Until 1990 biokinetic studies of aluminium metabolism and biokinetics in man and other animals had been substantially inhibited by analytical and practical difficulties. Of these, the most important are the difficulties in differentiating between administered aluminium and endogenous aluminium-especially in body fluids and excreta and the problems associated with the contamination of samples with environmental aluminium. As a consequence of these it was not possible to detect small, residual body burdens of the metal following experimental administrations. Consequently, many believed aluminium to be quantitatively excreted within a short time of uptake in all, but renal-failure patients. Nevertheless, residual aluminium deposits in a number of different organs and tissues had been detected in normal subjects using a variety of techniques, including histochemical staining methods. In order to understand the origins and kinetics of such residual aluminium deposits new approaches were required. One approach taken was to employ the radioisotope (67)Ga as a surrogate, but this approach has been shown to be flawed-a consequence of the different biological behaviours of aluminium and gallium. A second arose from the availability, in about 1990, of both (26)Al-a rare and expensive isotope of aluminium-and accelerator mass spectrometry for the ultra-trace detection of this isotope. Using these techniques the basic features of aluminium biokinetics and bioavailability have been unravelled. It is now clear that some aluminium is retained in the body-most probably within the skeleton, and that some deposits in the brain. However, most aluminium that enters the blood is excreted in urine within a few days or weeks and the gastrointestinal tract provides an effective barrier to aluminium uptake. Aspects of the biokinetics and bioavailability of aluminium are described below.