In order to assess the latitudinal variability in growth of the eastern Atlantic skipjack ( Katsuwonus pelamis), conventional tagging data collected by the International Commission of the Conservation of Atlantic Tunas (ICCAT) since the 1960s are reanalyzed using a modified version of the von Bertalanffy-Fabens growth model. An analysis of the latitudinal patterns of recaptures showed that the lowest mixing rate between northern and southern regions was evidenced at 10 degrees N Latitude. Since it is theoretically admitted that growth parameters L-infinity and K are negatively correlated, we substituted L-infinity into the growth equation model by a simple bioenergetic function of K fitted from existing studies on skipjack growth parameters. The likelihood for the meta-analysis of growth is combined with the likelihoods from the two regional tagging data with the aim of estimating simultaneously K in both areas and to transfer adequately uncertainties associated with the different data sets. From this study, it was showed that fish inhabiting waters south of 10 degrees N latitude can be expected to grow generally larger but at slower rate than in the North (L-infinity = 112.34 cm vs. 89.38 cm, and K = 0.14 vs. 0.38, respectively). Our results are consistent with the range of growth estimates obtained in the Atlantic Ocean and in different parts of the World's oceans. In contrast, the estimates of L-infinity and K done in the Senegalese region in the 1980s within the framework of the Skipjack Year Program, and traditionally applied by ICCAT, are not supported by the present study. The findings concerning the phenotypic plasticity exhibited by growth of skipjack among latitudinal regions suggest potential differences in fish population productivity and consequently imply that non-conventional stock assessment methods such as catch-at size/growth transition matrix per region could be used to assess the responses of skipjack to exploitation.