1. Male Sprague-Dawley rats given (RS)-[3H]-8-OHDPAT by intraperitoneal (i.p.) or intravenous (i.v.) injection, or orally (p.o.) by gavage, excreted the majority of the dose in the urine (> 80% in 3 days and > 70% in the first 24 h). A smaller proportion of the dose was excreted in the faeces (> 10% in 3 days), mostly in the first 24 h. Total recovery was > 90% (mean: i.p. = 94.9; i.v. = 99 and p.o. = 92.9%). 2. Urinary metabolites were separated by reversed-phase hplc before and after treatment with beta-glucuronidase or sulphatase and quantitated by liquid scintillation spectrometry. Metabolites were identified by hydrolysis by specific enzymes, comparison of hplc retention time with those of authentic standards and by LC-MS. 3. Two major metabolites were identified and quantitated in the 24-h urine, namely 8-OHDPAT-glucuronide, accounting for some 45% of dose, and its N-despropylated metabolite, 8-hydroxy-2-(N-n-propylamino)tetralin, excreted as its glucuronide, which accounted for 15% of dose. Small amounts (< 1%) of two monohydroxylated metabolites were also identified, one eluting slightly earlier than and the other co-eluting with the mono-despropylated metabolite. When analysed by LC-MS-MS, the first of these exhibited a fragmentation pattern consistent with ring hydroxylation and the other appeared to be a side chain oxidized metabolite, which may constitute an intermediate in N-despropylation. However, these metabolites were present at too low a level to allow the exact position of hydroxylation to be determined. 4. These studies suggest that the low oral activity exhibited by 8-OHDPAT is most likely the result of rapid and extensive glucuronidation rather than poor absorption from the gastrointestinal tract.