Two separate feeding experiments were carried out to determine the effects of dietary carnitine supplements on growth rates, total ammonia nitrogen (TAN) excretion and respiratory quotient rates (RQ) in the African catfish, Clarias gariepinus (Burchell), juveniles fed various diets differing in protein energy:nonprotein energy ratio (PE:NPE). In Experiment 1, 540 fish (20.9 ± 1.0 g) were evenly distributed into six dietary groups and fed in duplicate aquaria (N = 12 aquaria) to apparent satiation over a 41-day period, until all fish attained an individual final weight of 156.9 ± 10.8 g. Diets were formulated to contain two levels of carnitine (80 and 660 mg kg1) and three levels of PE:NPE (0.7, 0.9 and 1.4). Sampling was divided into two size ranges, according to the predicted individual end-weight of 60 g (days 0-16) and 160 g (days 16-41). In Experiment 2, 240 fish (16.5 ± 0.2 g) were distributed in four 140-L aquaria connected to a respirometer and preconditionally fed twice a day at a restricted feeding level of 24 g kg0.8 day1 for a 25-day period (final weight of 77.5 ± 1.5 g). Four isonitrogenous diets (368 g kg1 crude protein) were formulated to contain two levels of fat (85 and 175 g kg and two levels of carnitine (40 and 600 mg kg1), in a 2 2 factorial design. From days 20-25, three respirometric trials were carried out, during which total ammonia nitrogen (TAN) and RQ were determined over 24 h. In both experiments, carnitine supplemented fish accumulated 2.54 times more carnitine in their tissues than fish fed a basal level (P < 0.05). In Experiment 1, the smallest fish (<60 g) fed 660 mg carnitine showed significantly higher growth rates and lower feed conversion rates than 80 mg carnitine supplemented fish only when the dietary PE:NPE ratio was low (0.7). Additionally, dietary carnitine significantly increased protein:fat ratios in the whole body, suggesting a protein-sparing action. In Experiment 2, high carnitine supplements decreased both TAN excretion and RQ rates in fish <80 g.