Abstract. Within the field of environmental radiology, the levels of potassium and creatinine in urine are sometimes used to estimate the 24h excretion of radiocaesium from single urine samples. This is done in order to estimate the whole-body content of radiocaesium and the associated absorbed dose rate in man. Data on human urinary potassium excretion from various sources are available in the literature, but there are clear discrepancies between the ICRP 23 values for the reference man and these data which vary with the geographical location of the population. The relationship between urinary potassium and creatinine excretion and the correlation to parameters such as age, sex, body weight and ethnicity were investigated through multiple regression analysis. The groups of adults studied were a group from the city of Lund in southern of Sweden (1987-1994) and a group from rural areas outside the city of Bryansk, Russia (1994-1997). A significant difference was found in the relationship between the urinary potassium concentration and the creatinine concentration between the Swedish and Russian subjects. The elevated potassium concentrations found in the rural Russian subjects indicate a high dietary intake of potassium, which is in accordance with results from a Russian food habit questionnaire in the Bryansk area. No significant correlations were found between urinary potassium concentration and weight or age. The significant difference in the relationship between urinary potassium and creatinine between the two populations could explain the difference of a factor of two found in estimations of the whole-body content of I37 Cs for the Russian group, depending on which of the two substances, creatinine or potassium, was used for estimation of the body burden of ,37 Cs from single urine samples.