REVIEWS OF GEOPHYSICS, VOL. 25, NO. 5, PAGES 961-970, JUNE 1987 U.S. NATIONAL REPORT TO INTERNATIONAL UNION OF GEODESY AND GEOPHYSICS 1983-1986

POLAR WANDER AND PALEOMAGNETIC REFERENCE POLE CONTROVERSIES

D. V. Kent

Lamont-Doherty Geological Observatory and Department of Geological Sciences, Co ..... ia "-:---rsity .LUmU

S. R. May

Exxon ProdtC• ion Research

Introduct ion Precambrian

The purpose of this paper is to review the progress made over the past four years in studies of Phanerozoic and Precambrian polar wander, with an emphasis on the contributions made by U.S. scientists. We perceive a general theme embra- cing the notion that the accuracy of cratonic "reference poles" and therefore of apparent polar wander (APW) paths is not as good as was gener- ally believed four years ago. A renewed interest in research concerning cratonic reference poles is largely the result of paleomagnetic studies in orogenic belts, namely the western Cordillera and Appalachians of North America, the Andes of South America, and elsewhere which seek to document the existence or non-existence of displacements of suspect terranes (i.e., rotation and/or trans- lation) with respect to their presently asso- ciated cratons. (Some of the results of this research are reviewed in this volume by Hillhouse and McWilliams). An important realization of which paleomagnetists have long been aware but which was demonstrated recently by a number of sobering examples for both the Paleozoic and Mesozoic APW paths for North America, is that the reliability of concordance/discordance parameters that define terrane displacement is dependent not only on the precision and accuracy of the indivi- dual paleomagnetic study, but also on the pre- cision and accuracy of the appropriate cratonic reference pole.

In the previous quadrennial report, McWilliams [1983] accurately predicted that as a future trend (i.e., 1983-1986) "more and better reference data" would be needed "to clarify and refine" APW paths "to provide a more reliable data base for estimating terrane movements." While progress has been made toward this goal, the suggestion is still a valid assessment of future work. The following discussion will briefly review progress in three areas: new paleomagnetic reference poles, analysis and implications of revised APW paths, and true polar wander as analyzed through comparison of paleo- magnetic and hot-spot reference frames.

New Reference Poles

The following section covers work over the past 4 years on reference poles from Precambrian and Phanerozoic rocks from cratonic North America

as well as other major plates.

Precambrian paleomagnetic research is a daunting enterprise, faced with a tract of time much longer than the Phanerozoic, complex thermal or deformational histories, and the resulting strong possibility of multiple remagnetizations. It is therefore not surprising that progress in this area is slow but the questions are of such fundamental importance (e.g., how far back in the Precambrian was plate tectonics operative?) that new and improved techniques have been actively sought to answer them. Of particular signifi- cance has been the continuing application of sophisticated geochronological techniques to document with precision not only the crystal- lization age of a rock unit, but also the details of its cooling history that are so critical in interpreting the paleomagnetic record. A case in point is the recognition that paleopoles from the Grenville province represent magnetizations acquired during cooling and reheating episodes in post-orogenic time, perhaps as young as the Ordovician [Dunlop and Stirling, 1985]. On the other hand, the use of 40Ar/39Ar thermochrono- metric calibration of magnetization ages has enabled Onstott et al. [1984] to more con- clusively demonstrate relative motion between the West African and Guayana Shields, and possibly also with the Kalahari Shield, since 1.9-2.0 Ga. It seems clear that high precision thermochrono- metric analyses will be an essential adjunct to paleomagnetic work not only in the Precambrian, but in metamorphic/igneous terranes in general.

Examples of other contributions to our under- standing of Precambrian paleomagnetism are studies by McCabe and Van der Voo [1983) on late Keweenawan rocks, Geissman et al. [1982, 1983] and Tasillo et al. [1982] on Archeart rocks from the Abitibi orogen, and Kodama [1984] on Precambrian basement rocks under Kansas recovered

from boreholes.

Phanerozoic of North America

Crossing into the Phanerozoic, we find that the reality of the Cambrian APW loop of Watts et al. [1980] for North America is still highly uncertain and suspicion grows that several key paleopoles are seriously contaminated by later Paleozoic remagnetizations. For example, both Late Paleozoic and Cambrian directions have been

found in the Upper Cambrian Bonneterre Fro. [Wis- niowiecki et al., 1983; Dunn and Elmore, 1985]

Copyright 1987 by the American Geophysical Union. although no results frm the Bonneterre had previously been used for the Cambrian APW loop.

Paper number 7R0127. Considering the added possibilities of unrecog- 8755-1209/87/007R-0127515.00 nized tectonic rotations and inclusion of poles

961

962 Kent and May: Apparent and True Polar Wander

based on data of marginal quality, Larson et al. [1985] propose a highly simplified APW path from the Late Precambrian right through to the Triassic, essentially a 20o wide swath that in gross terms encompasses all these uncertainties. Ironically, the path is reminiscent of pole paths drawn in the early stages of paleomagnetic re- search some 30 years ago. An interesting insight into the different philosophical approaches used in the interpretation of reference poles, and the Cambrian data in particular, can be found in the discussion and reply by Van der Voo [1986] and Larson [1986].

Things become somewhat clearer for the Ordovi- cian and Silurian with the reports of new data from the Lower Ordovician Oneota Dolomite

[Jackson and Van der Voo, 1985] and the Middle Silurian Wabash Reef Limestone [McCabe et al., 1985]. The Oneota Dolomite gives a pole position at low latitudes, near to what Larson et al. [1985] would regard as the Early Cambrian segment of their simplified APW path. Obviously this cannot be accounted for by remagnetization of the Oneota, moreover, no paleopoles from rocks younger than Early Ordovician from North America are known to resemble it. Optimistically, the Oneota can at least be considered to cap the enigma of the Cambrian poles at the Early Ordovician.

The Wabash pole is supported by positive tilt and reversal tests, and coming from the stable interior (Indiana), must be regarded as one of the most reliable Paleozoic reference poles for North America now available. The Wabash gives about the steepest directions found in the ?aleozoic of North America, and places the mid- continent in the southern hemisphere at about 25os latitude. The paleopole for the C magnetization from the Cordova Gabbro [Dunlop and Stirling, 1985], dated as Late Ordovician (according to geologic time-scale of Palmer, 1983) by 40Ar/39Ar plateau ages of 446 Ma [Lopez- Martinez and York, 1983], is however somewhat out of spatial sequence, leap-frogging the Wabash and falling closer to some Late Silurian-Early Devonian poles.

The Devonian and Early Carboniferous poles for North America have undergone major revision in this quadrennial. Roy and Morris [1983] chal- lenged prevailing wisdom that placed North America straddling the paleoequator and suggested on indirect evidence that the available Late

Devonian-Early Carboniferous cratonic reference poles in fact represented Kiaman (Pertoo-Carboni- ferous) remagnetizations, even though positive fold tests and mixed polarities supported key poles from the Upper Devonian Catskill Fro. [Van der Voo et al., 1979] and the Lower Carboniferous Mauch Chunk Fro. [Knowles and Opdyke, 1968]. The relatively steep directions that pass a fold test in the Lower Carboniferous Deer Lake Group (which overlies Grenville-type basement in western New- foundland) more directly undermined confidence in the reliabililty of the middle Paleozoic cratonic reference poles [Irving and Strong, 1984].

Renewed paleomagnetic investigations of the Mauch Chunk [Kent and Opdyke, 1985] and the Cat- skill Fins. [Miller and Kent, 1986a, b] confirmed that the earlier results were indeed seriously contaminated by Permian remagnetizations. An interesting aspect of the pervasive remagnetiza-

tion now documented in the Appalachians is that it often (but not always; Perroud and Van der Voo, 1984) appears to be of syntectonic origin, that is, acquired sometime during Alleghanian folding. Syntectonic remagnetization is becoming more frequently reported in Appalachian rocks (e.g., from the Lower Devonian Helderberg Group, McCabe et al., 1983; Upper Silurian Bloomsburg Fro., Kent, 1986; Catskill Fro., Miller and Kent, 1986a, b; Mauch Chunk Fro., Kent and Opdyke, 1985], but the exact mechanism of acqui- sition is unclear. For the redbeds, a chemical event seems most likely because of the very high unblocking temperatures but for the limestones either a chemical [McCabe et al., 1983] or a thermoviscous origin [Kent, 1985] for remagneti- zation has been argued. Strain during defor- mation could conceivably play a role in interpre- tation of the fold test [Kligfield, et al., 1986; Kodama, 1986; Van der Pluijm, 1986; Lowrie et al., 1986].

Remagnetization however is not always total and characteristic components often have been recovered at the very high end of the demagneti- zation spectra of samples, especially in redbeds. These prefolding magnetizations place North America at more southerly paleolatitudes in the Late Devonian [Miller and Kent, 1986a, b; Van der Voo and McCabe, 1986] and Early Carboniferous [Kent and Opdyke, 1985] than previously thought, although not quite as southerly as the Newfound- land results would imply [Irving and Strong, 1984, 1985]. Better isolation of these charac- teristic magnetizations also reopens the possi- bility of detecting oroclinal bending in parts of the Appalachians. Previous analyses that discount paleomagnetic evidence for oroclinal bending in the Appalachians [Knowles and Opdyke, 1968; Schwartz and Van der Voo, 1983; Eldredge et al., 1985] were largely based on results now known to be heavily biased by secondary compo- nents. Declination anomalies that appear to vary more or less systematically around the Penn- sylvanian salient have now been reported from Silurian, Devonian and lower Carboniferous forma- tions [Kent, 1986; Miller and Kent, 1986b; Kent and Opdyke, 1985]. While interesting from a structural geology point of view, the possibility of rotations in the Appalachians, where many of the Paleozoic reference poles tend to be obtained, lends an additional element of com- plexity to the delineation of an APW path repre- sentative of the stable craton of North America.

Resolution of the problem will require reliable paleomagnetic data from more interior parts of North America.

For Late Carboniferous time and younger, there is a marked improvement in the quantity and quality of the data because more rocks suitable for paleomagnetic. study are available and the severity of diagenesis and deformation generally diminishes with decreasing age. New results on Carboniferous rocks from Nova Scotia [Scotese et al., 1984] and Lower Permian redbeds from Colorado [Miller and Opdyke, 1985] tend to corro- borate the well established Late Carboniferous

and Early Permian segment of the APW path. In- direct information on this interval comes from

the remagnetized directions from older rock units, especially completely remagnetized carbo- nates, as recently documented by Wisniowiecki et

Kent and May: Apparent and True Polar Wander 963

al. [1983], Chen and Schmidt [1984], Horton et al. [1984], McCabe et al. [1984], Stead and Kodama [1984], Elmore et al. [1985], and Kent [1985]. Relevant to the problem of a Pangea reconstruction discussed below, Van der Voo and McCabe [1985] suggest that these more recently determined paleopoles place the North American craton about 5o more northerly in the Pertoo- Carboniferous than the earlier ones, due to the more complete removal of secondary normal po- larity overprints. No new paleopoles from Upper Permian rocks have been published recently and it would be of interest to see if Late Permian poles are similarly affected.

Paleomagnetic results from Triassic rocks of cratonic North America have been published in the past quadrennial period by Herrero-Bervera and Helsley [1983], Shive et al. [1984], and Mcintosh et al. [1985]. Both Herrero-Bervera and Helsley and $hive et al. studied the paleomagnetism of the Red Peak Formation of the Chugwater Group in Wyoming. Both studies resulted in paleomagnetic poles which are virtually identical to each other and which help to constrain further the now well known Early Triassic reference pole for North America. Shive et al. [1984] suggest that dis- cordance between the Red Peak Chugwater pole and a pole from the Moenkopi Formation on the Colorado Plateau may reflect relative rotation between the Plateau and the basement uplifts of the Wyoming foreland. Gordon et al. [1984] and Steiner [1986], however, discuss age uncertain- ties and correlation problems of these non- marine, redbed units and suggest that the respec- tive poles may not be exactly comparable. As will be discussed below, the problem of demon- strating age equivalency between sedimentary rocks on and off of the Colorado Plateau is

important for evaluating the suggestion that there has been a small clockwise rotation of the

Plateau with respect to other parts of the craton.

The work of Mcintosh et al. [1985] concerns Late Triassic and Early Jurassic Newark Basin rocks from the Eastern United States. These

authors studied redbeds, from which they describe a polarity stratigraphy, as well as the Watchung basalts which, as described by previous workers, are of entirely normal polarity. Mcintosh et al. [1985] conclude that the redbed results are com- plicated by unremoved secondary magnetizations (i.e., they fail a reversal test), and that the Watchung results may fail to average secular variation. Therefore, no results from this study can be used to constrain Late Triassic-Early Jurassic APW. Van Fossen et al. [1986] have also reported results from Early Jurassic redbeds and basalts exposed in the Newark Basin. Failure of the fold test is interpreted by these workers to reflect possible local block rotation along the western limb of the Watchung Syncline although no field evidence could be found to demonstrate the

existence of the predicted structures. Two new reference poles have been obtained

from Middle and Late Jurassic volcanic rocks in southeast Arizona. The welded tuffs in the

Patagonia Mountains studied by May et a l. [1986] (172 Ma) and the welded tuffs in the Canelo Hills studied by Kluth et al. [1982] (151 Ma) represent parts of an autochthonous segment of the Cordil- leran Jurassic magmatic arc. The magnetic

properties of these rocks are relatively simple and yield well-determined poles. Jurassic rocks on the Laramide margin of the North American craton were also studied by McCabe et al. [1982] and Schwartz and Van der Voo [1984]. Samples for these studies, however, were collected from structural allochthons and can well represent rotations within the Idaho-Wyoming overthrust belt. For this and other reasons discussed more

fully by May and Butler [1986], the Twin Creek [McCabe et al., 1982] and Stump Formation [Schwartz and Van der Voo, 1984] poles are probably not reliable cratonic reference poles. Nevertheless, these and other studies [Eldredge and Van der Voo, 1987] illustrate an interesting and very useful application of paleomagnetism for understanding the buttressing effects of foreland blocks, and the associated rotation history of thrust sheets, within fold and thrust belts.

Our understanding of Early Tertiary North American APW has been improved by the work of Diehl et al. [1983]. These authors studied Paleocene and Eocene volcanic and intrusive rocks

in north-central Montana and report two new reference poles which are consistent with previous Early Tertiary results. These poles were interpreted to document an Early Tertiary APW stillstand following a period of rapid APW in the Late Cretaceous. Gordon [1984] calculated an alternative Paleocene reference pole based on a technique which allows declination-only data from other studies to be included in the calculation

of a mean pole. His Paleocene pole for North America is a few degrees more southerly than the Diehl et al. [1983] pole, indicating slightly more APW in the Early Tertiary.

Eurasia

For Hercynian Europe, new data for the Ordovician [Perroud et al., 1983; Perroud and Van der Voo, 1985] and Late Precambrian-Cambrian [Perigo et al., 1983] further document the Gondwana-affinities of the Armorica plate [Perroud et al., 1984]. Similarly high paleo- latitudes have also been found for the Late

Precambrian-Early Paleozoic of Avalon [Irving and Strong, 1985; Johnson and Van der Voo, 1985, 1986; Van der Voo and Johnson, 1985; Wu et al., 1985; Rao et al., 1986], suggesting that the Avalonian terrane in the Appalachians was part of Armorica and Gondwana during this time.

The difference in $iluro-Devonian poles between Baltica and Britain has long been problematical, explained by either remagnetiza- tion affecting the Baltica poles or a paleo- latitudinal separation between these regions. A renewed paleomagnetic investigation of the Lower Devonian Ringerike Sandstone in Norway reports dual polarity characteristic directions which pass a fold test [Douglass and Kent, 1986]. These results strengthen the case for a separation between Baltica and Britain, approximately along the Tornquist, prior to the Middle to Late Devonian.

The scarcity of Jurassic reference poles for Europe was addressed by Johnson et al. [1984] who report on the paleomagnetism of Oxfordian lime- stones from the Jura Mountains. The only other Late Jurassic paleopole for western Europe [Heller, 1977] is from similar limestones in

964 Kent and May: Apparent and True Polar Wander

southern Germany, however, the pole positions of these two studies lie some 10o apart. All but two of the "stable" sites reported by Johnson et al. are of normal polarity, yet the limestones sampled are believed to span the Oxfordian, at least the latter part of which is apparently a time interval of numerous polarity reversals [Kent and Gradstein, 1985]. Either the magneti- zation, although pre-folding and in magnetite, is a somewhat younger remagnetization, or the Oxfordian is actually mostly of normal polarity. These problems illustrate the need for further work on the Mesozoic APW path for Europe.

Across the continent in eastern Asia, it has been evident that China is made up of a number of cratonic elements, principally the North China (NCB) and South China (SCB) or Yangtze blocks, but very little was known about their relative motion history until fairly recently. Modern paleomagnetic work in China was initiated by McElhinny et al. [1981] who showed that the NCB and SCB plus other Asian blocks had not yet docked by the Late Permian. The paper by Lin et al. [1985] reports work on extensive collections of Phanerozoic rocks from throughout China and lays out in broad outline the APW paths for both the NCB and SCB. Unfortunately, few details of the paleomagnetic survey work are described in the paper but rapid progress is being made to document and refine these preliminary APW curves, especially from the SCB. Chan et al. [1984] report results from Upper Permian, Lower Triassic and Middle to Upper Triassic rocks that show low paleolatitudes for SCB in Pertoo-Triassic time. More extensive studies of Triassic rocks from

South China by Opdyke et al. [1986a] also indicate low paleolatitudes and support the argument that the final suturing of SCB with mainland Asia occurred sometime after the

Triassic. A Late Cretaceous paleopole from the northeastern part of South China agrees well with Eurasian reference poles [Kent et al., 1986a]. However preliminary results from Cretaceous sedi- ments from more southerly parts of South China are anomalous and may reflect deformation of Asia associated with the impingement of India in the Tertiary [Kent et al., 1986a; Chan and Zich, 19861.

Pervasive remagnetization has also been reported in paleomagnetic studies in both South .and North China [Kent et al., 1986b; Zhao and Coe, 1985] but in contrast to the Kiaman (Permian) age of remagnetization in the Appala- chians, the magnetizations of many of the Paleozoic formations in China, especially the carbonates, appear to have been reset in Late Mesozoic and Cenozoic times. More ancient remag- net izat ions may also be involved, for example, Cambrian directions reported by Lin et al. [1985] from South China are now seen to closely resemble Silurian directions that pass a fold test [Opdyke et al., 1986b].

Gondwana

Paleomagnetic work on the Gondwana continents has received renewed attention in this quadren- nial but the new Paleozoic data are problematical to say the least. The long-known salient feature of APW for an assembled Gondwana is that the

(South) pole must traverse somehow from a posi-

tion in the vicinity of northern Africa in the Early Paleozoic (corresponding for example to Ordovician glacial deposits in the Sahara), to a position around southern Africa by the Late Paleozoic (corresponding to the Gondwana glacia- tions). A pole from Lower to Middle Devonian redbeds in Mauritania falls off southern Africa

[Kent et al., 1984] and disagrees with the only other then reported Devonian pole from Africa (Msissi riorite, cited as Late Devonian; Hailwood, 1974] which plots in northern Africa. Hurley and Van der Voo [1986a] determined a Late Devonian pole from the Canning Basin of western Australia which agrees closely with the Msissi pole in a conventional Gondwana fit. A plausible interpre- tation of the Devonian data is that the Mauri-

tania pole represents a Carboniferous or younger remagnetization. However, the Msissi has recently been dated radiometrically as 135 Ma (late Jurassic or early Cretaceous) and the mag- netizations were found to be very complex [Salmon et al., 1986]. Although the agreement of the Canning Basin pole with the old Msissi pole must now be seen as fortuitous, the Canning Basin results which include a positive reversal test and a magnetostratigraphy [Hurley and Van der Voo, 1986b] still argue that the pole did not get to southern Africa until after the Devonian. But

then there are the paleomagnetic directions from ring-dikes in Niger, radiometrically dated as 400-440 Ma (nominally Silurian), which give a pole off southern Africa [Hargraves et al., 1986]. How all these and other Paleozoic paleo- poles from Gondwana fit into a coherent scheme of apparent polar wander is still far from clear. Current paleomagnetic work on the Paleozoic Cape Supergroup, South Africa, may shed some light on the dilemma [Bachtadse et al., 1986].

While Paleozoic poles for Gondwana have a direct bearing on the tectonic evolution of the Appalachian-Caledonide orogen and the formation of Pangea, precise Mesozoic and Cenozoic reference poles from the Gondwana continents are needed to evaluate tectonic rotations and dis-

placements in the Andean Cordillera and Alpine- Himalayan belt. A significant obstacle to the interpretation of paleomagnetic data being obtained from the Andean margin of South America [Irwin et al., 1985; Kent et al., 1985; Sheriff, 1985; Forsythe et al., 1986; May and Butler, 1985; Beck et al., 1986; Jesinsky et al., 1986] is the uncertainty concerning the reliability of the South American reference path. The available Mesozoic and Cenozoic reference poles, as listed for example in the compilation of Irving and Irving [1982], generally cluster close to the geographic pole and ostensibly show very little APW over this time. Fortunately, the fit and subsequent sea-floor spreading history between South America and at least Africa are well

established so it should be possible to transfer good qualilty poles from other continents into South American coordinates for any selected time interval. The recent critical analysis of African Late Mesozoic and Cenozoic poles, with the additional constraint of paleolatitudes derived from well-dated Deep Sea Drilling Project cores from the eastern South Atlantic [Tauxe et al., 1983], should be useful in this regard. We should also mention that tectonic conclusions

drawn from current paleomagnetic work on Mesozoic

Kent and May: Apparent and True Polar Wander 965

rocks in West Antarctica [Watts et al., 1984; Grunow et al., 1986, 1987] are also dependent on well-founded reference poles from East Antarctica as well as South America and Africa.

Analysis and Implications of Revised APW

Here we draw attention to developments in syntheses of the paleomagnetic reference pole data sets in terms of APW paths and their paleo- geographic and geodynamic implications.

Pre-Carboniferous APW

In a recent and comprehensive summary of Precambrian paleomagnetic data for North America, Roy [1983] observes that even though the data base for this craton is much greater than for any continent it is still difficult to draw a

meaningful APW path. He provides an interesting perspective on the problem by noting that over 2000 pole determinations would be required for North America alone (compared to the approxi- mately 350 available at his writing) to obtain a polar path of reliability and definition comparable to the APW path for the last 300 Ma for North America, even assuming that the continent was not an agglomeration of separate blocks over the Precambrian. It seems that the

testing of specific hypotheses of relative positions of cratons at well-defined times, based on precise thermochronometric dating of magneti- zations, is going to be the most practical and fruitful approach to the problem of the Pre- cambrian for the foreseeable future.

As outlined above, there have been significant changes over the past quadrennial in the inter- pretations of the paleomagnetic reference poles for the Paleozoic of North America, which have necessitated revisions of previously postulated paleocontinental reconstructions and their impli- cations, particularly for the Atlantic-bordering land masses. Unfortunately, convincing paleomag- net ic evidence is lacking for the involvement of cratonic North America in an Eocambrian super- continent [Bond et al., 1984; Van der Voo et al., 1984a] which would make a convenient starting point for Paleozoic (as well as Precambrian) paleocontinental reconstructions and tectonic analysis. But despite the problems in drawing an exact APW path through the Early Paleozoic data, virtually all the available poles place North America near to the paleoequator, with the possible exception of preliminary results from the Cambrian Caborca Sequence in Sonora, Mexico, which would suggest moderate paleolatitudes for the east coast of North America [Barr and Kirschvink, 1983]. In contrast, the Early Paleo- zoic poles for the Avalon terrane indicate high paleolatitudes and suggest that Avalon was separate from North America and more likely asso- ciated with Armorica and Gondwana. Differences

still exist in Late Silurian-Early Devonian poles from Baltica, Britain and North America [Van der Voo, 1983] but are no longer supported for the Late Devonian and Early Carboniferous due to the revisions in the North American cratonic

reference poles. In a synthesis of Silurian to Permian tectonic and paleomagnetic data for the Atlantic-bordering continents, Kent and Keppie [1986] therefore suggest that Euramerica, and

possibly even Pangea, formed during the Devonian, producing the late Caledonian/Scandian/Acadian/ Ligerian orogenic events. The alternative inter- pretations of the width of the Devonian ocean between Gondwana and Euramerica, due to the large uncertainties in the Gondwana poles, are illustrated in a set of global maps by Scotese [1984]. Briden et al. [1984, 1986] discuss further the paleomagnetic data base related to Paleozoic paleocontinental reconstructions of the Atlantic-bordering continents, including estimates of the width of the Iapetus Ocean and the scale and timing of motion on the Great Glen Fault.

Pangea Problem

A key ingredient in attempts at Phanerozoic paleocontinental reconstructions is a good model for Pangea, the supercontinent whose break-up history started in the Late Triassic. The time of formation of Pangea in the Paleozoic is still uncertain as it depends strongly on the problema- tical Paleozoic paleomagnetic data from Gondwana [see also Boucot and Gray, 1983], but while there is broad consensus that it had occurred by the Late Carboniferous there has been an ongoing debate regarding the exact configuration(s) of the supercontinent. There is first the familiar configuration of Wegener, Pangea A, quantita- tively fit by Bullard et al. [1965]. There are however disagreements among Carboniferous, Permian and Triassic paleopoles from Euramerica and Gondwana when these continents are assembled

in this fashion. Tectonic suggestions that have been made to minimize the disagreements involve clockwise rotation of Gondwana relative to the

northern continents, by amounts ranging from about 20o [Pangea A2, Van der Voo and French, 1974] to 35o [Pangea B, Irving, 1977], and even greater in the rather extreme Pangea C fit of Smith et al. [1980].

Irving [1983] reviewed the history of the Pangea fit problem and restated his case for Pangea B prior to the Jurassic. Van der Voo et al. [1984b] reevaluated the paleomagnetic data base for the critical time period from the Late Carboniferous until the earliest Triassic. They conclude that even though the data for the Late Carboniferous and Early Permian are in better agreement with the Pangea A2 fit, the Pangea B fit is also paleomagnetically permissible; for the far fewer paleopoles available for the Late Permian than the earlier intervals examined, Pangea B fits better. Livermore et al. [1986] suggest that the Pangea B and C are less likely alternatives, possibly resulting from inaccura- cies of pole ages. They favor a supercontinent resembling Pangea A2 which formed in the Late Devonian and subsequently evolved to the Pangea A configuration by the Late Triassic. More recently, Ballard et al. [1986] have shown that many of the Late Permian-Early Triassic paleomag- netic data from southern Africa that were incor-

porated in the various Pangea analyses represent post-Triassic remagnetizations. They suggest that the paleomagnetic data for the Late Permian- Early Triassic are insufficient for a valid test of the alternative Pangea reconstructions. In any case, it seems evident that the Pangea super- continent did not include large parts of eastern

966 Kent and May: Apparent and True Polar Wander

Asia which as mentioned above had yet to dock by the time the circum-Atlantic Pangea began to break up.

Windows, Cusps and Data Selectivity

The lack of a definitive paleomagnetic test of Pangea reconstructions notwithstanding, the Late Carboniferous (ca. 300 Ma) to Recent time interval is certainly represented by the best reference pole data set available, especially for North America. It has therefore been possible to apply more sophisticated analytical techniques to construct APW paths for this time interval.

The sliding time-window technique has proved very useful for illustrating the first-order features in APW and such analyses, especially by Irving and Irving [1982] for all the major conti- nents, have been widely used for reference paths. This method is designed to average out 'noise' contributed from paleopoles that are either more poorly determined or less well dated than others, giving a smoothed, empirical model of APW.

An important contribution to understanding plate motion information contained within APW paths was made by Gordon et al. [1984]. Expan- ding and quantifying previous models by Francheteau and $clater [1969] and Irving and Park [1972], they suggested that APW paths are composed of "tracks" which record long periods of constant direction plate motion. Tracks should therefore follow small-circle segments, each of which defines a paleomagnetic Euler pole (PEP) in the same way as hot-spot tracks and transform faults can be used to determine Euler poles. Successive tracks are joined at "cusps" which record plate reorganization events reflected as a change in the direction (and angular velocity) of plate motion.

Gordon et al. [1984] analyzed Carboniferous through Cretaceous poles for North America and recognized a Carboniferous - Late Triassic track, an Early Jurassic cusp, and a Jurassic through Early Cretaceous track. Based on a slightly different data base that also included a new

Middle Jurassic reference pole, Butler and May [1985] and May and Butler [1986] recognized the same Early Jurassic cusp which they labeled "Jl", but were able to statistically demonstrate a second cusp in the Late Jurassic labeled "J2". Both J1 and J2 cusps correlate temporally with known plate tectonic and intra-plate defor- mat ional event s.

As was illustrated by Gordon et al. and applied by May and Butler, the angular distance along the best fit small-circle, plotted as a function of age, reflects the angular velocity of a plate about the associated PEP. This type of angular velocity analysis can be interpreted as a test of the PEP corollary that not only do plates undergo rotat ion about fixed Euler poles for significant periods of time, but they do so with nearly constant angular velocity.

Gordon et al. [1984] found that the rms angular velocities for North America were similar to those calculated by Schult and Gordon [1984] using hot-spot tracks and sea-floor spreading data. May and Butler [1986] also found that the Early to Late Jurassic (J1-J2) and Late Jurassic to mid-Cretaceous (J2-K) APW tracks are well described by simple, unconstrained linear regres-

sions in angular velocity space. Such results argue strongly for the validity of the PEP model and support the suggestion by Gordon et a l. [1984] that the best fit model can then be used to generate synthetic reference poles for use in other analyses such as for comparison with data from suspect terranes.

It should however be noted that the rather

stringent data selection employed in PEP analysis provides a list of reference poles which may differ significantly from those generated with a "sliding-window" averaging technique. The philosophy of the latter allows somewhat less rigor in data selection, relying on the belief that the best estimation of a reference pole is obtained by averaging lots of data. May and Butler [1986] have shown that systematic biases among poles of questionable reliability have tended to displace Late Triassic and Jurassic North American reference poles toward the present geographic axis, thereby predicting anomalously high paleolatitudes for the craton. These diffe- rences lead to quite different estimates of the relative latitudinal motion of suspect terranes along the western Cordillera.

Colorado Plateau Rotation

A controversial topic involving North American APW which has come to light in the past quadren- nial period and of importance to PEP models is the question of whether or not the Colorado Plateau has undergone significant rotation with respect to other parts of cratonic North America [Hamilton, 1981]. Steiner [1986] argues that comparisons of individual poles from rocks of similar age on and off the Plateau indicate about 10o of clockwise rotation in post-Late Triassic time, probably reflecting intraplate deformation during Laramide compression and Rio Grande rifting. It has also been noted, however, by Gordon et al. [1984] that various Permian poles do not show a significant discordance. Bryan and Gordon [1986a] argued that "uncertainties in the precise age of characteristic magnetization and errors in paleomagnetic poles make comparisons of individual poles too inaccurate to discern small rotations". They present an alternative method for simultaneously comparing all available paleo- magnetic data from on and off the Plateau which is not dependent on demonstration of exact age equivalency. Their method evaluates the disper- sion of data generated by various hypothetical rotation cases with respect to the North American APW model of Gordon et al. [1984]. Bryan and Gordon [1986a] find a value of 3.9o clockwise rotation With a 95% confidence interval of 1.4-

6.6o. May and Butler [1986] evaluated the effect of the approximately 4o of clockwise Colorado Plateau rotation o.n the North American Jurassic

APW path and found that the same tracks and cusps were apparent in the PEP analysis, but that the "sharpness" of the J1 cusp was decreased and the age of this cusp became less distinct if this amount of Colorado Plateau rotation was assumed.

True Polar Wander

Research concerning true polar wander (TPW) has received continued interest by U.S. scientists. Two reference frames are commonly

Kent and May: Apparent and True Polar Wander 96?

compared to evaluate the possibility of TPW, namely the hot-spot and the time-averaged geomag- netic field. Hotspots are thought to track plate motions with respect to the mantle; the paleomag- net ic framework as defined for example by Jurdy and Van der Voo [1975] allows tracking of plate motions with respect to the dipole axis (i.e., the core). Since there does not appear to be convincing evidence for net motion of the plates with respect to the core or dipole axis, any differences between the hotspot and paleomagnetic reference frames ascribed to TPW may indicate that only the mantle moves or 'rolls' (e.g., Hargraves and Duncan, 1973).

Harrison and Lindh [1982b] compared the two reference frames using a global "mean" APW path based on paleomagnetic data from a number of plates combined with a relative motion model [Harrison and Lindh, 1982a], and the Morgan [1981] hot-spot model. Their general conclusion was that there has been approximately 20o of relative motion between the two reference frames

since 200 million years ago, and that this dis- placement has varied with time. Harrison and Lindh ascribe this discordance to a TPW model

wherein the hot-spot generating portion of the Earth's mantle rotates with respect to the spin axis (to which the geomagnetic field is fixed).

Andrews [1985] attempted a similar analysis, using a somewhat different global paleomagnetic data set that excluded the Pacific, and the relative motion and hot-spot models of Morgan [1983]. She concludes that there has been a relative displacement of 22 +_ 10o since 180 Ma, including episodes of both relatively rapid and relatively slow motions. Andrews interprets the motion of global mean paleomagnetic poles within a fixed hot-spot reference as TPW, due to a "shifting of the entire Earth in response to a change in the principal axes of the moment of inertia of the mantle."

An alternative TPW model was discussed by Davis and Solomon [1985] who suggest that a net torque on the lithosphere can be exerted both by ridge push and trench pull forces. When oriented properly, these forces are considered potentially significant enough to induce rotation of the global lithosphere with respect to the mantle.

Relatively rapid TPW during the late Neogene is compared by Andrews [1985] to astronomical observations for the last 75 years which document motion of the Earth's rotational axis at a rate

of ,about lo/m.y. If such a high rate was operative over a long time, then a relationship with processes other than lithospheric plate motion is implied. Schneider and Kent [1986] propose that Andrews [1985] overestimated the rate of Plio-Pleistocene TPW because of the

uncompensated effect of long-term non-dipole components. After correcting for these second and third order axisymmetric features of the geomagnetic field within the same data set as used by Andrews, Schneider and Kent find only 1.5o +_ 1.1o of TPW in the last 5 million years (i.e., significantly less than the 4o to 5 ø suggested by extrapolating ¾ecent astronomical data or by the uncorrected Plio-Pleistocene pa leomagnetic data).

Gordon [1983] compared paleomagnetic and hot- spot reference frames for Late Cretaceous and Early Tertiary data from the Pacific plate. He

found evidence for a rapid shift of the Pacific Ocean basin hot-spots with respect to the geomag- netic field between 90 and 81 Ma. Gordon

believed that this shift represented an event of TPW and that the temporal proximity with anoma- lous geomagnetic field behavior observed at the end of the Cretaceous normal polarity superchron (anomaly 34) might reflect a causal relationship between the motion of the hot-spot reference with respect to the spin axis and the end of a long lived geomagnetic field configuration. However, Ogg [1986] recently found that a relatively constant rate of latitudinal motion for the

Pacific plate is instead in better agreement with the Late Aptian through Early Tertiary paleomag- netic results from DSD? sediment cores. Sager [1983] used paleomagnetic data from DSDP cores on the Pacific plate in conjunction with seamount magnetic anomaly studies and sediment facies data to calculate a Late Eocene reference pole. One of the results of this work is that Late Eocene

Pacific plate pole is concordant with a synthetic 40 Ma pole predicted by various hot-spot constrained motion models. Sager concludes that there has been no significant displacement between the Pacific hot-spot and geomagnetic reference frames since the Late Eocene.

One of the assumptions in TPW studies such as those described above is that the hot-spots are relatively fixed with respect to the mantle and with respect to each other. Chase and Sprowl [1984] suggested that groups of hot-spots exhibit relative motion with respect to each other (North Pacific vs. South Pacific in their study) and that this motion may be associated with geoid anomalies which they interpret as related to convective phenomena in the mantle. If they are correct, then discordance between hot-spot and geomagnetic field reference frames need not necessarily reflect TPW. This emphasizes a larger problem with studies of the above type, namely the reliability of pre-anomaly 34 (approx. 84 Ma) hot-spot models, so that what appears as discordance between reference frames may simply reflect inaccuracies of either the pre-Late Cretaceous hot-spot tracks, or paleomagnetic reference poles.

Summary

Research concerning cratonic reference poles, particularly from North America but from other plates as well, has received increased, and some would argue overdue attention in the past quadrennial period. For the Mesozoic and Ceno- zoic this largely reflects the great interest in using paleomagnetic data to constrain motion histories of suspect terranes, coupled with the understanding that the resolution of tectonic models derived from such data are directly depen- dent on the reliability of cratonic reference poles. Fundamental problems still exist con- cerning major-plate reconstruct ions in the Paleozoic and there are correspondingly greater uncertainties in the paleomagnetic documentation of displacements of more ancient terranes. However, significant advances have been made in generating new high quality reference poles and at the same time in demonstrating the prevalance of remagnetization not fully appreciated in many of the earlier studies. For research concerning

968 Kent and May: Apparent and True Polar Wander

Precambrian APW, sophisticated chronometric techniques in close conjunction with detailed paleomagnetic studies are being successfully used to decipher complex thermal histories and thereby rejuvenating interest in this difficult but important area.

As more paleomagnetic data become available and the demands for higher precision and accuracy in the reference pole data set increase, more stringent and comprehensive reliability criteria or weighting schemes are being enacted. The current PEP models of APW for North America in

particular are predicated on the application of severe data selectivity. A strongly compensating factor is that the PEP conceptual model is pre- dictive, based on the critical assumption that APW should follow small-circle segments, in contrast to the highly empirical nature of con- ventional APW paths. Nevertheless, the question remains open as to which poles will be of suffi- cient quality to constitute a realistic test or a basis for refinement of the PEP models. While

there does not appear to be any satisfactory scheme of data selection free of subjectivity, we expect that the importance of data quality indices, based on multiple criteria such as recently applied to the North American paleomag- netic data base by Van der Voo [1987], to increase. Although such efforts at objective pass-fail or graded criteria may not meet with universal acceptance, at the very least they effectively define the quality level of work needed to address the important issues of the nature and significance of true and apparent polar wander that remain outstanding.

Acknowledgments. We thank Van der Voo and K. Kodama for useful comments on the manuscript and various authors for preprints and reprints of their work. Supported by the National Science Foundation, Division of Earth Sciences, Grant EAR85-07046 (DVK). Lamont-Doherty Geological Observatory Contribution •4074.

References

Andrews, J.•., True polar wander: an analysis of Cenozoic and Mesozoic paleomagnetic poles, J• Geophys. Res., 90, 7737-7750, 1985.

Bachtadse, V. and R. Van der Voo, Paleomagnetic evidence for crustal and thin-skinned

rotations in the European Mercyhides, Geo•h¾s. Res•Lett., 13, 161-164, 1986.

Bachtadse, V., R. Van der Voo and I.W. Maelbich, Further paleomagnetic results from the Paleozoic Cape Supergroup, western Cape Province, South Africa, (abs.), EO__S Trans. AGU, 67, 267, 1986.

Ballard, M., R. Van der Voo and I. W. Maelbich, Remagnetizations in Late Permian and Early Triassic rocks from southern Africa and their

implications for Pangea reconstructions, Earth and Planet. Sci. Lett. 79, 412-418, 1986.

Barr, T.D. and J.L. Kirschvink, The paleoposition of North America in the Early Paleozoic: New data from the Caborca Sequence in Sonora, Mexico, (abs.) EOS Trans. AGU 64, 689, 1983.

Beck, M.E. Jr., R.E. Drake and M.F. Butler, Paleomagnetism of Cretaceous volcanic rocks from central Chile and implications for the tectonics of the Andes, Geolozv, 14, 132-136, 1986.

Bogen, N.L., D.V. Kent and R.A. Schweickert, Paleomagnetism of Jurassic rocks in the western Sierra Nevada metamorphic belt and its bearing on the structural evolution of the Sierra Nevada block, J_•. Geophys. Res., 90, 4627-4638, 1985.

Bond, G.C., P.A. Nickeson, and M.A. Kominz, Breakup of a supercontinent between 625 MA and 555 MA: New evidence and implications for continental histories, Earth and Planet Sci. Lett., 70, 325-345, 1984.

Bouco.t•, A.J. and J. Gray, A Paleozoic Pangsea, Science, 222, 571-581, 1983.

Briden, J.C., D.V. Kent, P.L. Lapointe, R.A. Livermore, J.L. Roy, M.K. Seguin, A.G. Smith, R. Van der Voo and D.R. Watts, Paleomagnetic constraints on the evolution of the Caledonian-Appalachian Orogen, Geol. Soc. London, in press, 1986.

Briden, J. C., M. B. Turnell, and D. R. Watts, British paleomagnetism, Iapetus Ocean, and the Great Glen Fault, Geolozv, 12, 428-431, 1984.

Brown, P.M. and R. Van der Voo, A paleomagnetic study of Piedmont metamorphic rocks from northern Delaware, Geol. Soc. America Bull. 94, 815-822, 1983.

Bryan, P. and R.G. Gordon, Rotation of the Colorado Plateau: an analysis of paleomagnetic data, Tectonics, 5, 661-667, 1986a.

Bryan, P. and R.G. Gordon, Errors in minimum plate velocity determined from paleomagnetic data, J_•. Geophys. Res., 91, 462- 470, 1986b.

Bullard, E.C., J.E. Everett and A.G. Smith, A symposium on continental drift. IV. The fit of the continents around the Atlantic, Phil. Trans. Roy. Soc. London, A258, 41-51, 1965.

Burrett, C.F., Comment on '•aleozoic evolution of the Armorica plate on the basis of paleomagnetic data", Geology, 13, 380, 1986.

Butler, R.F. and S.R. May, Implications of a

revised Triassic-Jurassic APW path for western Cordilleran terrane displacement (abs.), E0S Trans. AG___•U, 66, 862, 1985.

Chan, L.S., C.¾. Wang and X.Y. Wu, Paleomagnetic results from some Permian-Triassic rocks from

Southwestern China, Geophys. Res. Lett. 11, 1157-1160, 1984.

Chan, L.S. and C. Zich, Paleomagnetism of Port Island Fm. (Cretaceous), Mong Kong: Implications for extrusion tectonics in Asia, (abs.), E0S Trans. AGU 67, 267, 1986.

Chase, C.G. and D.R. Sprowl, Proper motion of hotspots: Pacific Plate, EOS Trans. AGU, 65, 1099, 1984.

Chen, D. L. and V. A. Schmidt, Paleomagnetism of the Middle Mississippian Greenbrier Group in West Virginia, USA, in Plate Reconstruction from Paleozoic Paleomagnetis•., Van der Voo, R., C.R. Scotese and N. Bonhommet eds, AGU Geodynamics Series, 12, 48-62, 1984.

Cox, A. and M.G. Gordon, Paleolatitudes determined from paleomagnetic data from vertical cores, Rev• Geophys. Space Physics, 2_•2, 47-72, 1984.

Dankers, P., Implications of Early Devonian poles from the Canadian Arctic Archipelago for the North American apparent polar wander path, Can. J. Earth Sci., 19, 1802-1809, 1982.

Davis, D.M. and S.C. Solomon, True polar wander and plate- driving forces, J_•. Geophys. Res., 90, 1837-18441, 1985.

Diehl, J. F., M. E. Beck, S. Beske-Diehl, D. Jacobson and B.C. Hearn, Jr., Paleomagnetism of the Late Cretaceous-early Tertiary north- central Montana alkalic province, J• Geophys. Res., 88, 10,593-10,609, 1983.

Douglass, D.N. and D.V. Kent, Multicomponent magnetization of the Upper Silurian-Lower Devonian Ringerike Sandston, adjacent dikes and Permian lavas, Oslo, Norway, (abs.), EOS Trans. AGU, 67, ?67, 1986.

Dunlop, D. J. and J. M. Stirling, Post-tectonic magnetizations from the Cordova Gabbro, Ontario and Paleozoic reactivation in the

Grenville Province, Geoph¾s. J•Ro•. astr. Soc., 81, 521-550, 1985.

Dunn, W. J. and R. D. Elmore, Paleomagnetic and petrographic investigation of the Taum Sank Limestone, southeast Missouri, J__. Geophys. Res., 90, 11,469-11,483, 1985.

Eldredge, S., V. Bachtadse, and R. Van der Voo, Paleomagnetism and the orocline hypothesis, Tectonoph¾sics, 119, 153-179, 1985.

Eldredg e, S. and R. Van der Voo, Paleomagnetic study of thrust sheet rotations in the Melena and Wyoming salients of the northern Rocky Mountains, in Interaction of the Rocky Mountain Foreland and Cordilleran Thrust Belt Geol. Soc.

America Special Paper, Schmidt C. and W. J. Perry eds., in press, 1987.

Elmore, R. D., W. Dunn, and C. Peck, Absolute dating of a diagenetic event using paleomagnetic analysis, GeologY, 13, 558-561, 1985.

Forsythe, R.D., D.V. Kent, C. Mpodozis and J. Davidson, Paleomagnetism of Pertoo-Triassic

rocks, Central Chilean Andes, Proc. VI Gondwana Symposium, in press, 1986.

Francheteau, J. and J.G. Sclater, Paleomagnetism of the southern continents and plate tectonics, Earth Planet. Sci. Lett., 6, 93- 106, 1969.

Geissman, J.W., D.W. Strangway, A.M. Tasillo, and L.S. Jensen, Paleomagnetism and structural history of the Ghost Range intrusive complex, central Abitibi Belt, Ontario: further evidence for the Late Archean geomagnetic field of North America, Canad. J. Earth Sci. 19, 2085-2099, 1982.

Geissman, J.W., D.W. Strangway, A.M. Tasillo, and L. S. Jensen, Paleomagnetism of Late Archean metavolcanics and metasediments, Abitibi Orogen, Canada, Toleiites of the Kinojevis Group, Canad. J. Earth Sci., 20, 436-461, 1983.

Gordon, R. G., A Paleocene North American paleomagnetic pole incorporating declination- only data, Geophys. M•s. Lett., 11, 477-480, 1984.

Gordon, R.G., Late Cretaceous apparent polar wander of the Pacific plate: evidence for a rapid shift of the Pacific hotspots with respect to the spin axis, Geo•h¾s. Res. Lett. 10, 709-7 12, 1983.

Gordon, R. G., A. V. Cox, and S. O'Mare, Paleomagnetic Euler poles and the apparent polar wander and absolute motion of North America since the Carboniferous, Tectonics, 13, 499- 537, 1984.

Grunow, A., I. Dalziel and D.V. Kent, Ellsworth Mountains- Whitmore Mountains Crustal Block, Western Antarctica: New paleomagnetic results and their tectonic significance, Proc. VI Gondwana Symposium, in press, 1986.

Grunow, A., D.V. Kent and I.W.D. Dalziel, Evolution of the Weddell Sea Basin: New

paleomagnetic constraints, Nature, in press, 1987.

Hailwood, E.A., Palaeomagnetism of the Msissi Norite (Morocco) and the Palaeozoic reconstruction of Gondwanaland, Earth and Planet. Sci. Lett 23, 376-386, 1974.

Hamilton, W., Plate-tectonic mechanism of Laramide deformation, Contrib. Geol. 19, 87- 92, 1981.

Hargraves, R.B., E.M. Dawson and F.B. Van Houten, Paleomagnetism and age of mid Palaeozoic ring complexes in Niger, West Africa, and tectonic implications, Geophys. J. Roy. astr. Soc., in press, 1986.

HargraVes, R.B. and R.•. Duncan, Does the mantle roll?, Nature, 245, 361-363, 1973.

Marland, W.B., A.V. Cox, P.G. Llewellyn, C.A.G. Pickton, A.G. Smith and R. Walters, A-- GeoloKic Time Scale Cambridge University Press, 1982.

Harrison, C.G.•. and T. Lindh, A polar wandering curve for North America during the Mesozoic and Cenozoic, J. Geophys. Res., 87, 1903-1920, 1982a.

Harrison, C.G.•. and T. Lindh, Comparison between the hot spot and geomagnetic field reference frames, Nature, 300, 251-252, 1982b.

Kent and May: Apparent and True Polar Wander 969

Hartnady, C., Comment on "Paleozoic evolution of the Armorica plate on the basis of paleomagnetic data," Geology, 13, 589, 1985.

Heller, F., Palaeomagnetism of Upper Jurassic limestones from southern Germany, J_•. GeoDh¾s., 42, 475-488, 1977.

Herrero-Bervera, E. and C.E. Helsley, Paleomagnetism of a polarity transition in the Lower (?) Triassic Chugwater Formation, Wyoming, J_•. Geophys. Res., 88, 3506-3522, 1983.

Horton, R. A., J. W. Geissman and R. J. Tschauder, Paleomagnetism and rock magnetism of the Mississippian Leadville (carbonate) Formation and implications for the age of sub- regional dolomitization, Geophys. Res• Lett., 22, 649- 652, 1984.

Hurley, N. F. and R. Van der Voo, Paleomagnetism of Upper Devonian reefal limestones, Canning Basin, Western Australia, Geol. Soc. America Bull., in press, 1986a.

Hurley, N.F. and R. Van der Voo, Late Devonian magnetostratigraphy from a condensed limestone, Canning Basin, Western Australia, (abs.), EOS Trans. AGU 66, 265, 1986b.

Irving, E., Drift of the major continental blocks since the Devonian, Nature, 270, 304-309, 1977.

Irving, E., Fragmentation and assembly of the continents, Mid-Carboniferous to present, Geophys. Surv., 5, 299-333, 1983.

Irving, E. and GR. Irving, Apparent polar wander paths Carboniferous through Cenozoic and the assembly of Gondwana, Geophys. Surv.. 5, 141- 188, 1982.

Irving, E. and J.K. Park, Hairpins and superintervals, Can___•. J_•. E•rth Sci., 9, 1318- 1324, 1972.

Irving, E. and D. F. Strong, Palaeomagnetism of the Early Carboniferous Deer Lake Group, western Newfoundland: no evidence for mid-

Carboniferous displacement of "Acadia", E•rth and Planet. Sci. Lett., 69, 379-390, 1984.

irving, E. and D. F. Strong, Paleomagnetism of rocks from Burin Peninsula, Newfoundland: hypothesis of Late Paleozoic displacement of Acadia criticized, J_•. GeoDhvs. Res., 90, 1949- 1962, 1985.

Irwin, J.J., R.E. Drake, and W.D. Sharp, Some paleomagnetic constraints on the tectonic evolution of the coastal cordillera of central

Chile (abs.), EOS T•ns. AGU, 66, 866, 1985. Jackson, M. and R. Van der Voo, A Lower

Ordovician paleomagnetic pole from the 0neonta Dolomite, Upper Mississippi Valley, J• Geo•hvs. Res., 90, 10,449-10,461, 1985.

Jackson, t4. and R. Van der Voo, A paleomagnetic estimate of the age and thermal history of the Kentland, Indiana Cryptoexplosion structure, J. Geology. in press, 1986.

Jesinsky, C., R. Forsythe, C. Mpodozis and J. Davidson, Paleomagnetic results from Late Paleozoic sequences of the Atacama Dessert (Chilean Andes), (abs.), EO_•S Trans. AGU, 67, 270, 1986.

Johnson, R.J.E. and R. Van der Voo, Paleomagnetism of the Mid- Devonian Fisset Brook Formation, Cape Breton Island, Nova Scotia (abs.), EOS Trans. AG.__•U, 64, 216, 1983.

Johnson, R.J.E. and R. Van der Voo, Middle Cambrian paleomagnetism of the Avalon terrane in Cape Breton Island, Nova Scotia, Tectonics, 4, 629-651, 1985.

Johnson, R.J.E. and R. Van der Voo, Paleomagnetism of the Late Precambrian Fourchu Group, Cape Breton Island, Nova Scotia, Can. J• E•r;h Sci., in press, 1986.

Johnson, R.J.E., R. Van der Voo and W. Lowrie, Paleomagnetism and late diagenesis of Jurassic carbonates from the Jura mountains, Switzerland and France, Geol. $oc, Americ• Bull., 95. 478- 488, 1984.

Jurdy, D.M., and R.G. Gordon, Global plate motions relative to the hot spots 6& to 56 Ma, J•Geo•hvs. Res., 89, 9927-9936, 1985.

Jurdy, D.M. and R. Van der Voo, True polar wander since the Early Cretacetous, Science. 187, 1193-1196, 1975.

Kent, D.V., Paleocontinental Setting for the Catskill Delta, in D.L. Woodrow and W. Sevon, eds., C•skill Delta Symposium Volume. Geol. Soc, AmeT. Special Pa•er 201. 9-13, 1985.

Kent, D.V., Thermoviscous Remagnetization in some Appalachian Limestones, Geophysical Res. L•tters. 12, 805- 808, 1985.

Kent, D.V., Separation of prefolding and secondary magnetizations from the Bloomsburg Formation from the Southern Limb of the

Pennsylvanian Reentrant, (abs.), EOS TFans, AGU. 67, 270, 1986.

Kent, D.V., O. Dia and J. Sougy, Paleomagnetism of Lower-Middle Devonian Rocks and Upper Proterozoic-Cambrian(?) Rocks from Mejeria (Mauritania, West Africa), AGU Geodynamics Series, 12, American Geophysical Union, Washington, D.C., 99-115, 1984.

Kent, D.V., R.D. Forsythe, C. Mpodozis, and J. Davidson, Paleomagnetism of Pertoo-Triassic rocks, central Chile (abs.), EOS T•ns. AGU, 66, 865, 1985.

Kent, D.V. and F.M. Gradstein, A Cretaceous and

Jurassic Geochronology, Geol. $oc, Ame•. Bull., 9•6, 1419-1427, 1985.

Kent, D.V. and F.M. Gradstein, A Jurassic to Recent Chronology, in B.E. Tucholke and P.R. Vogt, eds., The Western North Atlan•ig Reiion. The Geology of North America, Geol. Soc. Amer., in press, 1986.

Kent, D.V. and J.D. Keppie, Silurian-Permian Paleocontinental Reconstructions and Tectonics, Geol. Soc. London in press, 1986.

Kent, D.V. and N.D. Opdyke, Multicomponent magnetizations from the Mississippian Mauch Chunk Formation of the Central Appalachians and their tectonic implications, J_•. Geo•hvs. Res.. 90, 5371-5383, 1985.

Kent, D.V., G. Xu, K. Huang, W.Y. Zhang and N.D. Opdyke, Paleomagnetism of Upper Cretaceous Rocks from South China, E•r•h Planet Sci. Lett., 79, 179-184, 1986a.

Kent, D.V., X. Zeng, W.Y. Zhang and N.D. Opdyke, Paleomagnetism of Paleozoic sedimentary rocks from Nanjing (Jiangsu Province) and Yichang (Hubei Province), South China, Tectonophysics, in press, 1986b.

Kligfield, R., W.B. Lowrie and A. Hirt, Deformation effects on Remanence in Rock: A

review, (abs.), EOS Tr.ans. AGU, 67, 268, 1986. Kluth, C.F., R.F. Butler, L.E. Harding, t4.

Shafiqullah and P.E. Damon, Paleomagnetism of Late Jurassic rocks in the northern Canelo

Hills, southeastern Arizona, J• Geophys. Res., 87, 7079- 7086, 1982.

Knowles, R.R. and N.D. Opdyke, Paleomagnetic results from the Mauch Chunk Formation: A test

of the origin of curvature in the folded Appalachians of Pennsylvania, J•. Geophys. Res., 73, 6515-6526, 1968.

Kodama, K.P., Palaeomagnetism of granitic intrusives from the Precambrian basement under

eastern Kansas: orienting drill cores using secondary magnetization components, Geophys. J• R__•. astr. Soc., 76, 273-287, 1984.

Kodama, K.P., The effect of deformation on the fold test, (abs.), EOS Trans. AGU 67, 268, 1986.

Larson, E.E., reply t_•o R. Van der Voo,"Petrolo- gic, paleomagnetic and structural evidence of a Paleozoic rift system in Oklahoma, New Mexic( Colorado, and Utah," Geol. Soc. America Bull. 97, 909, 1986.

Larson, E.E., P.E. Paterson, G. Curtis, R. Drake and F.E. Mutschler, Petrologic, paleomagnetic, and structural evidence of a Paleozoic rift

system in Oklahoma, New Mexico, Colorado, and Utah, Geol. Soc. America Bull. 96, 1364-1372, 1985.

Lin, J-L., t4. Fuller and W-Y. Zhang, 1985, Preliminary Phanerozoic polar paths for the North and South China blocks, Nature, 313. 444- 449, 1985.

Livermore, R.A., A.G. Smith and F.J. Vine, Late Paleozoic to early Mesozoic evolution of Pangea, Nature, 322, 162-165, 1986.

Lopez-Martinez, t4. and D. York, Further thermochronometric unravelling of the age and paleomagnetic record of the southwest Grenville Province, Can. J. Earth Sci. 20, 953-960, 1983

Lowrie, W., A.M. Hirt, and R. Kligfield, Effects of tectonic deformation on the remanent

magnetization of rocks, Tectonics, 5, 713-722, 1986.

Lynnes, C. S. and R. Van der Voo, Paleomagnetism of the Cambro- Ordovician McClure Mountain

alkalic complex, Colorado, Earth and Planet. Sci•Lett., 71, 163-172, 1984.

May, S.R. and R.F. Butler, Paleomagnetism of the Puente Piedra Formation, Central Peru, Earth Planet. Sci. Lett. 72, 205-218, 1985.

May, S.R. and R.F. Butler, North American Jurassic apparent polar wander: implications for plate motion, paleogeography and Cordilleras tectonics, J• Geophys. Res.. 91, 11,519-11,544, 1986.

May, S.R., R.F. Butler, t4. Shafiqullah and P.E. Damon, Paleomagnetism of Jurassic volcanic rocks in the Patagonia Mountains, southeastern Arizona: implications for the North American 170 Ma reference pole, J_•. Geophys. Res., 91, 11,545-11,555, 1986.

McCabe, C., and R. Van der Voo, Paleomagnetic results of the Late Keweenawan Chequamegon Sandstone and implications for red bed diagenesis and Late Precambrian apparent polar wander of North America, Can. J. Earth Sci.. 20, 105-112, 1983.

McCabe, C., R. Van der Voo, and t4. M. Ballard, Late Paleozoic remagnetization of the Trenton Limestone, Geophys. • Lett.. 11, 979-982, 1984.

McCabe, C., R. Van der Voo, D. R. Peacor, C. R. Scotese and R. Freeman, Diagenetic magnetite carries ancient yet secondary remanence in some Paleozoic sedimentary carbonates, Geolomv, 11, 221-223, 1983.

McCabe, C., R. Van Der Voo and B.H. Wilkinson, Paleomagnetic and rock magnetic results from the Twin Creek Formation (Middle Jurassic), Wyoming, Earth Planet. Sci. Lett 60, 140-146, 1982.

McCabe, C., R. Van der Voo, B. H. Wilkinson and K. Devaney, A Middle/Late Silurian paleomagnetic pole from limestone reefs of the Wabash Formation (Indiana, USA), J_•. Geophys. Res., 90, 2959-2965, 1985.

McElhinny, M.W., B.J.J. Embleton, X.H. Ma and XX. Zhang, Fragmentation of Asia in the Permian, Nature, 293, 212-216, 1981.

Mcintosh, W.C., R.B. Hargraves and C.L. West, Paleomagnetism and oxide mineralogy of Upper Triassic to Lower Jurassic red beds and basalts

in the Newark Basin, Geol. Soc. America Bull., 96, 463- 480, 1985.

McWilliams, t4.0., Paleomagnetism and the motion of large and small plates, Rev• Geophs. Space Phys., 21, 644-651, 1983.

Miller, J.D. and D.V. Kent, Synfolding and prefolding magnetizations in the Upper Devonian Catskill Formation of Eastern Pennsylvania, J_•. Geophys. Res., in press, 1986a.

Miller, J. and D.V. Kent, Paleomagnetism of the upper Devonian Catskill Formation from the southern limb of the Pennsylvania Salient: Possible evidence of oroclinal rotation, Geophys. Res. Lett. in press, 1986b.

Miller, J. D. and N.D. Opdyke, Magnetostrati- graphy of the Red Sandstone Creek section - Vail, Colorado, Geophys. Res. Lett 12, 133-136, 1985.

Morgan, W.J., Hotspot tracks and the opening of the Atlantic and Indian Oceans, in Thee Se•a, 7, Emiliani, C., ed., Wiley-Interscience, MY, 443- 487, 1981.

Morgan, W.J., Hot spot tracks and the early rifting of the Atlantic, Tectonophysics, 94, 123-139, 1983.

Morrison, J. and B. B. Ellwood, Paleomagnetism of Silurian- Ordovician sediments from the Valley and Ridge Province, northwest Georgia, Geophys. Res. Lett., 13, 189-192, 1986.

Murthy, G.S., Paleomagnetism of the Deadman's Bay diabase dikes from northeastern central

Newfoundland, Can. J. Earth Sci. 20, 195-205, 1983.

Murthy, G.S., Paleomagnetism of certain constituents of the Bay St. George Sub-basin, western Newfoundland, Phys. Earth Planet. Int., 39, 89-107, 1985.

Ogg, J.G., Paleolatitudes and magnetostratigraphy of Cretaceous and Lower Tertiary sedimentary rocks, DSDP site 585, Mariana Basin, western central Pacific, Init. Repts. DSDP, 89, in press, 1986.

Onstott, T.C. and J. Dorbor, 40Ar/39Ar and paleomagnetic results from Liberia and the Precambrian APW data base for the West African Shield, J•Africa Earth Sci., in press, 1986.

Onstott, T.C., R.B. Hargraves and P. Joubert, Constraints on the tectonic evolution of the

Namaqualand mobile belt II: Reconnaisance paleomagnetic and 40Ar/39Ar results from the Mamaqua, Kheis, and Natal provinces, Bull. Geol. Soc. S. Africa in press, 1986.

Onstott, T.C., R.B. Hargraves, P. Joubert and D. Reid, Constraints on the tectonic evolution of the Namaqualand mobile belt II: Paleomagnetic and 40Ar/39Ar results from the Haib- Vioolsdrif igneous complex, Bull. Geol. Soc. S•Africa, in press, 1986.

Onstott, T.C., R.B. Hargraves and D. Reid, Constraints on the tectonic evolution of the

Namaqualand mobile belt III: paleomagnetic nd 40Ar/39Ar results from the Gannakouriep dike swarm, Bull. Geol. Soc. S. Africa in press, 1986.

Onstott, T. C., R. B. Hargraves, D. York and C. Hall, Constraints on the motions of South American and African Shields during the Proterozoic: I. 40Ar/39Ar and paleomagnetic correlations between Venezuela and Liberia, Geol. Soc. America Bull 95, 1045-1054, 1984.

Opdyke, M.D., K. Huang, G. Xu, W.Y. Zhang, and D.V. Kent, Paleomagnetic Results from the Triassic of the Yangtze Platform, J_•. Geophys. Res., 91, 9553-9568, 1986a.

Opdyke, M.D., K. Huang, G. Xu, W. Y. Zhang and D.V. Kent, Paleomagnetic Results from the Silurian of the Yangtze Paraplatform, Tectonophysics, in press, 1986b.

Palmer, A.R., the Decade of North American Geology 1983 geologic time scale, Geology, 11, 503-504, 1983.

Perigo, R., R. Van der Voo, B. Auvray, and N. Bonhommet, Paleomagnetism of Late Precambrian- Cambrian volcanics and intrusives of the

Armorican Massif, France, Geophys. J• astron. Soc 75, 235-260, 1983.

Per•-ou•7 •.7 N. Bosh .... t, and R. Van der Voo, Paleomagnetism of the Ordovician dolerites of the Crozon peninsula, France, Geophys. J_•. astron. Soc 72, 307-314, 1983.

Perroud, H., and R. Van der Voo, Paleomagnetism of the Late Ordovician Thours Massif, Vendee province, France, J•Geophys. Res., 90, 2959- 2965, 1985.

Perroud, H, and R. Van der Voo, Secondary magnetizations from the Clinton-type iron ores of the Silurian Red Mountain Formation, Alabama, Earth Planet. Sci. Lett. 67, 391-399, 1984.

970 Kent and May: Apparent and True Polar Wander

Perroud, H., R. Van der Woo and N. Bonhommet, Paleozoic evolution of the Armorica plate on the basis of paleomagnetic data, Geology, 12, 579-582, 1984.

Rao, K. V., M.-K. Seguin and E. R. Deutsch, Paleomagnetism of Early Cambrian red beds on Cape Breton Island, Nova Scotia: evidence for a displaced Avalon zone, Can. J. Earth Sci. 23, in press, 1986.

Roy, J. L., Paleomagnetism of the North American Precambrian: a critical look at the data base, Precambrian Res.. 19, 319- 348, 1983.

Roy, J. L. and W. A. Morris, A review of paleomagnetic results from the Carboniferous of North America: the concept of Carboniferous geomagnetic field horizon markers, Earth and Planet. Sci. Lett., 65, 167-181, 1983.

Sager, W.W., A Late Eocene paleomagnetic pole for the Pacific plate, Earth. Planet. Sci. Lett., 63, 408-422, 1983.

Salmon, E., R. Montigny, J.B. Edel, A. Pique, R. Thuizat and M. Westphal, The Msissi Norite revisited: K/Ar dating, petrography and paleomagnetism, Geophys. Res. Lett. 13, 741-743, 1986.

Schneider, D.A. and D.V. Kent, Influence of non- dipole field on determination of Plio- Pleistocene true polar wander, Geophys. Res. Lett., 13, 471-474, 1986.

Schult, F.R.. and R.G. Gordon, Root mean square velocities of the continents with respect to the hot spots since the Early Jurassic, J• Geophys. Res., 89, 1789-1800, 1984.

Schwartz, S. Y. and R. Van der Voo, Paleomagnetic evaluation of the orocline hypothesis in the Central and Southern Appalachians, Geophys. Res. Lett., 10, 505-508, 1983.

Schwartz, S. Y. and R. Van der Voo, A paleomagnetic study of thrust sheet rotation during foreland impingement in the Wyoming- Idaho overthrust belt, J• Geophys. Res., • 10,077- 10,086, 1984.

Scotese, C.R., An introduction to this volume: Paleozoic paleomagnetism and the assembly of Pangea, in Plate Reconstruction fro• Paleozoic Paleoma•netism. Geodynamics Series. 12. Van der Voo, R., C.R. Scotese and N. Bonhommet, eds., 1-10, 1984.

Scotese, C.R., R. Van der Voo, R. J. Johnson and P. J. Giles, Paleomagnetic results from the Carboniferous of Nova Scotia, in Plate Reconstructions from Paleozoic Paleomainetism. Geodynamics Series 1__2, Van der Voo, R., C. R. Scotese and N. Bonhommet, eds., 63-81, 1984.

Scotese, C. R., R. Van der Voo and C. McCabe, Paleomagnetism of the Upper Silurian and Lower Devonian carbonates of New York State: evidence for secondary magnetizations residing in magnetite, Phys. Earth Planet. Int., 30, 385-395, 1983.

Seguin, N.-K. and E. Gahe, Paleomagnetism of Lower Devonian volcanics and Devonian dykes from northcentral New Brunswick, Canada, Phys. Earth Planet. Int. 38, 262-276, 1985.

Seguin, M.-K., A. Singh and L. Fyffe, New paleomagnetic data from Carboniferous volcanics and red beds from central New Brunswick, GeoDh¾s. Res. Lett. 1__2, 81-84, 1985.

Sheriff, S.D., Paleomagnetism of the Triassic Pucara Limestone, Peru (abs.), E0S Trans. AGU, 66, 865, 1985.

Shire, P. N., M. B. Steiner and D. T. Huycke, Magnetostratigraphy, paleomagnetism, and remanence acquisition in the Triassic Chugwater Formation of Wyoming, J• Geophys. Res., • 1801-1815, 1984.

Smith, A.G., A.LM. Burley and J.C. Briden,

Phane•9zoic Pal•9•ontinental Wo•!4 Ha•s., _ _

Cambridge Univ. Press, Cambridge, 1980. Spariosu, D.J. and D.V. Kent, Paleomagnetism of

the Lower Devonian Traveler Felsite and the

Acadian orogeny in the New England Appala- chians, Ge9 •. Soc. Amer. Bull.. 94, 1319-1328, •983.

Spariosu, D.J., D.V. Kent, and J.D. Keppie, Late Paleozoic motions of the Megurea Terrane, Nova Scotia: New paleomagnetic evidence, in Plate RecPns•ruct•pns from Paleozoic Palcomagnetism, American cleowhys. Union Geodvnamics Series. 12. Van der Voo, R., C. R. Scotese and N. Bonhommet eds., 82-98, 1984.

Stead, R. J. and K. P. Kodama, Paleomagnetism or the Cambrian rocks of the Great Valley of east- central Pennsylvania: fold test constraints on the age of magnetization, in Plate Reconstruct ions from P• P• leomagnet ism, Am•ricgn Geovhvs. Un•pn Geodvnamics Series, 12, Van der Voo R., C. R. Scotese and N. Bonhommet eds., 120-130, 1984.

Steiner, M.B., Mesozoic apparent polar wander and plate motions of North America, in Mesozoic PaleoReo•raDhv of West-central United States. Reynolds, M.W. and E.D. Dolly, eds., Rocky Mountain Sec. S.E.P.M., 1-11, 1983.

Steiner, M. B., Rotation of the Colorado Plateau, ?ec•uics, 5_• 649-660, 1986.

_

Tasillo, A.M., J.W. Geissman, D.W. Strangway, and L. S. Jensen, Paleomagnetism of Late Arcbean metavolcanics and metasediments, Abitibi Orogen, Canada: Volcanics of the Blake River Group and sediments and volcanics of the Timiskaming Group, Canad. J_•. Earth SC•. , 19. 3000-3014, 1982. •

Tauxe, L., J. Besse and J. L. LaBrecque, Paleolatitudes from DSDP Leg 73 sediment cores: implications for the apparent polar wander path for Africa during the late Mesozoic and Cenozoic, Geoph¾s. J•R_•. •str. S9c., 73, 315-324, 1983.

Van der PlutOn, V., S-•perimposed bulk homogeneous strain and the fold test, (abs.), EOSTrans. AGU, 67, 268, 1986.

Van der Voo, R., Paleomagnetic constraints on the assembly of the Old Red Continent, Tectonovhvsics. 91. 271-283, 1983.

Van der Voo, R., A plate tectonic model for the Paleozoic assembly of Pangea on the basis of paleomagnetic data, in Cpn;Fibu•ipns to the Tectonics and GeQphysics of M9•ntain C• Geol. Soc. Americ• Memoir 158, 19-23, 1983.

Van der Voo, R., Paleomagnetism of continental North America: The craton, its margins, and the Appalachian Belt, in Monograph o_p_n Geophysica• Framework of the Continental United States, Pakiser, L.C. and W. D. Mooney eds., Geol. Soc. America Memoir, in press, 1987.

Van der Voo, R., Triassic-Jurassic plate migrat ions and paleogeographic reconstruct ions in the Atlantic domain, in Triassic-Jurassic riftinK: North America and Africa American Assoc. of Petrol. Gepl. Memoir, Manspeizer, W., ed., in press, 1986.

Van der Voo, R., Discussion of "Petrologic, paleomagnetic and structural evidence of a Paleozoic rift system in Oklahoma, New Mexico, Colorado, and Utah," Geol. Soc. America Bull., 9--7, 908-909,1986.

Van der Voo, R. and R.B. French, Apparent polar wandering for the At lant ic-bordering continents: Late Carboniferous to Eocene, Earth-Sci. Rev., 10, 99-119, 1974.

Van der Voo, R., A.N. French and R.B. French, A paleomagnetic pole position from the folded

Upper Devonian Catskill redbeds, and its tectonic implication, Geology, 7, 345-348, 1979.

Van der Voo, R. and R.J.E. Johnson, Paleomagnetism of the Dunn Point Formation (Nova Scotia): high paleolatitudes for the Avalon terrane in the Late Ordovician, Geophys.Res. Lett., 12, 337-340, 1985.

Van der Voo, R. and C. McCabe, Implications of remagnetized limestone paleomagnetic poles for Late Paleozoic APW of cratonic North America

(abs.), EOS Trans. AGU 66, 875, 1985. Van der Voo, R. and C. McCabe, The age of the

Catskill magnetizations: Results of a conglomerate test, (abs.), EOS Trans. AGU, 67, 270, 1986.

Van der Voo, R., C. McCabe, and C. R. Scotese, Was Laurentia part of an Eocambrian supercontinent ?, in Plate Reconstruct ions from Paleozoic PaleomaKnetism, Geodynamics Series 12, 131-136, 1984a.

Van der Voo, R., J. Peinado and C. R. Scotese, A paleomagnetic teevaluation of Pangea reconstructions, in Plate Reconstructions from Paleozoic PaleomaKnetism. Geodynamics Series, 12, Van der Voo, R., C. R. Scotese and N. Bonhommet eds., 11-26, 1984b.

Van der Voo, R., H. Perroud and N. Bonhommet, Reply t__o Comment on 'Paleozoic evolution of the Arnofica plate on the basis of paleomagnetic data', Geolol¾, 13, 380-381, 1985a.

Van der Voo, R., H. Perroud and N. Bonhommet, Reply t__o Comment on 'Paleozoic evolution of the Armorica plate on the basis of paleomagnetic data', Geoloz¾, 13, 589-590, 1985b.

Van Fossen, M.C., J.J. Flynn and R.D. Forsythe, Paleomagnetism of Early Jurassic rocks, Watchung Mountains, Newark Basin: Evidence for complex rotations along the border fault, Geophys. Res. Lett., 13, 185-188, 1986.

Watts, D.R., G.C. Watts and A.M. Branall, Cretaceous and Early Tertiary paleomagnetic results from the Antarctic Peninsula, Tectonics 3, 333-346, 1984.

Watts, D.R., R. Van der Voo and S.C. Reeve, Cambrian paleomagnetism of the Llano Uplift, Texas, J. Geophys. Res., 85, 5316-5330, 1980.

Weisse, P. A., S. E. Haggerry and L. L. Brown, Paleomagnetism and magnetic mineralogy of the Nabant Gabbro and tonalire, eastern Massachusetts Can. J. Earth Sci., 22, 1425-1435, 1985.

Wisniowiecki, M., R. Van der Voo, C. McCabe, and W. C. Kelly, A Pennsylvanian paleomagnetic pole from the mineralized Late Cambrian Bonneterre

Formation, southeast Missouri, J__•. Geophys. Res., 88, 6540-6548, 1983.

Wu, F., R. J. Johnson and R. Van der Voo, Paleomagnetism of the Eocambrian Roxbury Formation, Boston Basin, Massachusetts (abs.), EOS Trans. AGU, 66, 257, 1985.

Zhao, X. and R. Coe, Further paleomagnetic results from eastern China, (abs.), EOS Trans. AG__•U, 66, 864, 1985.

Dennis V. Kent, Lamont-Doherty Geological Observatory, Palisades, NY 10964

Steven R. May, Exxon Production Research, P.O. Box 2189, Houston, TX 77252

(Received September 18, 1986; accepted February 2, 1987.)