The doubly labeled water (DLW) method for the measurement of the field metabolic rate (FMR) of free-ranging vertebrates has limitations that are most evident when dealing with very small mammals, such as the nectarivorous marsupial honey possum (Tarsipes rostratus), that exhibit high metabolic rates and elevated rates of water turnover. We tested the efficacy of the radioisotope rubidium-86 (Rb-86) as an alternative method for estimating FMR in this species in winter in Scott National Park in extreme southwestern Western Australia. We carried out a trial with 58 honey possums injected with a cocktail containing Rb-86, tritium, and oxygen-18 and correlated estimates of the FMR using the DLW method with repetitive measurements of the biological elimination rate constant (kb) of Rb-86. The 2 variables were significantly correlated and a linear regression explained 86% of the variation, which increased to 91% with a power curve fit to the data. Successive measurements of kb in individuals that were repeatedly recaptured over a 12-day period did not differ significantly, nor did the time interval between release and recapture affect the estimate of kb. The mean error involved in predicting the FMR from the kb for Rb-86 was less than 5% and of the same order of magnitude as for the error inherent in the DLW method itself. Although the basic mechanism is unknown, the major advantages of the Rb-86 method include the much longer time span over which animals may be recaptured and still provide reliable data, and the absence of the need to collect blood from the injected animals. We suggest that further research measuring the k(b)s of radioisotopes such as Rb-86 in other species may lead to viable alternatives for the measurement of FMRs in very small mammals with high rates of metabolism and water turnover.