MEIAJOU 289that during long-term therapy the plasma digoxin ooncentration reflects the tissue concentration because of a fairly predictable plasma: tissue distribution ratio. All the patients with digoxin toxicity reported here had been previously digitalized and intoxication occurred during the process of chronic therapy as a result of the slow accumulation of digoxin. In this clinical context it is possible to cautiously plan a therapeutic approach to the problem on the basis of the results obtained.If an adult patient has been digitalized with digoxin and has been taking digoxin orally for more than three days, and presents with symptoms or signs suggestive of digoxin toxicity and the plasma potassium is not less than 3-5 mEq/l., then:(1) If the plasma digoxin concentration is less than 4 miug./ml., digoxin intoxication is very unlikely.(2) If the plasma digoxin concentration lies between 4 and 5 mng./ml., then digoxin intoxication is very likely and digoxin therapy should be temporarily stopped and restarted at a lower dose. Certainly increasing the dose of digoxin would be contraindicated. If in this situation emergency therapeutic measures were indicated, a fairly safe assumption of digox n intoxication could be made and appropriate measures instituted.(3) If the plasma digoxin concentration is greater than 5 mug./ ml., then a definite diagnosis of digoxin toxicity could be made.The nine patients whose plasma digoxin concentrations are depicted in Fig. 2 as toxic have been handled according to these criteria with impressive clinicaltimprovement.In two of the illustrative cases (Nos. 1 and 2) sequential E.C.G. tracings are shown (Figs. 3 and 4) and though in these cases there are other possible causes of increasing S-T segment sagging these changes progressed as the plasma digoxgn concentrations rose and the patients became intoxicated.The measurement of plasma digoxin concentration would seem to offer a rational approach to the diagnostic problem of digoxin intoxication and helps in distinguishing those cases in which the presenting s ptoms and signs might be due either to the intrinsic heart disease or to digorin toxicity. The alternatives to a method such as that described here are double isotope derivative dilution assays, such as that described by Lukas and Peterson (1966) for digitoxin, and radioimmunoassay -techniques as described by Butler and Chen (1967) for digoxin and by Oliver et al. (1968) for digitoxin. Double isotope derivative dilution assays take far too long to be of much use in a clinical situation, and the radioimmunoassay techniques seem to offer little advantage over the procedure described and are at present too complex to be of widespread applicability.We are grateful to the physicians of St. Mary's Hospital who allowed us to study patients undu their care, and also to the Endowment Fund of St. Mary's Hospital and Pfizer Limited for financial support.