An analysis of avian reproduction studies submitted for pesticide registration.

Avian reproduction studies are currently part of the regulatory ecotoxicology requirements for pesticides in many countries. In the study, 134 avian reproduction studies were reviewed to determine their ability to identify pesticides that have the potential to affect reproduction in wild birds. A clustering procedure was first used to assign measured variables to parental, developmental, or eggshell effects. This assignation was found to be identical in the two bird species tested. Nineteen of 69 pesticides tested were found to cause developmental effects at levels lower than those giving rise to detectable parental toxicity. At least some of these should not have been registered without some assurance that developmental effects would not occur in the wild. The analysis also found very little similarity in the effects of pesticides on the two bird species commonly used in avian reproduction tests. This casts serious doubts on the ability to extend the results of avian reproduction studies to any potentially affected bird species. Modifications to the avian reproduction test, based on results of the analysis conducted, are suggested. It is recommended that the avian reproduction study be recognized as a rough screening tool only and that efforts not be made to make it more realistic, e.g., such as through a reduced exposure period.