Serum corticosterone is used as a biomarker of stress in laboratory rats for the evaluation of the response to experimental stressors. Environmental (animal handling) influences can confound the experiment by introducing extraneous stressors. The purpose of this study was to evaluate the use of carbon dioxide-oxygen anesthesia, with two different animal handling techniques, as an appropriate procedure for the collection of serum samples for corticosterone determinations in male and female Fischer 344 rats. The design of the study mimicked the initiation of a toxicity study, but only sham dosing occurred prior to blood collection. Animals were divided into two groups, acclimated and nonacclimated. Acclimated animals were anesthetized in their home cage, whereas nonacclimated animals were anesthetized in a separate anesthesia chamber. Blood was obtained from the retro-orbital plexus under carbon dioxide-oxygen anesthesia after the fourth day of sham administration of 0.5% aqueous methylcellulose. Serum corticosterone was measured by radioimmunoassay. Mean serum corticosterone concentrations in acclimated male and female rats were approximately 25% and 50% those of nonacclimated animals, respectively. Serum corticosterone concentrations in the acclimated groups were similar to those published for Fischer 344 rats after blood collection by decapitation and were lower than concentrations in studies using other collection methods. A minor change in the handling technique by anesthetizing animals in their home cage can result in reliable baseline corticosterone concentrations in male and female Fischer 344 rats. This refinement in technique will allow for decreased animal stress and decreased animal numbers (reuse of animal for future bleeds). These improvements in animal use can lead to better corticosterone data and lower variability on a given toxicology study.