THE FOLD TEST - AN EIGEN ANALYSIS APPROACH

We combine eigen analysis and parameter estimation techniques for a ne wly constituted, more versatile fold test. The method is automatic, re quiring no assumptions about the polarity or distribution of data, and gives confidence limits on the degree of unfolding required to produc e the tightest grouping of data. We illustrate the method using severa l published data sets that show the tightest data groupings before, af ter and during correction for bedding tilt. The latter case is usually ascribed to acquisition of remanence during folding, but we show that this behavior can also arise from undetected multiple rotations. In o ur simulation, the beds undergo rotation about a vertical axis as well as a horizontal one, a case likely to occur in nature. These data, wh en rotated back to horizontal around what would be the observed strike , exhibit a peak in concentration at about 60% unfolding, very like th e behavior of many published data sets. Thus, the origin of remanence in many such cases may not be syn-folding at all, but the behavior may purely be the result of an artifact of structural complications.