The direction of a secondary magnetization component is found from the intersection point of converging remagnetization circles using a method based on the least-squares fitting of great circles to points on a sphere. The technique may be applied to any problem that requires the best intersection point of convergent great circles and is thus useful in other fields besides palaeomagnetism, such as structural geology, plate tectonics and astronomy.

The Midcontinent rift system is a 1.1‐b.y.‐old structure extending from Kansas, through the Lake Superior region, and into southern Michigan. The rift is filled with thick sequences of basaltic volcanic rocks and clastic sediments. For most of its extent it is buried beneath Paleozoic rocks but can be traced by its strong gravity and magnetic anomalies. The rocks of the rift system are exposed only in the Lake Superior region and comprise the Keweenawan Supergroup. Much of the geology of the Keweenawan is beneath Lake Superior and has only been inferred from potential field studies and seismic refraction studies and extrapolation from on‐shore geology. Seismic reflection surveys by the Great Lakes International Multidisciplinary Program on Crustal Evolution in 1986 imaged much of the deep structure of the rift beneath the lake in detail. The reflection profiles across the rift reveal a deep, asymmetrical central graben whose existence and magnitude was not previously documented. They show that, in addition to crustal sagging documented by previous investigations, normal faulting played a major role in subsidence of the axial region of the rift. A sequence of volcanic and sedimentary rocks, in places greater than 30 km thick, fills the graben. Thinner volcanic and sedimentary units lie on broad flanks of the rift outside of the graben. Near the axis, the prerift crust is thinned to about one fourth of its original thickness, apparently by a combination of low‐angle extensional faulting and ductile stretching or distributed shear. The sense of asymmetry of the central graben changes along the trend of the rift, documenting the segmented nature of the structure and suggesting the existence of accommodation zones between the segments. The location of the accommodation zones is inferred from abrupt disruptions in the Bouguer gravity signature of the rift. Uplift of the central graben occurred when the original graben‐bounding normal faults were reactivated as high‐angle reverse faults with throws of 5 km or more in places. The Midcontinent rift has some striking similarities to some younger passive continental margins. We propose that it preserves a record of nearly complete continental separation which, had it not been arrested, would have created a Middle Proterozoic ocean basin.

No abstract

No abstract

No abstract

A palaeomagnetic pole position, derived from a precisely dated primary remanence, with minimal uncertainties due to secular variation and structural correction, has been obtained for China’s largest dyke swarm, which trends for about 1000 km in a NNW direction across the North China craton. Positive palaeomagnetic contact tests on two dykes signify that the remanent magnetization is primary and formed during initial cooling of the intrusions. The age of one of these dykes, based on U–Pb dating of primary zircon, is 1769.1 ± 2.5 Ma. The mean palaeomagnetic direction for 19 dykes, after structural correction, is D = 36°, I = − 5°, k = 63, α95 = 4°, yielding a palaeomagnetic pole at Plat=36°N, Plong=247°E, dp = 2°, dm = 4° and a palaeolatitude of 2.6°S. Comparison of this pole position with others of similar age from the Canadian Shield allows a continental reconstruction that is compatible with a more or less unchanged configuration of Laurentia, Siberia and the North China craton since about 1800 Ma

No abstract

A paleomagnetic study involving alternating field (AF) and thermal cleaning, baked contact tests, and the measurement of hysteresis and other magnetic properties has been carried out on 21 Keweenawan diabase dikes (age ~1.1 Ga) from Baraga and Marquette Counties, northern Michigan. The main results are as follows.(1) The dikes exhibit a steep, negatively inclined remanence, which is shown to be a thermoremanent magnetization (TRM) acquired parallel to the ambient field at the time of initial magma cooling.(2) A secondary component, more resistant to magnetic cleaning than the primary TRM, and with west-northwest declination and positive inclination is occasionally found in the dikes, especially those from Baraga County.(3) Despite an early pole position obtained by Graham, which suggested that the dikes in Michigan were potentially important in defining the depth of the Logan Loop, most of them, from Marquette County, yield a pole that lies on the loop's western arm. The region of the projected apex of the loop thus still remains without reliable data.(4) The pole position of the Marquette dikes (48.4°N, 213.5°E; dp = 5.2°; dm = 6.2°) obtained from 14 sites is virtually identical to that obtained from 17 reversely magnetized dikes from the Thunder Bay district in Ontario, but appears to be distinct from poles derived from other Keweenawan units. Both dike swarms are thus interpreted to represent parts of the same magmatic pulse, and one that occurred in the centre of Lake Superior during an early opening phase of the Keweenawan rift system.