The pharmacokinetics of widely used and of investigated aminoglycosides, namely kanamycin, gentamicin, tobramycin, amikacin, sisomicin and netilmicin, in normal volunteers and in patients with physiological states and disturbances known to alter their disposition are critically examined in view of recent developments. Analytical methods for the assay of aminoglycosides in biological fluids are briefly but critically described. Emphasis is placed on multicompartmental characterisation of aminoglycoside kinetics, especially as it relates to certain heretofore unexplained observations such as tissue accumulation and delayed elimination. Other components of the kinetic profile of aminoglycosides are reviewed. For practical purposes, aminoglycosides are neither protein bound nor biotransformed. The major route of elimination is glomerular filtration, and aminoglycosides undergo some tubular reabsorption. Relevant pharmacokinetic parameters from significant studies have been tabulated. Some practical aspects of distribution kinetics are then presented. All aspects considered, and even though therapeutic concentrations can be obtained in some patients, the penetration of aminoglycoside into the cerebrospinal fluid, bronchial secretions and biliary tract remains somewhat unpredictable, depending in particular on the underlying diseases. When feasible, local treatment such as aerosols, intrathecal or intraventricular supplementation of parenteral administration are to be recommended.Due to the almost exclusive excretion from the body by glomerular filtration, the elimination rate of aminoglycoside antibiotics is greatly affected by impairment of renal function. The mathematical relationships between pharmacokinetic parameters and renal function indicators are tabulated and discussed. The predictive potential of nomograms can be greatly improved by taking into consideration some physiological parameters such as sex, age, lean body mass, haematocrit and characterisation of disposition kinetics by multicompartmental models. However, actual serial monitoring of aminoglycoside concentrations in serum appears to be the most reliable method for therapeutic monitoring. Half-lives in patients with minimal renal function during haemodialysis, during peritoneal dialysis and during interdialysis periods are tabulated and the practical implications of these data discussed.In the neonatal period, the absorption rate after intramuscular injection appears to be faster than reported in adults and the volume of distribution is significantly larger. The major pharmacokinetic difference between neonates, infants and adults is the slower elimination: for instance, half-lives which average around 2 hours in adults with normal renal function can reach and sometimes exceed 5 to 6 hours during the very first days of life.The volume of distribution per kilogram of body weight is relatively smaller in obese subjects and in dehydrated patients than in normal subjects and can lead to overdosing when dosages are calculated on a body weight basis. Anaemia, fever, hypoxaemia and major burns are other pathological states which significantly influence pharmacokinetics of aminoglycosides. Inactivation by heparin, carbenicillin and ticarcillin is reviewed.The methods of administration currently favoured are briefly examined and a procedure which may be more applicable to the kinetics of aminoglycosides is suggested so that further comparative clinical trials may be undertaken of the assessment of the relative merits of different modes of administration.