A non-lethal sampling method for stable carbon and nitrogen isotope studies of tropical fishes

Despite prior studies showing good agreement between fin and muscle isotope ratios in temperate fishes, the non-lethal method of fin sampling has yet to become a standard technique in isotopic food-web studies, and the relationship between the two tissues has never been tested in the tropics. We hypothesised that fin and muscle delta(13)C and delta(15)N would be strongly correlated in tropical fishes, thus allowing non-lethal sampling of these species. To test this hypothesis, we analysed fin and muscle tissues from 174 tropical fishes representing 27 species from the Mitchell River, Queensland, Australia. Fin tissue was a strong predictor of muscle-tissue delta(13)C (r(2) = 0.91 for all species) and was slightly enriched in (13)C (0.9 parts per thousand), consistent with the results of studies on temperate species. Fin tissue was a poorer predictor of muscle-tissue delta(15)N (r(2) = 0.56 for all species) although the mean difference between the tissues was small (<0.1 parts per thousand). Differences were smallest in the largest fish, possibly because the elemental composition (%N) of fin more closely resembled that of muscle. These measurements provide more impetus for increased use of fin tissue as a non-destructive means of testing hypotheses about fish food webs in the tropics and elsewhere.