Abstract The methods of constructing life tables using mortality data From capture-mark-recapture studies are reviewed. A stochastic model was produced to simulate cohort mortality and produce mortality estimates derived by three methods: (i) fusion-cohort analysis, entering each individual from the time of capture to the time of death; (ii) pseudo-cohort analysis, placing each individual in the table From time of birth; and (iii) pseudo-cohort plus cadaver analysis, including unmarked individuals where they can be successfully aged. The accuracy of the three construction methods was compared for differing levels of mortality, cadaver recovery, probability of capture, and number of capture attempts. The accuracy of these methods is presented graphically and general recommendations are made.