In the last decade, there have been many reports of geomagnetic excursions, in which VGPs deviate more than 40 or 45 from the axial geocentric dipole field direction. It has become popular to think of excursions as aborted reversals because of the short-lived, but large, swing in VGP towards values of opposite polarity. Some authors refer to Brunhes Chron (0--780 kyr) excursions as subchrons (a subchron is a zone bounding two closely spaced polarity transitions; e.g. Spell and McDougall [1992] accept eight subchrons during Brunhes time). If such zones are real subchrons, they have important implications for geomagnetic field behavior because they imply that the reverse polarity state has been less stable than the normal polarity state for the past 780 kyr. Publication of a substantially new geomagnetic polarity timescale for the late Cretaceous and Cenozoic by Cande and Kent [1992] provided Merrill and McFadden [1994] with the opportunity to re-examine the relative stabilities of the normal and reverse polarity states. Their analysis confirms the previous conclusions of McFadden et al. [1987] that there is no reason to reject the hypothesis of a common stability for the normal and reverse polarity states at any time during the Cenozoic. Given this confirmation, Merrill and McFadden [1994] conclude that it is necessary to question the reality of the numerous claims of short reverse polarity subchrons in the Brunhes chron. After examining the literature, they conclude that there is no convincing evidence for reversed subchrons in the Brunhes chron and that the only sufficiently well documented geomagnetic excursions are the Laschamp and the Blake excursions. Many authors make no distinction between excursions and aborted reversals [e.g. Tric et al., 1991; Hoffman, 1992; Quidelleur and Valet, 1994] and excursion records are commonly included in statistical analyses of transitional field behavior. Merrill and McFadden [1994] consider the evidence for excursions as aborted reversals as equivocal. For example, the Laschamp and Blake excursions appear to occur when the non-dipole to dipole field ratio was large with a low dipole field intensity, without evidence for reversal of the dipole. As a consequence of the possible difference in causal mechanism for polarity reversals and excursions, Merrill and McFadden [1994] advocate that polarity transitions and excursions should be treated separately when the data sets are examined to determine if systematics are present.
The Pringle Falls (Oregon) record of Herrero-Bervera et al. [1989]
has been widely cited as the most detailed record of the Blake excursion.
Herrero-Bervera and Helsley [1993] have confirmed the earlier Pringle
Falls study by obtaining a second record from another sedimentary
sequence, 1.5 km away from the original site, which duplicates all of the
features seen previously. Herrero-Bervera and Helsley [1993] and
Zhu et al. [1994b], who describe a record of the Blake excursion from a
Chinese loess sequence, suggest that the common features observed in
numerous records support the global nature of the Blake excursion.
Herrero-Bervera et al. [1994] and Negrini et al. [1994] have
demonstrated, on the basis of correlation of paleomagnetic directions,
tephrochronology, and radiometric dating, that the Pringle Falls record is
synchronous with excursion records from Summer Lake (Oregon) and Long
Valley, California. These records represent a distinct and synchronous
geomagnetic excursion, with concomitant low field intensities [
Roberts et al., 1994], which must be considered one of the best
documented multiple records of a single excursion. However,
Ar/
Ar dating of a tephra layer that is closely associated with
the excursion at all three western North American localities,
demonstrates that the excursion is considerably older than the inferred
age of the Blake excursion [ Herrero-Bervera et al., 1994; Negrini
et al., 1994]. Clearly, this finding has important implications for the
inferred global nature of the Blake excursion.