The activities in a mixture of two independently decaying radionuclides emitting indistinguishable radiation but with different half-lives, is determined by measuring the mixture twice. A computer simulation program has been set up to gain insight in the influence of the time interval between the two measurements on the accuracy of the activity determination for the shortest lived radionuclide. The influence of the activity ratio, the half-life ratio and the total time available for measurement has also been investigated. On the average, it appears that the inaccuracy can be minimized by taking a time interval between the two measurements of about 4–5 half-lives of the shortest lived radionuclide. This result can be applied usefully to the measurement of many samples with one detector.