Search

The deeply eroded Acraman structure in the Mesoproterozoic Gawler Range Volcanics, Gawler Craton, South Australia, ranks among the world’s largest known impact structures, with a transient crater ∼...

We use small-scale surveys in a three-axis, uniform field generating coil set (a Rubens cage) to emulate the Black Hill Norite magnetic field anomaly in South Australia. We then run further surveys with the source rotated to provide a vertical magnetization and in a vertical applied field to investigate the FFT reduction-to-pole (RTP) transform, which for this anomaly faces challenges arising from the low-inclination magnetization. Survey-based data imperfections are an important influence on performance of the RTP transform and this study provides a valuable extension of computer-based investigations. We show that striations in the RTP anomalies of both the measured and emulated BHN anomalies are due to FFT artefacts avoided in physical rotation of the magnetization to the vertical., Open-Access Online Publication: March 03, 2023

The Christmas Island Seamount Province is an extensive zone of volcanism in the Northeast Indian Ocean, consisting of numerous submerged seamounts and flat-topped guyots. Within this region lies two subaerial island groups, Christmas Island, and the Cocos Keeling archipelago. Christmas Island has experienced multiple episodes of volcanism that are exposed sporadically along its coastline. Here, we dated these volcanics using 40Ar/39Ar geochronology and analysed them for paleomagnetism. The oldest exposed volcanism occurred in the Eocene between 43 and 37 Ma. This is followed by a time gap of ~ 33 million years, before the eruption of a younger episode of Pliocene age (4.32 ± 0.17 Ma). It has, however, been suggested by previous workers that there is a much older Late Cretaceous event beneath the limestone which is unexposed. In addition, this study conducted the first paleomagnetic analysis of samples from Christmas Island to determine its paleoposition and the paleomagnetic polarity of the sampled sites. Two normal and two reversal magnetic events have been recorded, that agree with the geomagnetic reversal timescales. Late Eocene (38–39 Ma) palaeomagnetic data suggest a palaeolatitude of − 43.5°− 11.2°+ 9.0°, which is further south than palaeolatitudes (around 30° S) derived from existing plate reconstruction models for the Australian plate. However, the Late Eocene palaeomagnetic data are limited (only two sites) and secular variation may not have been averaged out. During the Pliocene (ca. 4 Ma) we estimate a palaeolatitude of approximately 13° S. The presence of the Late Eocene ages at Christmas Island correlates well with the cessation of spreading of the Wharton Ridge (~ 43 Ma), the initiation of spreading along the South East Indian Ridge, and the transit of Christmas Island over a broad low velocity zone in the upper mantle. This suggests that changes in stress regimes following the tectonic reorganisation of the region (prior to~ 43 Ma) may have allowed deeper-origin mantle melts to rise. Similarly, changes in the plate's stress regime at the flexural bulge of the Sunda–Java subduction zone may be implicated in renewed melting at ~ 4 Ma, suggesting that tectonic stresses have exerted a first-order effect on the timing and emplacement of volcanism at Christmas Island.

New paleomagnetic data for red beds (dolomitic siltstone and fine-grained sandstone) from the Angepena Formation of the late Cryogenian interglacial succession between deposits of the Sturt glaciation (ca 660 Ma) and Elatina glaciation (ca 635 Ma) in the Adelaide Geosyncline, South Australia, together with previous paleomagnetic data, indicate that the region lay in low paleolatitudes during the ca 25 Ma-long interglacial interval. Thermal demagnetisation of samples from the Angepena Formation at Depot Flat in the western Flinders Ranges and Nepouie Creek and Balcanoona Creek in the northeastern Flinders Ranges revealed a high-temperature component that decayed to the origin at ∼680°C, consistent with hematite being the only magnetic mineral present. Restoring the strata to the paleohorizontal yielded a positive fold test at the 99% confidence level. The concordance of results for widely separated sites and the positive fold test argue for the early acquisition of remanence before folding of the strata du...

Rocks from the Musgrave Block (Figure 1) are mainly composed of granulite facies quartzofeldspathic metasedimentary and meta-igneous rocks, but also include a suite of layered mafic to ultramafic intrusions, known as the Giles Complex. The Giles Complex comprises peridotites, pyroxenites and gabbronorites, which collectively form one of the largest suites of this type on Earth. The geophysical characteristics of the Giles Complex are poorly known, but have received renewed interest since discovery of the NeboBabel Ni-Cu-PGE deposit, which sits 130 km west of the NW corner of the study area, in Western Australia.

A magnetic body has a resultant magnetisation that is the vector sum of its induced and remanent components. Ignoring the role of remanence may lead to erroneous resultant magnetisation direction and hence inaccurate geophysical models. Determination of this resultant magnetisation direction may be obtained from total magnetic intensity data for a well separated magnetic anomaly. We present a new method to recover the resultant magnetisation direction that proceeds by iteratively calculating an approximate source layer, and cross correlating trial magnetisation directions. This method is tested with a number of remanent component directions for compact and elongate sources. For a compact source, the accuracy to which the correct resultant magnetisation direction can be recovered is generally found to be less than 5°. The method retains accuracy for low inclination resultant magnetisation directions, however, begins to lose sensitivity as the direction approaches the pole. Application of the method to the case study of the Black Hill Norite recovers resultant magnetisation directions in agreement with paleomagnetic results. Here the resultant directions from the two methods are found to be consistent, with minor differences possibly due to limited paleomagnetic sampling of magnetisation directions that changed as the intrusion cooled. A higher bulk Q value, than found from the limited sampling, is suggested for one of the three anomalies studied, as supported by the results of other authors.

Magnetic field data forms the most detailed and comprehensive geophysical coverage of Australia but at present there are challenges in reliable interpretation of that data. One major concern is the recognition and correct treatment of remanent magnetization which can result in mislocation of drill targets. To both reduce hazards of incorrectly interpreting magnetic field data and also to derive new geological information we have developed a web-delivered database of Australian magnetic anomalies recognised as being at least in part due to remanent magnetization. The database is linked with databases of model studies, magnetic moment analyses (MMA) and palaeomagnetic and rock magnetic investigations. These linked databases presented in the context of magnetic field imagery provide geoscientists with a rapid means of recognising the expression of remanent magnetization and of accessing information about that magnetization to assist in their interpretations. The databases have been lightly populated and can be accessed with a web utility. Community involvement is sought to further populate the databases with information currently distributed through geological survey, company and academic records.

The Mesoproterozoic Pandurra Formation and Blue Range Beds of the Gawler Craton, South Australia, are undeformed, low-dipping (typically 10–15°) arenites of fluvial origin displaying pale to moderate red colours with light grey mottles and streaks and local Liesegang banding. Eighty-nine core samples and seven block samples were collected from 18 sites in the Pandurra Formation at Whyalla and 70 core samples from 16 sites in the Blue Range Beds at Talia. Specimens from all samples were stepwise thermally demagnetised, revealing hematite as the dominant magnetic carrier with a Neel temperature of 680°C. The mean direction for the Pandurra Formation after tilt-correction (N = 90 specimens) is declination = 248.2°, inclination = 46.7° (α95 = 4.4°), indicating a paleolatitude of 27.9 +3.9/–3.4° and a pole position at latitude 33.6°S, longitude 64.5°E (A 95 = 4.9°). The mean direction for the Blue Range Beds after tilt-correction (N = 61) is declination = 233.9°, inclination = 50.2° (α95 = 4.0°), with a paleol...

The ∼1590 Ma Gawler Range Volcanics (GRV) and 1595–1575 Ma Hiltaba Suite Granitoids (HSG), from the central Gawler Craton of South Australia, can be divided into an oxidised Roxby Downs-type containing hematite and magnetite, and a less oxidised Kokatha-type that commonly contains ilmenite and titanomagnetite. In both types the predominant magnetic phase is pure end-member magnetite, except for a minor amount of titanomagnetite present in the Kokatha-type Black Yardea Dacite. The characteristic remanence observed in the GRV is directed upward to the northeast. A fold test is strongly negative, indicating that this characteristic remanence is a post-folding overprint although the age of folding is uncertain. Rock magnetic properties identify magnetically unstable multidomain magnetite particles. The magnetisation has been blocked as the crust cooled from above multidomain (MD) unblocking temperatures probably during a mid-phase of the Alice Springs Orogeny (ASO). The in situ mean direction, not corrected f...

Airborne magnetic surveys have improved dramatically over the past three decades with advances in both data acquisition and image processing techniques. Magnetic surveys form an integral part of exploration programs and are now routinely undertaken before geological mapping programs. These advances have been made despite treating the magnetic field as a scalar, wherein various processing procedures that assume a potential field are compromised. If the vector information could be retrieved, either by direct measurement or by mathematical manipulation, magnetic surveys could be improved even further. For instance, the total magnetic intensity (TMI) could be corrected so it represents a true potential field.

Potential field analysis, with focus on magnetic interpretation is ideally suited to geological problems where the targets are deep, totally undercover, or when there is little or no outcrop available for mapping or sampling. In these scenarios understanding the location, orientation and relationships between deep geological structures is the key to developing a clear picture of an areas tectonic history. In this project, we conduct a comprehensive magnetic analysis of the Lachlan Orocline and Macquarie Arc, both located in the Lachlan Orogen, eastern Australia to test the potential relative rotation of this block to the surrounding units. These results are compared with palaeomagnetic data from the region and a series of geologically constrained crustal-scale forward models of lithospheric magnetisation for the project area, with focus on the large-scale structural components of the Lachlan Orocline. The forward model method treats each geological unit present in both the Macquarie Arc itself and the surrounding areas as independent parameterised stratigraphic units. Using a Geographical Information System (GIS) approach, known geological, structural and physical properties are used to produce a series of vertically integrated value grids. These grids are then used as input to model the regional magnetisation in the global scale magnetic field forward model.

SUMMARY Palaeomagnetic results from late Middle Cambrian‐Early Ordovician carbonate sequences sampled at Black Mountain (Mt Unbunmaroo), Mt Datson and near Chatsworth Station (southeastern Georgina Basin) are presented. A palaeomagnetic reassessment of these carbonates was designed in an effort to constrain regional magnetization ages as results from an earlier study, conducted at Mt Unbunmaroo, play a pivotal role in a proposed Cambrian inertial interchange true polar wander (IITPW) event. Remanent magnetizations within these carbonates were found to be variably developed with most specimens displaying two of the five isolated components. Component PF, for which goethite is the identified remanence carrier, is thought to reflect a chemical remanent magnetization of recent origin. Component TR, held by haematite, has a palaeomagnetic pole consistent with the Tertiary segment of Australia’s apparent polar wander path (APWP) and most probably was acquired as a consequence of prolonged weathering during this period. The A component has a palaeomagnetic pole at 54.7 ◦ S, 262.3 ◦ E( dp = 2.3 ◦ , dm = 4.5 ◦ ) after unfolding. This direction, constrained by positive fold and reversal test statistics, is consistent with Australia’s Early Devonian APWP, perhaps reflecting a remagnetization event associated with the intracratonic Alice Springs Orogeny. A Late Ordovician‐Early Silurian remanence, component B, is also described; with 100 per cent unfolding the associated palaeopole lies at 8.0 ◦ S, 216.8 ◦ E( dp = 2.6 ◦ , dm = 5.1 ◦ ). A third Palaeozoic, and presumed primary or early diagenetic, component, C, also passes applied fold and reversal tests and has a palaeomagnetic pole at 48.6 ◦ N, 186.0 ◦ E( dp= 2.1 ◦ S, dm = 4.0 ◦ ). This palaeopole is dissimilar from younger magnetizations, is consistent with Cambrian poles from other parts of cratonic Australia and falls within a cluster of Middle‐Late Cambrian (515‐500 Ma) palaeopoles from other Gondwanan continents. The age attributed to the palaeopole associated with the C component, ∼510 Ma, provides a tight constraint on the younger boundary of the proposed Cambrian IITPW event and its agreement with other Gondwanan palaeopoles is incompatible with the IITPW hypothesis. Components A, B and C are analogous to palaeomagnetic results reported in the earlier investigation of this region, and a comparison of results from the two studies, coupled with rigorous statistical analyses of the new findings, is presented.

Palaeomagnetic data are used to study the configurations of continents during the Proterozoic. Applying stringent reliability criteria, the positions of the continents at 12 times in the 2.45- to 1.00-Ga period have been constructed. The continents lie predominantly in low to intermediate latitudes. The sedimentological indicators of palaeoclimate are generally consistent with the palaeomagnetic latitudes, with the exception of the Early Proterozoic, when low latitude glaciations took place on several continents. The Proterozoic continental configurations are generally in agreement with current geological models of the evolution of the continents. The data suggest that three large continental landmasses existed during the Proterozoic. The oldest one is the Neoarchaean Kenorland, which comprised at least Laurentia, Baltica, Australia and the Kalahari craton. The protracted breakup of Kenorland during the 2.45- to 2.10-Ga interval is manifested by mafic dykes and sedimentary rift-basins on many continents. The second 'supercontinental' landmass is Hudsonland (also known as Columbia). On the basis of purely palaeomagnetic data, this supercontinent consisted of Laurentia, Baltica, Ukraine, Amazonia and Australia and perhaps also Siberia, North China and Kalahari. Hudsonland existed from 1.83 to ca. 1.50-1.25 Ga. The youngest assembly is the Neoproterozoic supercontinent of Rodinia, which was formed by continent-continent collisions during similar to 1.10-1.00 Ga and which involved most of the continents. A new model for its assembly and configuration is presented, which suggests that multiple Grenvillian age collisions took place during 1.10-1.00 Ga. The configurations of Kenorland, Hudsonland and Rodinia depart from each other and also from the Pangaea assembly. The tectonic styles of their amalgamation are also different reflecting probable changes in sizes and thicknesses of the cratonic blocks as well as changes in the thermal conditions of the mantle through time. (C) 2003 Elsevier B.V. All rights reserved.

Magnetic tensor gradiometry provides gradient components of true potential fields which enables unique depth estimates and improves analytic signal methods as well as providing a number of other advantages. A high temperature SQUID (HTS) gradiometer can provide measurements of the components of the earth's field tensor creating a new tool for mineral exploration. A successful comparison between a HTS SQUID gradiometer and a Cs-vapour gradiometer under survey conditions has been conducted. Both instruments were configured vertically. The HTS gradiometer measured the B/sub zz/ component of the gradient tensor, while the Cs-vapor gradiometer measured the vertical gradient of the total magnetic intensity. The HTS gradient measurement was the difference in output between two coaxial SQUID sensors. Effective noise levels achieved were 0.16-0.3 nT/m RMS, compared with 0.1-0.5 nT/m RMS for the Cs-vapor system. The SQUID noise was dominated by vibration with additional contributions from the multiplexed sampling between the SQUIDs. This paper reports on the system development, design issues, trial results and the implications for geophysical exploration.

The results of rock magnetic, thermal demagnetization and Thellier palaeointensity studies are presented for two high-level intrusions from southern New South Wales, Australia. The Gingenbullen Dolerite (GB) (34.5°S, 150.3°E) and the Gibraltar Microsyenite (GS) (34.5°S, 150.4°E) were emplaced during the second of three major phases of igneous activity that affected the southern Sydney Basin and are K/Ar dated at 172 and 178 Ma, respectively. The magnetic mineralogy of the two intrusions is different: the GB is dominated by single-domain (SD)/pseudo-single-domain grains of magnetite, whereas the GS has both SD magnetite and haematite components, although the haematite component does not record a stable remanence. The GB records two opposed components of magnetization, above 100 °C, occupying distinct parts of the blocking temperature spectrum. The lower blocking temperature (LBT) component resides between 150 and 400 °C and is of normal polarity, having a mean direction of D/I = 021°/−79° (α95 = 8°, k = 63), with a corresponding VGP at Lat/Long = 54°S/137°E (dp = 12, dm = 13). This component is believed to be a TCRM or TVRM related to the initial opening of the Tasman Sea at ~90 Ma. The higher blocking temperature (HBT) ChRM component is recognized between 450 and 580 °C and is of reverse polarity, with a mean direction of D/I = 174°/+81° (α95 = 11°, k = 33) and a corresponding VGP at Lat/Long = 52°S, 153°E (dp = 17, dm = 18). The ChRM for the GS resides in the 100–450 °C region and, although the scatter is high, has a similar mean direction (D/I = 167°/+79°; α95 = 24°, k = 26) to the HBT component of the GB, with a pole position at Lat/Long = 54°S/158°E (dp = 30, dm = 31). There is no evidence of the GB LBT component in samples from the GS, suggesting that this intrusion was magnetically unaffected by the later event, which is also seen for other intrusions in the Sydney Basin. The VGPs calculated from the ChRM directions both plot on the Jurassic segment of the Australian APWP (Dunlop et al. 1997), providing further palaeomagnetic confirmation for the K/Ar ages. Thellier palaeointensity results were of variable quality, as expected for rocks of this age. The mean VDM values calculated for the ChRMs of the GB and GS were 1.3 × 1022 and 4.1 × 1022 Am2, respectively (approximately 16 and 51 per cent of the present value). On application of a cooling-rate correction estimate, the GB result falls to a minimum of 0.87 × 1022 Am2 (approximately 11 per cent of the present-day value), representing an overestimate of up to 56 per cent. These values, arising from a time window previously lacking in Thellier estimates, provide further confirmation of the existence of a pre-Cenozoic dipole low. The VDM values derived from the overprint in the GB samples yield an average of 4.8 × 1022 Am2 (from a range between 44 and 81 per cent of the present-day mean VDM) and are tentatively attributed to the Cretaceous Normal Superchron (CNS). These results may document evidence for a previously unrecognized low dipole moment during the bulk of the CNS, a possibility that requires further investigation by performing palaeointensity work on igneous rocks with radiometric ages between 110 and 90 Myr that carry unequivocal primary TRMs.

Paleotemperature controls the maturation of coal and hydrocarbons in sedimentary basins and is also important in determining paleogeothermal gradient and hence tectonic style in exhumed metamorphic terrains. One method of estimating paleotemperature analyses the partial remagnetization of a rock due to heating in thick volcanic or sedimentary sequences, over subcrustal heat sources such as plumes, or at convergent plate margins. The overprinted natural remanent magnetization (NRM) of a rock records both the age and the paleotemperature of remagnetization, but a temperature correction from laboratory to geological time scales is required, using theoretical time–temperature relations. Time–temperature relations are well known for magnetite (Fe 3 O 4 ) but are reported here for the first time for pyrrhotite (Fe 7 S 8 ), another common NRM carrier. Data for each mineral separately yield independent estimates of paleotemperature if geologically reasonable estimates of heating time can be made. Paleotemperature can be estimated without geological input if data for both minerals are combined. Together with the age of remagnetization, determined from the paleomagnetic pole of the NRM overprint, these paleotemperature estimates can be used to infer the history of heating and uplift following burial. As a test case, we examine thermally acquired NRM overprints carried by pyrrhotite (Fe 7 S 8 ) and magnetite (Fe 3 O 4 ) in the Milton Monzonite of southeastern Australia. These overprints record a heating event about 100 Ma ago, probably thermal doming prior to rifting of the Tasman Sea, that upgraded coal rank in the Sydney Basin. Extrapolating from laboratory to geological times, using the new time–temperature contours for pyrrhotite, we estimate that the presently exposed Sydney Basin in the vicinity of the Milton Monzonite was remagnetized by heating to 165±30°C for ≈100 ka. Assuming a paleogeothermal gradient of 70°C/km appropriate for young or incipient rifts, the depth of burial at the time of remagnetization is estimated to have been 2.3±0.4 km. This figure is in excellent agreement with independent estimates based on reflectance data for the coal accessory mineral vitrinite.

The Early Triassic (∼245 Ma) Milton Monzonite of the Sydney Basin, Australia, has four distinct components of natural remanent magnetization (NRM) with only slightly overlapping ranges of unblocking temperatures. The low-temperature (LT) component, the first to be thermally demagnetized, is thought to be a Late Cretaceous (≈100 Ma) thermoviscous overprint acquired in slow cooling during uplift. The high-temperature (HT) component, the second to be demagnetized, is probably the primary thermoremanent magnetization (TRM) of the Milton intrusion but could possibly be a Jurassic overprint. LT and HT are usually carried by magnetite and occasionally by pyrrhotite. Samples from nine sites have a further NRM component which unblocks at higher temperatures than HT but below the magnetite Curie temperature of 580°C. This component is argued to be a chemical remanent magnetization (CRM) because of its discrete range of high unblocking temperatures, above those of the thermal components HT and LT, and is called CRM1. CRM1 has almost the same direction as LT and is likely carried by authigenic magnetite produced during uplift ∼100 Ma. Samples from five sites have a fourth NRM component, with a direction resembling that of HT but carried by hematite. This fourth component could be a primary TRM but is more likely a CRM and is therefore called CRM2. The HT-CRM2 mean direction is D=50°, I=75.5°, defining a paleopole at 16°S, 172°E. The HT-CRM2 paleopole falls near 150 Ma on the Australian apparent polar wander path but is a considerable distance from paleopoles of Permian and Early Triassic age. There is no known tectonic or other remagnetizing event in the Sydney Basin around 150 Ma. For this reason, we propose that the HT-CRM2 paleopole defines a new Triassic segment of the Australian polar wander path. The LT-CRM1 mean direction is D=348°, I=−79°, with a paleopole falling at 56°S, 1580°E, near 100 Ma on the polar wander path. This age is consistent with uplift and cooling related to initial rifting of the Tasman Sea.

The Milton Monzonite of southeastern Australia was thermoviscously remagnetized as a result of Cretaceous burial and uplift. Thermal demagnetization separates the low unblocking temperature (LT) overprint from the high unblocking temperature (HT) primary remanence, with a relatively sharp junction between LT and HT components in vector projections. For single-domain grains, the junction temperature TL between two such vectors corresponds to the maximum blocking temperature Tr reactivated in nature, apart from a correction for the difference between natural and laboratory timescales. However, measured TL values are distributed over an implausibly wide range (>250°C) for burial remagnetization of an untilted intrusion like the Milton Monzonite. Furthermore, many TL values are anomalously high compared to the predictions of single-domain theory. Multidomain grains are the cause of these anomalies. Samples pretreated before thermal demagnetization by zero-field cycling to liquid nitrogen temperature, so as to erase multidomain remanence and isolate single-domain remanence, do have the theoretically expected TL values. In these samples, realistic remagnetization time and temperature (tr, Tr) conditions in nature are predicted using the t-T contours of Pullaiah et al. [1975]. The anomalously high TL values before low-temperature treatment are due to multidomain grains, which carry ≥50% of the LT overprint. The LT thermal demagnetization curve in samples dominated by multidomain grains is quasi-exponential in shape with a high-temperature tail extending almost to the Curie point, as predicted by multidomain theory. These high LT unblocking temperatures, which are much greater than plausible remagnetization temperatures reached in nature, overlap and mask the lower part of the HT unblocking temperature spectrum, driving up TL values and leading to inflated estimates of Tr. Although multidomain remanence is a sufficient explanation of anomalously high unblocking temperatures of thermoviscous overprints in the Milton Monzonite, chemical overprinting may be a factor in other lithologies and tectonic settings.

SUMMARY Low-temperature demagnetization (LTD) experiments with TRMs and pTRMs using liquid nitrogen were carried out on samples with different (MD and PSD) domain structures to infer and to compare the type of domain structure of samples before and after LTD judged by the thermomagnetic criterion. This is based on testing materials against Thellier’s law of independence of pTRMs. The results demonstrate good agreement between the commonly used criteria for differentiating domain structure, such as Qrt, Hc,/Hc and ZrS/Zs factors, and more sophisticated ones, such as the thermomagnetic criterion and LTD treatment. It is noteworthy that there is no observable tendency of MD or PSD samples to display SD-type pTRM =f(7‘) curves after LTD treatment. Hence our experiments failed to detect a quasi-SD component of remanence acquiring TRM by thermofluctuations in MD or PSD grains in the manner of Verhoogen (1959) and Stacey & Banerjee (1974) hypotheses

Eigen-analyses of the magnetic gradient tensor provide powerful tools for characterising and targeting magnetic bodies. A very useful invariant known as the scaled moment or normalised source strength peaks directly over a source body for a number of source geometries. Using this invariant, we investigate an angular minimisation method to triangulate the direction to source using a single string of data measured downhole with no assumption about the source geometry or position. The results are excellent for constraining the centre of magnetisation for the ideal case of an isolated dipole. When applied to multiple sources or non-dipolar geometries the observation string should be post-analysed as a series of smaller observation strings. Then, the estimated directions to source pass through the source body or bodies. Gradient tensor analyses, such as those presented here, will continue to become more important as new gradiometer systems start to come online.

Rock magnetic properties and palaeomagnetism of weakly metamorphosed banded-iron formations (BIFs) of the Palaeoproterozoic Hamersley Group, Western Australia, and Proterozoic BIF-derived iron ores have been investigated. The BIF units sampled here are slightly younger than 2500 Ma. At Paraburdoo, Mount Tom Price and Mount Newman iron ore formation was completed before 1850 Ma. Sampling was mainly from the Mount Tom Price and Paraburdoo mining areas and for the first time a palaeomagnetic fold test on fresh (unweathered and unaltered) BIF samples has allowed the nature of the remanence of the BIFs to be defined. The remanence of the BIFs is carried by late diagenetic/low-grade metamorphic magnetite after primary haematite. This remanence is pre-folding and is unlikely to be greatly affected by the high anisotropy because the palaeofield inclination was demonstrably low. Determination of palaeofield directions from measured remanence directions is complicated by self-demagnetization effects in strongly magnetic, highly anisotropic BIF specimens. We present a method for correcting measured directions for the effects of self-demagnetization and anisotropy. For typical BIFs, the effect of magnetic anisotropy on measured remanence inclinations and inferred palaeolatitudes is minor for low palaeolatitudes, but can lead to large errors in calculated palaeopoles for intermediate to moderately steep palaeolatitudes. Anisotropy also causes cones of confidence to be underestimated, due to compression of the range of inclinations. In principle, deflection of post-folding remanence towards the bedding plane by high magnetic anisotropy can produce an apparent syn-folding signature, with best agreement between directions from different fold limbs after partial unfolding. Thus high anisotropy cannot only bias estimated palaeofield directions and cause underestimation of errors, but can also mislead interpretation of the time of remanence acquisition. The anisotropy of anhysteretic remanent magnetization (ARM) probably yields an upper limit to the anisotropy of the chemical remanent magnetization (CRM) carried by the BIFs. Therefore, from the anisotropy of ARM, a maximum inclination deflection of 9° is suggested for the sampled BIFs. This corresponds to less than 5° change of palaeolatitude. The palaeomagnetic pole position calculated for BIFs at Paraburdoo is 40.9°S, 225.0°E (dp=2.9°, dm=5.8°) after tilt correction, but without correction for anisotropy. Other pole positions reported include that from flat-lying BIFs from Wittenoom at 36.4°S, 218.9°E (dp=4.6°, dm=9.1°), from Mount Tom Price iron ore at 37.4°S, 220.3°E (dp=5.7°, dm=11.3°) and from Paraburdoo ore at 36.4°S, 209.9°E (dp=4.7°, dm=8.8°). The poles from the BIFs, the Paraburdoo ore and the part of the Tom Price deposit that was sampled in this study are indistinguishable from each other and from the Mount Jope Volcanics overprint pole. The magnetization of the BIFs was probably acquired during burial metamorphism of the Hamersley Group, soon before the main folding and uplift event in the southern part of the Hamersley Province. This tectonic event exposed magnetite-rich BIFs to near-surface oxidizing conditions, producing extensive martite-goethite orebodies and also appears to have produced the syn-folding overprint magnetization recorded by the Mount Jope Volcanics of the underlying Fortescue Group. Ages of magnetization are tentatively interpreted as ∼ 2200±100 Ma for the BIFs, ∼ 2000±100 Ma for the supergene enrichment of BIF to martite-goethite ore, recorded by the Parabudoo and Mount Tom Price orebodies, and ∼ 1950±100 Ma for the metamorphic martite-microplaty haematite ore, recorded as an overprint by the Tom Price orebody and as the only surviving magnetization of the Mount Newman orebody.

Palaeomagnetic orientation of drill core from the Sydney-Bowen Basin is feasible. A consistent magnetisation is observed in coal measures in the basin enabling the remanence of an unoriented sample to be utilised to orient the sample. Palaeomagnetic orientation of drill core for fracture analysis has been used successfully in the Sydney-Bowen Basin. The direction of magnetic remanence in the sediments of the southern Sydney Basin is consistent across a wide area, this direction being north and up. Similarly, the remanence direction observed in Bowen Basin coal measures is consistent from the south (Moura mine) to the north (Goonyella mine) of the basin, and this direction is also north and up. The consistency of the remanence direction is lithology dependent. Fine sandstone and siltstone/mudstone units gave the most reliable results. Coarse sandstones gave inconsistent remanence directions and should not routinely be used for drill core orientation using palaeomagnetism. A longcore magnetometer constructed by the CSIRO Rock Magnetism Group enables measurement of the remanence of HQ drill core without the need to subsample. The measurement of the remanence of HQ drill core using a longcore magnetometer shows similar results to the remanence measured using laboratory magnetometers indicating that drilling does not alter the NRM. Thus, the noninvasive orientation of drill core by measurement of remanence using a longcore magnetometer is feasible for samples from the Permian coal measures of the Sydney-Bowen Basin. As well, storage of the drill core has no or limited effect on the remanence of the drill core. A small component is sometimes present but is easily removed by AF demagnetisation. Weathering has a severe effect on the remanence of samples, inhibiting orientation by palaeomagnetism. Weathering can produce hematite/goethite which retains a strong chemical remanent magnetisation and thus dominates the NRM of the sample.

Palaeomagnetic studies of strata within the southern New England Fold Belt show evidence of a northward migration of remagnetisation of the Belt during the Permian. In some rock units, such as the Kiah Limestone a syn-deformational remagnetisation is observed, while in others a pre-folding magnetisation is observed.

While paleomagnetic pole paths and the palaeo-distribution of the continents is reasonably well understood for the past 200 Ma, there is an urgent need for more reliable Palaeozoic palaeomagnetic poles, especially from the Gondwana continents. The distribution of continents and the disposition and convergence of their margins is an important consideration in conceptual models of terrane accretion. Some continental distributions suggest only small oceans between Laurentia and Gondwana throughout the Palaeozoic, while others suggest a wide oceanic separation existed in the Devonian-Carboniferous. Current thoughts on the Gondwana Palaeozoic pole path may be grouped into, the autochthonous school, the allochthonous school and the Kanimblan overprinting push. The pole position from the Black Hill Norite, South Australia, suggests that some Australian Cambrian pole positions which plot nearby may be Early Ordovician overprints.

Cleaning is a term used in palaeomagnetism to refer to laboratory methods designed to demagnetize preferentially the less stable components of magnetization. Although the diversity of methods suggests that there is a wide choice available, this is illusory. Except in simple cases, the wrong cleaning strategy may yield misleading results. Cleaning methods include thermal, magnetic (alternating field or a.f.), low-temperature, chemical, shock and microwave cleaning. A cleaning strategy involves one or more cleaning methods applied in a specific sequence. The design of a strategy is made easier with the a priori knowledge of magnetic mineralogy and granulometry, or in other words, rock magnetic properties. Although not a substitute for other rock magnetic experiments, the variation of low-field susceptibility with temperature ( k − T ) not only provides information on magnetic mineralogy and granulometry, but it also draws attention to chemical alteration that may occur during thermal cleaning. Apart from the rapid, though reversible, decrease of k associated with Curie temperatures, other diagnostic k − T features include the low-temperature (about −140°C) peak typical of multidomain (MD) magnetite, the high-temperature (Hopkinson) peak typical of fine-grained magnetite (and other magnetic minerals) and the classic lepidocrocite-maghemite-haematite profile, which is irreversible. Examples are given where the remanence caused by lightning can be effectively eliminated only by applying a.f. or low temperature (LT) pre-treatment. Such remanence may be an isothermal remanent magnetization (IRM), or a combination of IRM and anhysteretic remanent magnetization (ARM). As the unblocking temperature of relatively low-coercivity MD grains may extend to high-temperature, it is often desirable to suppress MD remanence. An example is given of palaeointensity determination, with the without LT pre-treatment. The nature of the LT transition is briefly addressed in the light of a rare double low-temperature peak, which may reflect both the isotropic point and the Verwey transition.

Interpretation of magnetic surveys in terms of geology is hampered by poor correspondence between broad lithological categories and magnetic properties, and by lack of knowledge of the geological factors that influence the magnetisation of rocks. Magnetic petrology is the integrated application of rock magnetic and conventional petrologic techniques to identify and characterise the magnetic minerals in rocks. This information elucidates the factors that produce, alter and destroy magnetic minerals and thereby influence the bulk magnetic properties of the rocks and their associated magnetic anomalies. Improved understanding of magnetic petrology is therefore essential for maximising the geological information that can be obtained from magnetic anomaly patterns.

SUMMARY The Kulgera Dyke Swarm consists of olivine tholeiites which have intruded late Proterozoic transitional-granulite gneisses and granites of the eastern Musgrave Block, in central Australia. Preliminary Rb/Sr results suggest that the dolerites were emplaced at 1054 f 14Ma. In addition, a Rb/Sr age of 1060 f 10Ma on a biotite from a pegmatite indicates thermal resetting of the country rock minerals during dyke emplacement. Palaeomagnetic investigations of the dykes yield a primary thermoremanent magnetization direction corresponding to a palaeomagnetic pole at 17"S, 266"E (Ag5 = 12"). In addition to this primary magnetization, an overprint component was present in many of the samples, providing a palaeomagnetic pole at 30"S, 138"E (AgS=24"), which is similar to previous results from other central Australian rocks affected by the Alice Springs Orogeny. The results extend the area of influence of the Carboniferous Alice Springs Orogeny southward into the Musgrave Block. Further, the results provide no evidence for an earlier, Late Proterozoic, Petermann Orogeny affecting the Musgrave Block in the Kulgera region. However, the possibility that a Petermann Orogeny thermal overprint has been erased by the Alice Springs Orogeny cannot be dismissed.

Previous palaeomagnetic study of a ∼ 10 m thick rhythmite member of the Elatina Formation, part of the Late Proterozoic (∼ 650 Ma) Marinoan glacial succession in South Australia, argued strongly for a low palaeolatitude (∼ 5°) of deposition. However, a fold test was thwarted by the approximate alignment of the remanence with the axis of tectonic folding. Here we report results of a fold test on soft-sediment slump folds with a wavelength of 14–22 cm in the Elatina rhythmites, the axes of which are approximately perpendicular to the direction of remanence. Fifty-one standard palaeomagnetic specimens sampled across the slump structure gave α95 = 3.7° prior to unfolding, α95 = 1.8° with 67% unfolding, and α95 = 2.4° with 100% unfolding. The results indicate that the remanence was acquired very soon after deposition; apparently it was acquired prior to soft-sediment slumping and then slightly sheared by the disturbance. The Elatina pole must be considered a virtual geomagnetic pole because the tidal rhythmite member studied represents only 60–70 years of deposition. However, the mean pole position is similar to other Late Proterozoic poles for South Australia, which implies that the very low inclination (< 10°) for the Elatina rhythmites does not record a geomagnetic excursion or reversal but does indeed indicate deposition in low palaeolatitudes. The similarity of many of the Late Proterozoic pole positions is evidence also that inclination error is not significant. The low palaeolatitude of Late Proterozoic glaciation is one of the major enigmas in contemporary Earth science, raising questions concerning the nature of the geomagnetic field, climatic zonation, and the Earth's rotational parameters in Late Proterozoic time.

SUMMARY Gently folded strata of the Late Devonian Merrimbula Group along the south coast of New South Wales are similar to numerous deposits of Late Devonian to Early Carboniferous, subaerial to shallow marine, quartzose sandstone that are known as the Lambie Facies of SE Australia. Because the Lambie sands overlap the early Palaeozoic tectonic elements of the Lachlan Fold Belt, large displacement since the Late Devonian of any Lachlan terranes, with respect to interior Australia, is precluded. We collected oriented core samples from 37 sites primarily in reddish, quartzose litharenites of the Worange Point Formation. The remanent magnetization is carried by haematite. Incremental thermal demagnetization reveals a dominant, well-defined, steep-upward-north component of magnetization that post-dates the mid-Carboniferous folding. The south pole position (146.4"E, 68.6"S, Ag5 3.1") derived from the overprinted specimens is close to both the Late Carboniferous and mid-Cretaceous reference poles as well as the spin axis of today. The overprint is attributed to both viscous partial thermoremanent and chemical remanent magnetization (VPTRM and CRM). Its exclusively normal polarity is consistent with a mid-Cretaceous acquisition, perhaps related to the rifting of the SE Australian margin. A characteristic component of magnetization is isolated between -660" and -680 "C in -30 per cent of the samples. Although a regional fold test of the overall result is inconclusive because the directions of characteristic magnetization are generally close to the axes of folding, two sets of samples from opposite limbs of a local syncline pass a fold test at 90 per cent confidence, indicating a pre-folding origin of the remanence. The presence of both polarities in thin horizons, some of which are palaeosols, and diagenetic haematite associated with incipient cleavage formed in an early stage of deformation before folding, indicates that the acquisition of remanent magnetization occurred over a protracted period. An analysis that combines great circle demagnetization paths and set points, facilitates the incorporation of more data but gives a biassed result. Although normal and reverse polarity subsets are roughly antipodal, residual post-folding components cause the normal subset to fail a fold test. A south pole position (19.7"E7 70.8"S, Ag5 7.1") is derived from the specimens with characteristic component of reverse polarity in which the isolation of a characteristic component is clear. The pole position is close to that of a previous palaeomagnetic study from a widely separated occurrence of Lambie Facies sediments; both results are applicable not only to the Lachlan Fold belt, but also to the rest of Australia, and Gondwana as a whole. The position of the two poles derived from the Lambie Facies overlap assemblage in the Lachlan Fold Belt supports the interpretation that the progression

A magnetic polarity reversal near the top of the Tumblagooda Sandstone, in the Carnarvon Basin, Western Australia, has been intensively investigated in five coastal sections and four inland sections, all located near the top of the formation. A further four inland sections towards the base of the formation were also sampled. Of the inland sections, all but two are suspected to have been largely remagnetized in similar fashion to Late Palaeozoic and Mesozoic sediments to the south. At the remaining seven sections a zone of normal magnetization overlies a zone of mixed (coexisting normal and reversed) magnetization, and the boundary between them coincides with a unique marker horizon of pebbly sandstone. The intimate relationship of the (virtual) reversal and the pebble marker horizon lends strong support to the suggestion that erosional stripping accompanied the formation of the pebble marker. The pebble marker is best interpreted as a lag deposit. The stratigraphic coincidence does not necessarily signify the concurrence of a reversal with the formation of the marker horizon, but rather that the relationship represents a variation on the unconformity test. The record of the actual geomagnetic reversal has been physically removed from most sections. These results show conclusively that the age of the magnetization of the Tumblagooda Sandstone is very close to the age of deposition, and can therefore provide an important pole for the definition of the Gondwana apparent polar wander path. However, the suggested Silurian age must be questioned on the basis of the palaeomagnetism which best accords with an Ordovician age. The lower parts of the sequence may be as old as the Cambrian. While the age of the palaeomagnetic pole from the Tumblagooda Sandstone is therefore not well constrained, its position is well defined and in combination with other Gondwana poles indicates a period of rapid polar motion during the Ordovician/Silurian. A collision with Laurasia at the end of the Silurian and beginning of the Devonian may explain the timing of the Caledonian/Acadian Orogenies.

A review of reliable palaeomagnetic data from Gondwana Palaeozoic rocks supports the apparent polar wander path (APWP) initially proposed by Morel and Irving. This path, or versions of it, has recently gained favour with a number of groups. The APWP suggests that during the mid-Palaeozoic Gondwanaland was very mobile. No palaeomagnetic pole position(s) has been yet reported that confirms the contentious segment of APWP from the Late Ordovician to the Early Silurian. The APWP implies that the south pole moved rapidly from north Africa in the Ordovician to a position off southern South America by the Silurian, back to central Africa by the Early Carboniferous and across Gondwanaland to Australia by the Late Carboniferous. The distribution of Palaeozoic tillites independently supports such a mobilistic Gondwanaland. The palaeomagnetic and tectonic evidence are compatible with the Lachlan Fold Belt of Australia having been in place since the mid-Devonian.

Algorithms for calculating the equilibrium energy state of a three-dimensional ferromagnetic particle according to an earlier micromagnetic formulation can now be implemented on a personal computer (PC). Some PC enhancement boards are now very fast by normal standards. For instance, the Mercury 6400 is rated at 20 MFlops (million floating point operations per second). The 20-MHz Definicon DSIT4-1/800 transputer, used for much of the present work, although slower (1.5 MFlops), is still able to perform the numerical convergence in a respectable time by using a progressive minimization procedure. Other numerical techniques which have employed supercomputers arrive at similar estimations of energy levels and magnetic structure; this indicates that the present technique is dependable. Several examples that demonstrate this are discussed.

The Tasman Fold Belt (TFB) of Eastern Australia can be divided into three meridional orogenic realms: the Kanmantoo, Lachlan-Thomson and New England Orogens. The geological histories of the orogens overlap, but each is distinctive. The Kanmantoo Orogen was provenance-linked to the Australian craton in the Early Cambrian, and accreted to Australia by Late Cambrian. There are many possible tectonostratigraphic terranes in the Lachlan Fold Belt (LFB) but these can be simplified to two major amalgamated terranes by the Middle Silurian. All the LFB terranes appear provenance-linked in the Ordovician, and were progressively covered, from the west, during the Late Silurian to Late Devonian, by a quartzose overlap assemblage. The New England Orogen has a fragmentary Early Palaeozoic history, but from the Devonian onwards its geology is related to a series of volcanic island and continental margin magmatic arcs. There is some evidence of provenance-linking between the Lachlan and New England Orogens in the Devono-Carboniferous but docking is not demonstrated until the mid-Carboniferous. The few reliable pre-Late Carboniferous palaeomagnetic poles available from the TFB come from the eastern LFB. The poles post-date accretion of the LFB to the Australian craton. Thus, the possibility that parts of the Lachlan-Thomson and New England Orogens contain exotic elements is yet to be tested palaeomagnetically.

New palaeomagnetic data from mid- to Late Palaeozoic rocks in Australia have enabled us to revise the Palaeozoic apparent polar wander path (APWP). This modified Australian APWP is supported by data from other parts of Gondwanaland. The palaeomagnetic poles indicate that during the Early and mid-Palaeozoic. Australia underwent rapid rotation: first clockwise during the mid-Ordovician to the Early Silurian, then counterclockwise from the mid-Silurian until the end of the Devonian, while it remained at low to equatorial latitudes. This was succeeded by a rapid southward movement during mid-Carboniferous times. The implications of the palaeomagnetic data for the tectonic relationship between the Lachlan Fold Belt (LFB) and cratonic Australia are consistent with the tectonic evidence that the LFB has been in place since the mid-Devonian.

[1] The Newcastle Range is an extensive (2500 km2) and well-exposed caldera system erupted on the trailing edge of Eastern Gondwana between 325 and 295 Ma. Paleomagnetic samples were collected from ignimbrites and associated microgranitoid intrusions from the central, northern and southern calderas from which three components of magnetization are recognized. Component 1 is considered to be a viscous magnetization acquired during the Brunhes Chron. A presumed Permian component, C2, is found in seven paleomagnetic sites with a mean pole at 30.9°S, 139.7°E (K = 13.9, A95 = 16.8°, ASD = 21.7°), agreeing with previously reported Permian data from Australia. Carboniferous units have a well-defined characteristic component, C3, distinguished by dual polarity (predominantly reversed) and moderate to steep inclination directions. Paleomagnetic polarities in the Newcastle Range Volcanics are formation dependent and new constraints on the timing of Carboniferous volcanism (∼325–317 Ma) are consistent with recent reanalysis of the base of the Permo-Carboniferous Reversed Superchron (PCRS). A mean paleomagnetic pole, calculated from 15 VGPs, lies at 63.4°S, 125°E (K = 26.22, A95 = 7.6°, ASD = 15.8°), suggesting that Australia remained at midlatitudes into the Middle Carboniferous. This paleomagnetic pole is consistent with similarly aged poles from Western Gondwana, the conformity of which indicates contributions from nondipole components of the Earth's paleofield were probably not significant in the time immediately preceding the PCRS.

Vector magnetometers may be used to determine the in situ magnetic properties of magnetic anomaly sources. The adaptation of caesium-vapour magnetometers to vector measurements has been challenging but rewarding. We describe the considerations that we have taken into account and summarise our success to date.

The marine Controlled-Source Electromagnetic (CSEM) method has proven itself an invaluable adjunct to the seismic method in petroleum exploration in the last few years. The problem is that it is not easily adapted to the Australian scene because most of Australia's petroleum deposits are in relatively shallow waters where CSEM does not work very well. Depending on the target, depths up to 500 m can be considered shallow water for CSEM. CSEM surveys in shallow water can have problems with the airwave dominating the received CSEM signal and also with electromagnetic noise induced by ocean waves. Modelling studies suggest that it is feasible to measure CSEM magnetic field gradients as a supplement to E and B field measurements and use this extra information to remove environmental noise (wave effects and magnetotelluric noise), airwave and direct wave components, and other unwanted contributions to the measured E and B fields. This can improve the signal/noise ratio of the CSEM measurements, thereby producing more reliable interpretations, and extend the useful range of the method, which can reduce survey costs. We have been developing a magnetic gradiometer that can be used for marine CSEM surveys. This poster will present the results from preliminary field trials conducted off Sydney's coast, in water depths up to 62 m, which show that the OceanMAG gradiometer was able to measure the magnetic field induced by ocean waves. Approaches for using magnetic gradient measurements as an adjunct to marine CSEM data will also be described.

Summary. A method that enables the objective resolution of almost parallel multi-component magnetizations is described and demonstrated. A feature distinguishing this method from others is its simultaneous analysis of demagnetization data from a group of specimens, rather than the analysis of data from one specimen at a time. The only prerequisite is that the specimens are derived from a homogeneous source. Thus for a formation carrying a simple single component magnetization, all specimens from the formation may be simultaneously reduced. For a more complicated two component magnetization it is shown that only specimens from a particular site can be considered homogeneous, and for a complex three component system each sample often requires undivided attention. Thus the workload is proportionally increased to achieve analyses of comparable reliability from data of variable quality. New pole positions from Mesozoic intrusions of the Sydney Basin, NSW are: from the Marsden Park Breccia pipe 48°S, 127°E (A95= 6°); the St Marys Breccia pipe 46°S, 150°E (A95= 8°); the Prospect Dolerite 60°S, 142°E (A95= 13°) and 53°S, 180°E (A95= 6°); and from the Dundas Breccia pipe 58°S, 162°E (A95= 36°) and 31°S, 195°E (A95= 16°). The last two formations possess multi-component magnetizations. These pole positions are consistent with previous results from south-eastern Australia.

Palaeomagnetic results from the late Middle to early Late Devonian Comerong Volcanics in the Budawang Synclinorium, Lachlan Fold Belt (LFB), New South Wales, Australia, satisfy a fold test and support the hypothesis that foliated parts of the LFB have been folded by megakinks during the terminal mid-Carboniferous orogenic phase. The preferred pole position is at 76.9°S, 330.7°E (A95 = 7.2°), allowing calibration of the Australian pole path at around 370 Ma ago. An alterative pole position at 71.3°S 283.1°E (same A95) takes into account a possible rotation of the megakinked terrane by 15° clockwise with respect to the Australian craton. Although there are still problems unravelling previously published results from Silurian rocks in the LFB, both because of unreliability and uncertainty as to where the LFB crustal units lay in relation to the Australian craton, there is no geological reason to suspect any large relative movement between the LFB and the craton since the Middle Devonian, and thus the Comerong Volcanics pole position is representative of the Late Devonian for Australia. The new pole requires reassessment of many other poles that have previously been accepted as representing the Siluro-Devonian for Australia, and Gondwanaland. We present reasons for believing that the magnetization, or remagnetization of many rock units (e.g., Mereenie Sandstone and Mulga Downs Group, Australia, and the Gneigura Supergroup, Africa) was later than sedimentation, probably during widespread Early Carboniferous deformation. Another rock unit (Msissi Norite, Africa) could be of Early Carboniferous, rather than Late Devonian, age.

In palaeomagnetic studies remagnetization circles are often observed to converge. The circles represent the simultaneous decay of two magnetic components, and the convergence zones represent estimates of their directions. It is shown that these estimates are biassed, the degree of biassing depending upon the distribution parameters. In particular the ratio of the precision of the underlying distributions ( κ 2 κ 1 ) is very important. Only when κ 2 κ 1 is very different from unity can converging great circle methods give accurate results. Of the examples available in the literature most appear to satisfy this criterion, but the indiscriminate use of these methods can lead to significantly biasssed results.

Summary. The difference between step-wise and continuous thermal demagnetization of rock specimens is examined with particular reference to the presence of multi-component magnetizations. The effect of the thermal dependence of the intensity of spontaneous magnetization is clearly evident in recordings of continuous demagnetization when a least-squares analysis of a two component system is performed. The differences between spontaneous magnetization, saturation magnetization and high-field magnetization are discussed and it is shown that the normalized thermal dependence of spontaneous magnetization can be determined with negligible error by measuring magnetization as a function of temperature in sufficiently hgh fields (> 0.1 T for magnetite, > 1 T for pyrrhotite). Using experimentally determined thermal dependence of high-field magnetization to correct the continuously observed intensities leads to analyses comparable to those of step-wise demagnetization data. Experiments designed to compare the temperature dependence of an apparent spontaneous magnetization, derived from the observed thermal decay of saturation remanence carried by a multidomain (MD) magnetite bearing sample, with the true spontaneous magnetization of magnetite reveal a systematic difference, with the apparent spontaneous magnetization decreasing more rapidly than the true spontaneous magnetization. This difference is minor, however, compared to the thermal decay of spontaneous magnetization and to a first-order approximation the use of the thermal decay of saturation magnetization to correct intensities should usually be adequate. Similar conclusions are supported by experiments with MD pyrrhotite bearing samples. These experiments serve to constrain models of MD remanence. The approximate proportionality of blocked remanence (TRM and saturation remanence) to spontaneous magnetization which is observed for these samples, which contain predominantly MD grains, resembles the behaviour expected for non-interacting single domain (SD) grains. Conventional MD theories assume that domain structure remains stable below the blocking temperature and attribute remanence carried by MD grains to Barkhausen discreteness of domain wall positions. However, if the domain multiplicity does not change with temperature, remanence due to Bark

Summary. Attempts to estimate palaeo-radii of the Earth, using palaeomagnetic data have necessarily been based on simplistic Earth models. It has been asserted that real geological processes are too complex to enable us to approach the problem quantitatively, and such attempts yield invalid results. We examine this and argue that, to the contrary, it appears that errors introduced by allowing for more realistic behaviour of the continents, e.g. ‘orangepeel effect’ and crustal extension, are smaller by an order of magnitude than the response of palaeomagnetic data to simplified expansion models. From a qualitative argument, it is shown that the observed Late Palaeozoic and Early Mesozoic palaeomagnetic data are not what should be expected from an expanded Earth. We conclude that it appears unlikely that the Earth has expanded significantly since the Early Mesozoic.