This paper re-examines studies of Na kinetics in canine carotid arterial wall previously reported by three groups of investigators. The similarities and differences in experimental and analytical approaches for Na arterial wall washout therein are reviewed. Major conclusions are: (1) A three-compartment model in series is adequate for all short and long (after subtracting the small slow exponential) records. (2) Values for the diffusion coefficient, however, varied markedly among the different groups. (3) A continuous sampling method as opposed to an intermittent one measures better the kinetics of a rapidly exchanging ion as Na+. (4) Models based on individual washout records are better than those based on ensemble averaging. Characterization of sets of results in terms of mean values for comparing parameters provides significant potential for distinguishing quantitative features between control and treated sets despite inadequacies in data sampling or model specifications.