[en] It was suggested that the ratio peak 3/peak 5 can be used to estimate the time elapsed between the irradiation and readout of a personal dosimeter of TLD-100.. Horowitz and Yossian[2] claim that this statement might be wrong since Moscovitch did not study the effect of the time between prior readout and the irradiation. The possible fading of unfilled traps can also affect the peak 3/peak 5 ratio. We measured the ratio peak 3/(peak 4 + peak 5) and found that it depends actually on the time between the prior readout and the last readout (the sum of the pre-irradiation and post-irradiation times) as can be seen in table 1. This observation eliminates the possibility of measuring the elapsed time after irradiation. The decay of peak 3/(peak 4 + peak 5) for short pre-irradiation time can be plotted as signal vs. log(t) (figure 1) as was done by Moscovitch. However this functional form has no physical meaning and if exponential decay was looked for it was found to be composed of the sum of two exponential terms (figure 2), indicating that probably peak 3 is composed of two kinds of traps