The fact that one cannot kill a bird twice makes it very difficult to determine the relative contributions of fat and non-fat components to increases in body mass before migratory flights in individual birds. Knowing the relative contributions of these components is of obvious energetic interest since fat yields about eight times as much energy as fat-free muscle tissue. Several recent studies have failed to demonstrate convincingly, due to flaws in their analyses, that fat-free mass in addition to fat is accumulated before long-distance flights. We point out that regressions of fat or the non-fat component on total body mass cannot yield reliable estimates of the composition of individual mass changes in view of inter-individual variation in structural size, reserve levels or timing of storage. We suggest that studies over time of synchronous populations or marked individuals will give better answers. A re-analysis of published data indicates the widespread existence of fat-free tissue deposition during migration, whereas in some species fat alone explained the increase in total body mass. Larger species tend to incorporate a relatively higher proportion of non-fat components when increasing in mass. However, the comparative data set is not yet of sufficient quality to allow general statements on why, and to what extent in different individuals and species, non-fat tissue in addition to fat is deposited before take-off on migratory flights.