mendla, and J. F. Vilas, Palaeomagnetismand K- Sandwell, D. T., and D.C. McAdoo, Marine gravity of the southern Ar age of Mesozoic and Cenozoicigneous rocks from Antarctica, ocean and Antarctic margin from Geosat,J. Geophys.Res., 93, Earth Planet. Sci. Lett., 45, 61-68, 1979. 10389-10396, 1988. Vandamme, D., V. Courtillot, J. Besse, and R. Montigny, Scharnberger,C. K., andL. Scharon,Palcomagnetism and plate tecton- Palcomagnetism and age determinations of the Deccan Traps ics of Antarctica,in AntarcticGeology and Geophysics,edited by R. (India): Resultsof a Nagpur-Bombaytraverse and review of earlier J. Adie,pp. 843-847,Universitetsfofiaget, Oslo, 1972. work, Rev. Geophys.,29, 159-190, 1991. Scharon, L., and I. C. Hsu, Palcomagneticinvestigation of some Van Fossen,M. C., and D. V. Kent, Palcomagnetismof 122 Ma plutons Arkansasalkalic igneousrocks, J. Geophys.Res., 74, 2774-2779, in New England and the Mid-Cretaceouspalcomagnetic field in 1969. North America:True polar wanderor large-scaledifferential manfie Schmidt,P. W., A new palaeomagneticinvestigation of Mesozoicig- motion?,J. Geophys.Res., 97, 19651-19661•1992. neousrocks in Australia,Teconophysics, 33, 1-33, 1976. Veevers, J. J. (Ed.), Phanerozoic Earth History of Australia, 418 pp., Schmidt,P. W., and B. J. J. Embleton,Magnetic overprintingin south- Clarendon, Oxford, 1984. easternAustralia and the thermal history of its rifted margin, J. Vugteveen,R. W., A. E. Barnes,and R. F. Butler, Paleomagnetismof Geophys.Res., 86, 3998-4008,1981. the Roskruge and Gringo Gulch volcanics, southeastArizona, J. Schmierer, K., and R. Burmester, PalcomagneticResults from the Geophys.Res., 86, 4021-4028, 1981. Cambro-OrdovicianBowers Supergroup,northern Victoria Land, Wade, F. A., and J. R. Wilbanks, Geologyof Marie Byrd and Ellsworth Antarctica,in GeologicalInvestigations in NorthernVictoria Land, Lands,in Antarctic Geologyand Geophysics,edited by R. J. Adie, Antarct. Res. Ser., vol. 46, edited by E. Stump, pp.69-90, AGU, pp. 207-214, Universitetsforlaget,Oslo, 1972. Washington,D.C., 1986. Watts, D. R., and A.M. Bramall, Palaeomagneticevidence for a dis- Schneider,D. A., and D. V. Kent, Testingmodels of the Tertiary palc- placedterrain in WesternAntarctica, Nature, 293, 638-641, 1981. omagneticfield, EarthPlanet. Sci. Lett., I01,260-271, 1990. Watts, D. R., G. C. Watts, and A.M. Bramall, Cretaceousand Early Schult, A., A. G. Hussain,and H. C. Soffel, Palcomagnetismof Upper Tertiary paleomagnetic results from the Antarctic Peninsula, Cretaceous volcanics and Nubian sandstonesof Wadi Natash, S. E. Tectonics,3, 333-346, 1984. Egyptand implicationfor the polar wanderpath for Africa in the Weaver, S. D., C. J. Adams, R. J. Pankhurst, and I. L. Gibson, Granites Mesozoic,J. Geophys.,50, 16-22, 1981. of EdwardVII Peninsula,Marie Byrd Land: Anorogenicmagmatism Stgoufin,J., Morphologicet structuredu canaldu Mozambique,th•se relatedto Antarctic-NewZealand rifting, Trans. R. Soc. Edinburgh d'ttat,236 pp., Univ. LouisPasteur, Strasbourg, France, 1980. Earth Sci., 83, 281-290, 1992. S•goufin, J., and P. Patriat, Existenced'anomalies m•sozoic dans le Wilch, T. I., D. R. Lux, G. H. Denton, and W. C. Mcintosh, Minimal bassin de Somalie. Implications pour les relations Afrique- Pliocene-Pleistocene uplift of the Dry Valleys sector of the Antarctique-Madagascar,C. R. Acad.Sc., 291, 85-88, 1980. TransantarcticMountains: A key parameterin ice-sheetreconstruc- Shive, P.N., and W. E. Frerichs,Palcomagnetism of the Niobrara tions, Geology,21, 841-844, 1993. Formationin Wyoming,Colorado, and Kansas, J. Geophys.Res., 79, Wilson, T. J., Mesozoic and Cenozoic kinematic evolution of the 3001-3007, 1974. TransantarcticMountains, in Recent Progress in Antarctic Earth Spbrli,K. B., and C. Craddock,Geology of the RuppertCoast, Marie Science,edited by Y. Yoshida, K. Kaminuma,and K. Shiraishi,pp. Byrd Land, Antarctica, in Gondwana Five, edited by M. M. 303-314, Terra Scientific,Tokyo, 1993. Cresswell,and P. Vella, pp. 243-250, A. A. Balkema, Rotterdam, Zijderveld, J. D. A., AC demagnetizationof rocks:An3alysis of results, Netherlands, 1981. in Methods in Palcomagnetism,edited by D. W. Collinson,K. M. SPRITE Group, and C. G. Boyer, The southernrim of the Pacific Creer,and S. K. Runcorn,pp. 172-191,Elsevier, New York, 1967. Ocean: Preliminary geologic report of the Amundsen Sea- BellingshousenSea cruiseof the Polar Sea, 12 February-21March I. W. D. Dalziel, Institute for Geophysics,University of Texas at 1992, Antarct. J. U.S., 27, ll-14, 1992. Austin,Austin, TX 78759-8435. (email:[email protected]). Steiger,R. H., andE. Jgger,Subcommission on Geology:Convention on V. J. DiVenere and D. V. Kent, Lamont-DohertyEarth Observatory, the use of decayconstants in geo- and cosmo-chronology, Earth P. O. Box 1000, Rt. 9W, Palisades, NY 10964. Planet. Sci. Lett., 36, 359-362, 1977. ([email protected]. edu). Stern,T. A., and U.S. ten Brink, Flexuraluplift of the Transantarctic Mountains,J. Geophys.Res., 94, 10,315-10,330,1989. (Received November 22, 1993; revised March 14, 1994; Storetvedt,K. M., A.M. Vage, S. Aase,and R. Lovlie, Palcomagnetism acceptedMarch 21, 1994.)