Gas chromatography mass spectrometry (GCMS) and inductively coupled plasma mass spectrometry (ICPMS) were evaluated for the measurement of selenium (Se) and Se stable isotope ratios. GCMS and ICPMS were found to be accurate for quantitative Se analysis in biological matrices by isotope dilution using Se-78 and Se-76 as internal standards, respectively. A higher precision was obtained for ICPMS than GCMS enabling a smaller quantity of the tracer to be administered to subjects in labelling experiments. The isotopes of choice for metabolic tracers were Se-76 when sample analysis was by GCMS and Se-77 and Se-82 when analysis was by ICPMS. The influence of copper (Cu) on endogenous fecal Se excretion and true absorption of Se in nonlactating Holstein cows was examined by the use of Se stable isotopes as tracers. The method involved the application of conventional balance techniques in conjunction with isotopic enrichment of the body Se pools. Selenium in several tissues following oral and intravenous routes of isotope administration were evaluated as the precursors of endogenous fecal Se. Two cows fed a Se deficient diet (0.035 mg kg⁻¹) were administered 4 mg Se-76 orally, daily, for 5 d. After a 10-d equilibration period total collection of feces was made daily for two 5-d periods. The animals were then sacrificed and samples obtained from all major tissues and fluids. Se-7 6 enrichment (tracer/tracee mass percent, TTMP) in tissues was variable (< 0.56 - 13.4). However, enrichment was similar (9.8 - 12.9) in the tissues considered as potential contributors to endogenous fecal Se (serum, epithelium of the stomach, liver, bile, pancreas, small intestine and colon). Enrichment in serum and liver was used to calculate endogenous fecal Se. Apparent absorption of Se in the two cows was negative (-37 and -147 µg d⁻¹). Correction of apparent absorption for the fecal Se of endogenous origin gave a true Se absorption (% of intake) of 10 and 16%. The percentage of total fecal Se of endogenous origin was 23 and 36%. In two trials, 5 or 6 cows were assigned to one of two Cu-supplemented treatment diets: 0 mg kg⁻¹ or 17 mg kg⁻¹. The basal diet contained 0.19 mg Se kg⁻¹ and 13 mg Cu kg⁻¹. To each cow ~4.6 mg Se-77 and ~1.3 mg Se-82 were administered by oral and intravenous routes, respectively. After a 14-d equilibration period, total collection of feces and urine were made daily for two 5-d periods. Serum was collected on the first, third and fifth days of each period. Liver biopsies were taken 2 d following the completion of the balance periods. The estimates of endogenous fecal Se ( d⁻¹) from enrichment in the serum (256) and liver (235) following oral administration of the tracer and from enrichment in serum (241) following intravenous administration were not significantly different (P>0.05) but were higher than the estimate from the enrichment in liver (197) (P<0.05). No significant differences (P>0.05) were present when true absorption ( µg d⁻¹) was determined from enrichment in serum (290) or liver (268) following oral administration or from enrichment in serum (274) or liver (230) following intravenous administration. It was concluded the analysis of serum or liver with oral administration or the analysis of serum with intravenous administration of the tracer would provide reliable methods for estimation of endogenous fecal Se and true absorption. There was no effect of Cu on endogenous fecal Se excretion or true absorption of Se. Apparent and true absorption were 3.2 and 11%, respectively. Approximately 90% of the total Se excreted was in the feces, of which, 9.7% was of endogenous origin. The use of Se stable isotopes as metabolic tracers in dairy cattle provided a safe alternative to the use of radioactive tracers and enabled experiments requiring multi-isotopic enrichment to be performed. / Land and Food Systems, Faculty of / Graduate