In the paper “Does the fall phytoplankton bloom control recruitment of Georges Bank haddock, Melanogrammus aeglefinus , through parental condition?”, Friedland et al. (Can. J. Fish. Aquat. Sci. 65(6): 1076–1086, 2008) examine a sizable number of hypotheses aiming to explain the recruitment patterns observed in Georges Bank haddock. The authors focus on a correlation between the size of the autumnal phytoplankton bloom and the survivor ratio (recruitment), concluding this to be the main factor determining recruitment, via the mechanism of adult condition at the time of spawning. Here we examine this result in close detail and re-analyse some of the data presented in the paper. We show that the recruitment metric upon which Friedland et al. base their conclusions inadvertently biases the analysis in favour of high recruitment events and against low recruitments. As a consequence, Friedland et al. disregard correlations that are, in fact, significant. Furthermore, we show that the parental condition hypothesis hinges upon a single, highly uncertain data point, without which the correlation is no longer significant. We find that evidence for the parental condition hypothesis is weak, and that in performing the analysis in the chosen manner, Friedland et al. have overlooked alternative hypotheses.
Revisiting the limping parental condition of unmarried fathers
