Testing reproductive allometry in fish

Energy allocation models have proved useful for clarifying the ecological mechanisms which influence growth and reproduction schemes in species with indeterminate growth and for understanding how traits correlate with generate the life history of specific species. In view of the need to understand how exploitation affects species for which few data on key population parameters such as reproductive potential and population growth rate are available, it would be helpful if simple energy allocation schemes parametrized by easily obtainable data could be used to estimate parameters such as reproductive effort and output. Testing the predictions of three such models on fecundity data from a range of species, we show how the simplifications required in a generalized energy allocation scheme are reflected in the deviations between model predictions and empirical data, and discuss the validity of the assumptions underlying these models. We caution against over-reliance on generalized and simplified models to predict reproductive effort.