On drawing conclusions nine years after the Exxon Valdez oil spill

We comment on the paper of Irons et al. (2000), which evaluated the status of several marine bird taxa in Prince William Sound, Alaska, nine years aft er the Exxon Valdez oil spill. We discuss concerns about the effects on the study design of inherent differences between the oiled and unoiled areas; about interpretations of results that use inconsistent criteria to define t he spatial scales of analysis; and about explanations of underlying causes that are not empirically founded. These comments highlight general difficul ties in assessing the effects of large-scale environmental perturbations. I t is important to draw conclusions about the effects of such events, but th e conclusions must be founded on accuracy in reporting study results, cauti on in interpreting the results, and adequate consideration of alternative c ausal explanations for the observed results.