Does condition of Atlantic cod (Gadus morhua) have a greater impact upon swimming performance at Ucrit or sprint speeds?

To compare the sensitivity of sprint and critical (Ucrit) swimming speeds to the condition of Atlantic cod (Gadus morhua) and to identify the best anatomic, behavioural and biochemical correlates of these types of swimming, we established two groups of cod that were fed or starved for 12 weeks. We evaluated sprint swimming and Ucrit performance as well as the speed at which repeated burst-coast movements began in the Ucrit test before measuring the metabolic capacities of red and white muscle sampled caudally, centrally and rostrally and the anatomic characteristics of the cod. White muscle lactate was measured directly after the Ucrit test. As expected, the twofold difference in Fulton's condition factor (0.5+/-0.04 for starved and 1.0+/-0.1 for fed cod) was accompanied by large differences in the anatomic and biochemical parameters measured. Despite the relative sparing of muscle aerobic capacity during starvation and despite the greater use of oxidative fibres during Ucrit compared with sprint swimming, these types of swimming differed by much the same extent between starved and fed cod. In the Ucrit tests, white muscle lactate levels and lactate accumulation per burst-coast movement were considerably higher in fed than starved cod, suggesting more intensive use of fast muscle fibres in cod in good condition. Multiple regression analysis indicated strong correlations between Ucrit, the speed at which regular burst-coasting began and the activity of pyruvate dehydrogenase (PDH) in red muscle (both caudal and central positions). PDH activity may limit the rate of oxidative ATP production by red muscle. The activity of cytochrome c oxidase in rostral white muscle was the strongest correlate of sprint swimming, suggesting that aerobic preparation of white muscle facilitates rapid contraction. The correlation between Ucrit and sprint swimming was weak, perhaps due to inter-individual differences in sensitivity during sprint tests.