We assessed the intraspecific mass scaling of standard metabolic rate (SMR), maximum metabolic rate (MMR), excess post-exercise oxygen consumption (EPOC), and erythrocyte size in grass carp (Ctenopharyngodon idellus), with body masses ranging from 4.0 to 459 g. SMR and MMR scaled with body mass with similar exponents, but neither exponent matched the expected value of 0.75 or 1, respectively. Erythrocyte size scaled with body mass with a very low exponent (0.090), suggests that while both cell number and cell size contribute to the increase in body mass, cell size plays a smaller role. The similar slopes of MMR and SMR in grass carp suggest a constant factorial aerobic scope (FAS) as the body grows. SMR was negatively correlated with FAS, indicating a tradeoff between SMR and FAS. Smaller fish recovered faster from the exhaustive exercises, and the scaling exponent of EPOC was 1.075, suggesting a nearly isometric increase in anaerobic capacity. Our results provide support for the cell size model and suggest that variations of erythrocyte size may partly contribute to the intraspecific scaling of SMR. The scaling exponent of MMR was 0.863, suggesting that the metabolism of non-athletic fish species is less reliant on muscular energy expenditure, even during strenuous exercise.

The metabolic-level boundaries (MLB) hypothesis and the cell metabolism (CM) hypothesis have been proposed to explain the body mass scaling of metabolic rate. The MLB hypothesis focuses mainly on the influence of the metabolic level on the relative importance of volume and surface area constraints. The CM hypothesis focuses on the variation of cell size as the body grows. The surface area to volume ratio of individual cells may vary among species with different cell sizes, by which surface area constraints on metabolic scaling may change according to the MLB hypothesis. The present study aimed to extend the MLB and the CM hypotheses by proposing that, in addition to metabolic level, the varying cell surface area constraints among species also influence the intraspecific scaling exponents. The red blood cell area (S), and intraspecific scaling exponents for resting (b R ) and maximum metabolic rates of four species of cyprinids were assessed. The scaling exponents varied among species, but mass-specific resting metabolic rates (RMR) of each species were similar. No significant correlation was found between S and mass-specific RMR among species. As predicted, a significantly negative relationship exists between S and b R among species. The results suggest that the varying b R could be attributed to cell size differences among species, as those with larger cells may face stronger surface boundary limits, as predicted by the MLB hypothesis. This mechanism represents an additional way of relating the MLB and the CM hypotheses and does not exclude another mechanism based on the recent contextual multimodal theory.

Metabolic scaling is hypothesized to be affected by ontogenetic changes in the body chemical composition because several body components do not consume significant amounts of oxygen but contribute more to body mass (M) as a fish grows. We tested this hypothesis by assessing the mass scaling of the resting metabolic rate (RMR) and body compositions of lipid, protein, ash, and water in four species of cyprinids, Hypophthalmichthys molitrix (M range 9.5-218.3 g), Cyprinus carpio (M range 14.6-226.9 g), Carassius auratus (M range 4.5-323.2 g), and Ctenopharyngodon idellus (M range 5.8-274.2 g). The oxygen consumption of fasting fish was determined using a continuous flow respirometer at 25 °C and was used to assess the RMR. Then, the fish bodies were dried to a constant weight and used to determine the proximate compositions. No significant correlations were detected between the RMR and the masses of water, lipid, or ash after controlling for the wet mass in any species. The scaling exponents for the masses of lipid and/or ash were significantly >1; however, lipid and ash represented only a small proportion of the body mass of all four fish species. These results suggest that the increasing proportion of lipid and ash only explains a limited part of the metabolic scaling, and factors other than inert components may have a greater contribution. There was no significant difference between the metabolic scaling exponent for the wet mass and dry mass in all four species, suggesting that using either the wet or dry body mass has only a limited effect on the metabolic scaling exponent.

The effects of feeding frequency on the specific dynamic action of the juvenile snakehead were examined. The fish were fed at four different frequencies with the same total meal intakes (4% body mass), and the postprandial oxygen consumption rate and ammonia excretion rate were determined for two days. The durations of oxygen consumption and ammonia excretion significantly increased with increasing feeding frequency. The specific dynamic action coefficient did not significantly change as the feeding frequency increased. The net energy gain of the fish fed 4 meals daily was lower than the fish fed at the other frequencies. No significant change was found for the post feeding ammonia quotient as the feeding frequency increased. The total post feeding ammonia excretion and its ratio to the total nitrogen intake of the fish fed 4 meals daily were higher than those of the fish fed at any other frequency. These results suggest that frequent feedings result in a prolonged digestion process and reduce net energy gain and dietary protein efficiency, and very frequent feedings with smaller meals may be less beneficial for food energy utilization in the sedentary ambush foraging snakehead fish.

Unfortunately, the symbol 'B' was mistyped in line ''increased with exponent values significantly…'' of the introduction. The correct sentence is provided in this erratum. However, only one study tested this hypothesis and found that the lipid or ash of an amphipod, Gammarus minus, increased with exponent values B1, suggesting that the scaling of ash or fat content and metabolic rate are unrelated (Glazier et al. 2011).