Perspective on prospective methods for obtaining life table data1

The parameter cI denotes the temporal standard deviation of the recruitment distribution.Parslow et al. (1979) assumed equal G for all stages in their lag-Manly model.With this assumption (which seems reasonable) we obtain 23 parameters to be estimated and 27 equations.All parameters can therefore be estimated without assuming constant survival rates over stages.We therefore conclude that it is possible to obtain both survivorship and development times in a single cohort analysis by stage-frequency distributional statistics.Furthermore, it is not necessary to introduce the limiting assumptions that Hairston and Twombly have shown the Rigler-Cooley method to be based upon.Finally, rather than the equations suggested by Rigler and Cooley or the generalizations suggested by Hairston and Twombly, Eq. 12-16 should be used for estimation purposes: They require less external information, they are less restrictive, and in addition they make it possible to estimate parameters of the recruitment distribution.Furthermore, indepen-dent estimates of some of the parameters (such as known development times) can easily be incorporated in the estimation procedure.