As a key parameter in the management of fish populations, individual growth rate (GR) variations were examined in the European eel (Anguilla anguilla; >150mm) using extensive mark-recapture surveys in the lotic habitats of two small rivers of western France: the Fremur, supposed to be saturated, and at the same latitude, the Oir with densities fivefold lower than those of the Fremur. In both systems, generalised linear models were used to test whether spatiotemporal factors such as dominant habitat type or local density affect GR variability. In the presumed unsaturated system, the Oir, GR variability is mainly explained by a set of habitat suitability drivers (density, dominant habitat type). In the Fremur, GRs appear independent of differences in habitat density or productivity. Below saturation, an increase in density will decrease the GR through intraspecific competition. At saturation, intraspecific competition reaches such high levels that regardless of eel density and productivity, the resources available by individual are similar throughout the system. In these circumstances, the effect of density on growth was presumed undetectable. Despite these contrasted results, mean GRs observed in both catchments were closed (similar to 20mmyear(-1)). This is an unexpected result as GR is expected to be higher in unsaturated systems. This similarity could be explained by the difference between the two systems in terms of: (i) sex ratio (the Fremur is dominated by male, whereas the female is dominant in the Oir), (ii) habitat type distribution or (iii) possible interspecific competition (important salmonid populations in the Oir).