The construction of ports has caused substantial habitat destruction in coastal areas previously used as nursery grounds by many fish species, with consequences to fish stocks. These artificial coastal areas might provide alternative nursery habitats for several species for juvenile fish abundances and growth in ports, although their contribution to adult stocks had never been estimated. The variability of otolith composition in the juveniles of two Diplodus species was investigated in three contrasting port areas and two adjacent coastal juvenile habitats of the Bay of Toulon (northwestern Mediterranean) in order to determine the possible use of otolith fingerprints as natural tags for the identification of juvenile fishes in ports. The global accuracy of discrimination between ports and coastal areas was very high (94%) irrespective of species, suggesting that otolith fingerprints can be used with confidence to retrospectively identify past residency in the ports of this bay. However, Ba was systematically the most discriminating element, since its concentrations in otoliths were generally higher outside ports than in inside them, probably due to river runoff. Moreover, otolith signatures varied greatly by species and between sampling sites. Furthermore, although Cu and Pb concentrations in water were at least 2.3–34-fold higher inside ports than outside, this was not consistently reflected in fish otoliths, confirming that spatial differences in otolith concentrations depend on the species and do not directly reflect differences in environmental contamination levels. Therefore, it seems unlikely that otolith microchemistry could provide a universal fingerprint capable of discriminating ports from other coastal areas. Nevertheless, the contribution of ports to adult fish populations can be determined well by establishing a library of otolith fingerprints for all juvenile habitats.