Criticality codes are in general adjusted to reproduce measured values of critical bucklings which furnish (k∞ − 1)/M2. The parameter adjusted is k∞, the calculation of leakage being assumed to be well made. However, in the case of heterogeneous systems in particular, the slowing down region is not easy to study, and one has perforce to make certain simplifying assumptions which reflect on the calculated value of the age. Since a proper estimation of leakage is necessary for a code adjusted on clean critical systems to be valid when extrapolated to large power reactors where leakages are different, it is of practical interest to be able to use some experimental data for checking age calculations and searching for improvements if necessary. Pulsed experiments furnish the necessary experimental data, since measurements made on a given lattice for different block sizes permit the separation of multiplication from leakage. A method of analysis is presented and applied to experiments on natural uranium/graphite lattices. An effective age-diffusion expression in which k∞/p, L2 and the mean lifetime lo are evaluated in terms of buckling-dependent spectra, is transformed into a linear equation which permits simultaneous adjustment of p and of the age. Our analysis shows that pulsed experiments can be sufficiently precise for age adjustments. However, since these experiments are performed at far-from-critical bucklings, the precision is not sufficient for adjusting p, and hence k∞. We conclude that these experiments are very useful for adjusting leakage, but this adjustment being made, critical experiments remain necessary for the subsequent adjustment of k∞ with precision.