Glymour, Bruce D.2013-08-222013-08-222013-08-22http://hdl.handle.net/2097/16320Glymour (2006) claims that classical population genetic models can reliably predict short and medium run population dynamics only given information about future fitnesses those models cannot themselves predict, and that in consequence the causal, ecological models which can predict future fitnesses afford a more foundational description of natural selection than do population genetic models. This paper defends the first claim from objections offered by Gildenhuys (2011).en-USThe final publication is available at http://link.springer.comCauseFitnessModelingNatural selectionPopulation geneticsThe wrong equations: a reply to GildenhuysArticle (author version)