| Abstract: | A recently developed technique to estimate the lipid content of pelagic fish, using macroscopic visual assessment of their mesenteric fat, was applied to samples of southern Benguela sardine Sardinops sagax collected from commercial catches over the period 1996–2004. The data were analysed using general linear modelling, and temporal and spatial variably in lipid content was examined and compared with that of condition factor derived from length/mass relationships. Lipid content showed a clear seasonal cycle, being highest in May and lowest in August. This pattern was comparable with seasonal cycles in condition factor, although lipid content peaked 1–5 months later than did condition factor. A strong spatial effect was apparent in model results, with both lipid content and condition factor being highest in sardine from the South-East Coast. The correspondence in results from the fat staging technique, and those derived using condition-factor data for longer time-series, validate the new technique, and the higher level of resolution of the former compared with the latter argue for broader application of this technique. |
