Abstract
The frequency distributions per unit area of the projected lengths (p-distributions) of fission tracks intersecting an internal and external apatite surface and the surface of a muscovite external detector have been established by measurement. Deviations from the ideal distributions on a number of points can be tied to the effects of track revelation and observation. The effect of track revelation, in particular, masks the effect of variations of true track length, and precludes temperature-time path modelling based on length measurements of surface tracks. These effects do not prevent calculation of the track counting efficiencies (etaq) in the track registration geometries of interest to fission-track analysis: etaq = 1.01 +/- 0.01 for induced tracks revealed in an external surface of Durango apatite, etaq = 0.91 +/- 0.01 for both fossil and induced tracks revealed in an internal surface of Durango apatite, and etaq = 0.91 +/- 0.01 for induced tracks revealed in a muscovite external detector. The fact that the latter are significantly less than unity is not due to an etching effect (critical angle theta(c).) but to an observation threshold, best described by a critical depth z(c). For tracks revealed in an internal surface, etaq decreases rapidly with decreasing track length. As a result, the apparent age of strongly annealed apatite samples may be underestimated by as much as 5%, irrespective of whether the absolute method, the Z-method, or the zeta-method is used for dating. (C) 2002 Elsevier Science Ltd. All rights reserved.
