SynopsisThis paper reviews investigations on the ecophysiology of a population of roach, Rutilus rutifus, from a subalpine oligotrophic lake in the Austrian Tirol. Metabolic responses to season and temperature were studied in whole animals, tissues and selected enzymes. The exponent of the relationship between body mass and three levels of the metabolic rate of acclimated fish was 0.82 + 0.02,0.60 f 0.15, and 0.75 f 0.01 at 4, 12, and 20" C respectively. Various combinations of long-term acclimation to constant or seasonally fluctuating temperatures and long-term (up to 14 days) monitoring of VO, at the acclimation temperature led to the conclusion that the aerobic power of fish swimming in the routine mode does not show any sign of being temperature compensated. On the other hand, there are several indications that the energy expenditure of spontaneously swimming fish is adjusted to the seasonal pattern of environmental change and that these responses of metabolism and behaviour are controlled by both endogenous and exogenous factors. The rate of oxygen consumption of gill and muscle tissue 'brei' from fish caught during a seasonal cycle and measured at 15" C appears to follow closely the reproductive and gonadal cycle of the living fish. The same holds for the activities of phosphofructokinase, acetoacetyl-CoA thiolase, and cytochrome oxidase. On the other hand, the Na+, K+-ATPase of the kidney shows near perfect temperature compensation when fish acclimated to 5 and 25" C are compared, whereas an equally pronounced case of inverse temperature acclimation has been reported for the activity of digestive enzymes in the gut. Summarizing these data it is pointed out that the temperature relationship of a poikilothermic organism is the sum of often very diverse temperature relationships of specific metabolic and behavioural functions. In the case of the roach, strong effects of acclimation temperature on the molecular level, sometimes in the opposite direction, combine with seasonal effects on enzyme activities and tissue respiration. However, on the whole animal level the fish behave as strictly non-compensating poikilotherms, the reproductive cycle being the only detectable influence capable of modulating the basic temperature relationship of energy expenditure.Introduction two phases and is now in a third phase. In the first phase poikilothermic animals were regarded as beAs pointed out in a review some time ago (Wieser ing at the mercy of the thermal environment, an 1973)) the interpretation of temperature relations interpretation best expressed by Krogh's so-called of poikilothermic animals has already run through 'normal curve' (Krogh 1914). In the second phase,