Records of recent polarity reversals recorded in sediments show a preponderance of transitional VGP's situated in the longitudinal sector over the Americas and, to a lesser extent, over its antipode. A series of tests based on the methods of circular statistics show that this distribution is not consistent with the statistical fluctuations of a random probability distribution. The longitudinal bias in the distribution, therefore, requires a physical explanation.