Determination of Critical Periods in Ontogenetic Trajectories

Identification of ontogenetic shifts and thus potential critical developmental periods allows the formulation of testable hypotheses relating to the constraints faced by an organism during growth. A moving regression analysis was developed to discover and quantify the departures from single power-curve scaling that are frequently observed in ontogenetic data. To illustrate this approach, data on the scaling of brain weight in the albino mouse are reanalysed as a demonstration of the validity and usefulness of moving regression analysis. Data on egestion rate in the deposit-feeding marine annelid Capitella species I are then explored and modelled using a suite of statistical methods combining moving regression, scatterplot smoothing and piecewise regression. The model is then used to construct hypotheses relating to the ontogeny of energetic constraints in Capitella sp. I. We conclude that these methods provide a powerful and general approach for the functional analysis of ontogenetic