Minimum number alive (MNA) is commonly used to assess population size with capture-mark-recapture data. However, MNA uses information from prior and subsequent capture sessions to assess the population at each point in a longitudinal study. Therefore, it is subject to negative bias that is greatest at the beginning and end of the study and least in the middle. Stochastic simulations performed with constant population size and capture rate showed that MNA peaked at the middle of the study. The tapering bias was greatest when survival rate between capture sessions was high. If indices (rather than statistical estimators) are used to assess population size, then the number of individuals captured should be chosen in preference to MNA.