In our studies on diurnal 6-sulphatoxymelatonin (aMT6s) rhythms in various species, we have sometimes obtained fluctuating patterns. In most of these, the volume of individual urine fractions was not accurately measured because of methodological problems. Here, we report a simple method to overcome these problems by using urinary creatinine to estimate urine volume. The benefit of this method is demonstrated in two representative examples of the diurnal aMT6s rhythms of rats, domestic pigs and humans. Because the human urine fractions were collected accurately, the qualitative pattern of the aMT6s rhythm was not altered by using urinary creatinine as a substitute for urine volume. The total creatinine excretion (urine volume x creatinine concentration) was constant within a small range and showed no diurnal rhythm. In rats and pigs, the highly variable aMT6s concentrations relative to urine volume throughout the 24-hr period were changed drastically by referring to creatinine. All aMT6s patterns became stable and qualitatively similar to those of the rest of the group. From these results it can be concluded that creatinine is an adequate substitute for urine volume and a beneficial parameter with which to overcome technical problems with urine collection from laboratory animals or unknown urine volumes in human studies.