Several species of marine fishes perform seasonal migrations alternating foraging and spawning habitats during their life cycles. The knowledge of these movements is essential to define their population structure which is, in turn, necessary for a sustainable management of any fishery resource. Parasite tags have been extensively applied for stock identification and host migration assessment, proving to be excellent markers. However, ontogenetic changes in the structure of parasite assemblages must be considered since, due to ontogenetic cumulative patterns, differences between host length can lead to a misinterpretation of the patterns in this type of studies. The aim of this work was to evaluate the asynchronous migrations of the Argentine croaker Umbrina canosai, a demersal fish that inhabits coastal waters of Brazil, Uruguay and Argentina, as a cause of ontogenetic differences in parasite loads, considering its impact in the interpretation of migratory patterns when parasite assemblages are used as biological tags. A total of 254 specimens distributed into 8 samples (3 newly collected and 5 from previous study) with differences in mean length were examined for their metazoan parasites. Only long-lived parasites were used for comparisons between samples. A strong influence of fish length on the structure and composition of the parasite assemblages was observed across samples. Indeed, a clear pattern of similarity decay between parasite communities was evident as the differences in host length increased, demonstrating the importance of ontogenetic changes in the structure of parasites assemblages. On the other hand, differences in parasite assemblages were, most noticeably observed in young fish, indicating probable variations in migratory routes, distance traveled and/or latitude reached, depending on environmental conditions and age. Consequently, differences among fish length classes must be considered when using parasite tags for those resources with temporally and spatially variable migratory patterns, especially when different cohorts are compared.
