The only clinically applicable method for the quantitative measurement of blood flow to the skeleton which has so far been developed rests on the assumption that the initial extraction of the bone-seeking isotope 18F from blood by bone is known and constant. Previous work to measure the single-passage extraction ratio of this isotope in rabbit bone has been criticised, so a novel technique has been developed in which isotope recirculation cannot occur. In addition, the post mortem uptake of isotope from surrounding tissues and the vascular tracer content of the bone can be monitored. The apparent extraction ratios of two bone-seeking isotopes, 85Sr and 18F, were measured in nine animals, and extrapolated to the time of death from both the time at which dissection of the bones began (giving a maximum value) and the time at which it ended (giving a minimum). Mean blood flow to the femur was 0.562 ml min-1 (SD = 0.170, n = 15) and mean blood flow to the tibia was 0.290 ml min-1 (SD = 0.140, n = 12). Statistical analysis of the results showed that there were significant differences in the extrapolated ratio between animals and significantly higher extraction of isotope in the tibia than in the femur. There was no significant difference between the extraction of 85Sr and that of 18F. The best estimate of the single-passage extraction ratio of 85Sr and 18F which can be made, therefore, is that in rabbit femur it is 0.88 (SE 0.04) and in the rabbit tibia it is 1.08 (SE 0.04). Post mortem uptake of isotope increases the apparent extraction ratio by about 1% per minute.