The aim of this comparative study was to gain more information about the metabolism and excretion of glucocorticoids in cats and dogs in order to establish non-invasive methods for evaluating stressful conditions. Therefore, in a first experiment, [14C]cortisol was administered intravenously to 8 animals (two of each sex and species). Over a period of 6 days, faeces and urine were collected immediately after spontaneous defecation and urination. Marked species differences were found, as cats mainly excreted cortisol in the faeces (82% +/- 4% of the total recovered radioactivity), whereas in dogs only a small portion was found there (23% +/- 4%). The highest urinary radioactivity was observed after 9 +/- 3 h in cats and 3 +/- 1 h in dogs. Peak concentrations in the faeces occurred after 22 +/- 6 h in cats and after 24 +/- 4 h in dogs. Most of the radioactivity was not extractable with diethyl ether, indicating that the metabolites excreted in urine and faeces were mainly of the conjugated or polar unconjugated types. This was confirmed by RP-HPLC, which also revealed marked differences between cats and dogs concerning the metabolites formed. In addition, the immunoreactivity of the metabolites was tested in cortisol, corticosterone and 11-oxoaetiocholanolone EIAs. The latter, measuring 11,17-dioxoandrostanes (11,17-DOA) detected the highest quantities of immunoreactive metabolites in cats, but not in dogs. In a second experiment, the adrenal cortex of both species was stimulated by ACTH and, three weeks later, suppressed by dexamethasone. In this study, only faeces were collected over a period of 7 days. In both species, inter-animal variability in the basal and maximal/minimal faecal cortisol metabolite concentrations and the time course was observed. The 11-oxoaetiocholanolone EIA in cats and the cortisol EIA in dogs proved best suited for monitoring changes in adrenocortical activity. ACTH injections resulted in an increase above baseline values of 355%, (median) in 11,17-DOA concentrations in cats and of 702% in the concentrations of cortisol equivalents in dogs by about 25 h and 22 h (median) after injection, respectively. Minimal concentrations after dexamethasone administration were about 17% in cats and 31% in dogs (in relation to baseline values) and were reached in 66 h and 72 h, respectively. It was concluded that measuring cortisol metabolites in faeces should be a useful non-invasive tool for monitoring stress in carnivores.