1. The use of a whole body counter in total body potassium estimations is described. The accuracy and value of the counter in measuring serial changes in total body potassium was assessed.
2. Serial study of twelve normal subjects at varying intervals revealed a normal variability in total body potassium of up to 1·23% (4-weekly measurements over 3 months). The error involved in counting statistics alone was 4%.
3. The whole body counter appears to be an accurate and useful method of assessing total body potassium in clinical studies.