We report a study with HPLC/ICPMS on the long-term stability of the major selenium metabolite in human urine, namely methyl 2-acetamido-2-deoxy-1-seleno-β-D-galactopyranoside (selenosugar 1). Three separate experiments were performed of 4–28 weeks duration and incorporating various storage conditions: room temperature, 4 °C, −20 °C, −80 °C, lyophilisation, deoxygenation, or addition of a bactericide (NaN3). Triplicate samples of urine or water, spiked with selenosugar 1 at 200 μg Se L−1, were processed in each case. Selenosugar 1 was stable in water under all investigated conditions. For the urine samples, no significant degradation (<2%) was observed after 17 weeks frozen storage at −80 °C, or after lyophilisation and frozen storage at −20 °C, whereas small quantities of degradation products (ca. 3%) were recorded for frozen storage of wet samples at −20 °C. At 4 °C, the selenosugar was essentially unchanged after storage for up to 2 weeks, but clear losses were observed thereafter ranging up to 75% loss after 28 weeks. Several decomposition products were detected by HPLC/ICPMS, one of which was identified as dimethyl diselenide. Although present as only a trace constituent in the urine, dimethyl diselenide was recorded as a large HPLC peak, presumably because of a marked vapour enhancement effect due to more efficient transfer of volatile analytes to the plasma. In addition, total Se analyses revealed that Se was lost from the solutions during storage/handling, presumably as volatile species. Qualitative analysis of volatile species using head space sampling with solid phase microextraction followed by GC/MIP-AES and GC/MS revealed the presence of dimethyl selenide and dimethyl diselenide, based on comparison with standard compounds, and indicated the presence of dimethyl selenylsulfide based on comparison with literature data. The stability of selenosugar 1 and its isomer methyl 2-acetamido-2-deoxy-1-seleno-β-D-glucopyranoside (selenosugar 2), which occurs as a minor species in urine, was also investigated under room temperature storage in the presence and absence of light. Although both species were moderately stable when stored in the dark, their degradation was rapid in the light with clear losses recorded within three days. The work indicates that urine samples should be cooled immediately after collection, and that they may be stored at 4 °C (often the easiest way) for up to 2 weeks before analysis with no appreciable loss of selenosugar. For longer-term storage, urine samples should be kept at −80 °C or, when such facilities are not available, at −20 °C after lyophilisation. The study has also revealed potential quantification problems in Se speciation analysis resulting from different responses for Se species during ICPMS analysis.