Abstract
Growth and survival rates during the early life stages are key vital parameters driving population dynamics of fish. Growth rates generally present a pattern of autocorrelation. Growth autocorrelation is stronger when faster and slower growing individuals continue to grow faster and slower. Thus, the extent of growth autocorrelation can be a tool for considering potential effects of early growth rates on subsequent growth rates in the life history of fish. In the present study, we applied a group-level growth autocorrelation analysis to Atlantic bluefin tuna (Thunnus thynnus) larvae in the northwest Mediterranean (MED) and the Gulf of Mexico (GOM). Based on the otolith increment width data compiled mainly from published datasets, the pattern of growth autocorrelation was described for the species and compared between the MED and GOM populations. Atlantic bluefin tuna showed the highest levels of growth autocorrelation during the early life stages compared with various fish species. Their characteristics supported the general hypothesis that the species and populations with higher growth rates have stronger growth autocorrelation, extending the hypothesis to cover large pelagic piscivorous fish. Therefore, the maternal effects and the environmental variability that larvae encounter right after hatching would be even more critical in survival dynamics and useful for predicting the recruitment dynamics than previously recognized.