Does the probabilistic maturation reaction norm approach disentangle phenotypic plasticity from genetic change?

Abstract

Two-dimensional probabilistic maturation reaction norms (PMRNs) define the probabil- ity that individuals mature as a function of age and size. PMRNs have recently been used to derive empirical support for the hypothesis that high fishing mortality has induced genetic change towards earlier maturation in exploited populations. However, the 2-dimensional PMRN method does not, strictly speaking, disentangle phenotypic plasticity from genetic change in maturation. Instead, it dis- entangles the contribution of variation in growth and mortality from other sources of variation con- tributing to changes in maturation. After removing the contribution of variation in growth and mortality, any remaining change is not necessarily purely genetic. Environmental factors may exist that directly affect the propensity to mature at a certain size and age. An observed trend in the size- age PMRN may therefore be partly or fully explained by a co-incidental trend in an (environmental) factor. As pointed out by Grift et al. (2007; Mar Ecol Prog Ser 334:213-224), such co-dependence is captured conceptually by multi-dimensional PMRNs, as opposed to 2-dimensional (size-age) PMRNs, where the higher dimensions represent the relevant variables. Dimensions that may partly or fully explain the observed trends in size-age PMRNs, other than evolution, include temperature, polluting endocrine-disrupting chemicals, the social structure of the population, and body condition; changes in the last two may be fisheries-induced. To illustrate these views, I re-analysed data on North Sea plaice, and demonstrated that temperature explains part of the shift in the size-age PMRN but that a residual shift remains. The latter finding supports the hypothesis of genetic change.