Reexamination of published data on growth and feeding of fishes shows that when fish are fed on one type of food, the logarithm of the gross growth efficiency (log K) decreases with increase in rations. For a number of species and experimental situations this relation is adequately described by the linear equation[Formula: see text]whence[Formula: see text]where ΔW is the growth, R the rations intake during the period of time Δt, and a and b are the parameters fitted to the linear form of the equation. With a single food-supply level, rations are highly correlated with body weight so that either one may be used to predict the growth efficiency. However, comparison with experimental situations which induce changes in rations intake at given body size, suggests that body size is not the determining factor as long as environmental conditions are within the normal biokinetic range. That is, the basic pattern of distribution within the animals of the energy intake is described by a knowledge of the rate of intake and two parameters.The action of various environmental factors appears to be reflected in this equation in different ways. In the experiments examined changes in temperature, which are known to affect the level of total metabolism, affected the value of R but the parameters a and b remained unchanged. That is, temperature changes appeared to affect only the energy turnover rate, not the distribution of energy in the fish. By contrast, changes in factors such as salinity and metabolic wastes, which are known to affect the metabolic load, affected the values of a and b, and showed interaction with body-size as well. The factor having the most important influence on the parameters seemed to be the type of food, especially with respect to particle-size. We infer that this simple effect on the linear equation is a characteristic expression of the complex integration of expenditures during searching and grazing with the success of this activity.These observations imply that the growth efficiency equation, or what we term K-line, provides a useful index of the costs of particular behaviour patterns in particular ecological situations. Together with observations on metabolic rates it may also be used to describe the growth of fishes in relation to their food supplies. It is thus a potential indicator of the relative positions of life-history phases of various species in the production system represented by a natural community.