A set of thin dikes from central New Mexico, dated at 469 ± 7 Ma (Rb-Sr; Loring and Armstrong, 1980), have yielded a virtual geomagnetic pole which lies on the Late Paleozoic segment of the North American apparent polar wander path. The remanence of the dikes appears to be a product of Late Paleozoic hydrothermal alteration. Paradoxically, however, the magnetization of the host rocks is most simply explained in terms of a positive contact test. Samples collected between 0.2 and 0.5 dike-widths from the contact contain a component of remanence parallel to the magnetization in the dikes, with unblocking temperatures which decrease with distance from the dikes. Host rocks from a distance of more than 1 dike-width show no evidence of the characteristic dike magnetization. There are two possible resolutions of this paradox: 1. (1) the magnetization of the host rocks is secondary, despite the apparent positive contact test, and is a product of hydrothermal fluid migration through the dikes or along the contact zones; or 2. (2) the magnetization of the dikes is primary, but not representative of the Ordovician paleofield for North America. Possible reasons for inaccurate representation include: 1. (a) incomplete averaging of secular variation; 2. (b) tectonic rotation with respect to the stable craton; or 3. (c) erroneous age determination for the rocks. We argue that explanation (1) is the most likely.
Bibliographical noteFunding Information:
Financial assistancefo r field work was provided by the Geological Society of America (Grant 316583),S igmaX i, the Scott Turner Fund of the University of Michigan, and the New Mexico Bureauo f Mines and Mineral ResourcesF. unding for laboratoryw ork was providedb y the Division of Earth Sciences,t he National ScienceF ounda-tion (Grant EAR 84-07007t,o R. Van der Voo). M. Jackson was a National ScienceF oundation Graduate Fellow during the course of this research.C ommentsb y R.B. Hargraveso n an earlier draft improvedt he paper.