With the stated aim of scanning small regions of interest in mice, several high-resolution positron emission tomographic (PET) systems are presently under development. Some, however, have low sensitivity and require high doses of radioactivity to achieve count statistics adequate to reconstruct small volumes. Using in vivo dissociation constants for three carbon-11 labelled ligands previously measured in rat brain, the present paper utilises simple saturation kinetics to estimate the limits on radioactivity and specific activity, to minimise the degree of receptor occupancy and achieve maximal specific binding of the radioligand. The extent of the problem is exemplified by considering a high-affinity ligand (dissociation constant in vitro approximately 0.1 nM; in vivo approximately 5 nmol/kg i.v. injected dose), where routinely produced levels of specific activity ( approximately 100 MBq/nmol) would limit the activity injected into mice to approximately 0.1 MBq for a 1% receptor occupancy. If, as is feasible, the new generation of high resolution PET systems requires an injected activity >10 MBq, then a >100-fold increase in specific activity would be needed for tracer kinetics to hold. The paper highlights the need to consider realistically achievable goals if high-resolution PET is to be accepted as a viable methodology to acquire pharmacologically and physiologically accurate ligand-receptor binding data in mice.