Aim We used a combination of new and previously published palaeoecological data to test three hypotheses: (

The effect of ration on the growth of the three-spined stickleback, Gasterosteus aculeutus, was studied over the temperature range 3-19" C, which corresponded to the range in their natural habitat. With an ad lib ration, the specific growth rate increased with temperature. The relationship between specific growth rate and ration at a given temperature was curvilinear. Regression models were used to predict the maximum, optimum and maintenance rations at each temperature. Whereas maximum ration increased rapidly with temperature, the optimum and maintenance rations were relatively insensitive to temperature. A regression model was used to describe the overall relationship between specific growth rate and ration, temperature and body weight. This model predicted that at low rations, growth rate decreased with temperature, whereas at high rations, growth rate increased slightly with temperature. The gross growth efficiency increased rapidly with rations above maintenance ration up to a maximum at the optimum ration and declined at higher rations. The maximum gross growth efficiencies were at high temperatures and rations about half of the maximum ration. The maximum net growth efficiencies occurred at relatively low temperature and at rations just above maintenance.

1. Monthly and diel patterns of food consumption by the three-spined stickleback, Gasterosteus aculeatus L., were studied in Llyn Frongoch, a small reservoir in upland mid-Wales.2. Copepods were particularly prominent in the diet in spring and autumn, with ephemeropteran nymphs prominent in summer. Chironomid pupae and stickleback eggs were also elements of the diet in summer, whereas algae, plant material and debris were more important in autumn and winter. Chironomid larvae and ostracods were present throughout the year, but there was some doubt as to the importance of ostracods as digestible food items.3. Stomach contents were heaviest in spring and late summer and lightest in late autumn and winter. 4. Samples for the diel samples were taken four times, once in each season. These diel samples largely reflected the seasonal changes in diet noted in the monthly samples. There were few cases of a clear switch in the composition of the diet during a 24 h period.

In a population of Gasterosteus aculeatus living in an infertile, upland lake, the fish bred at the age of one year and few survived to breed again. The highest growth rates were achieved in the first month of life, but in comparison with other populations of G. aculeatus growth was slow during the autumn and ceased during the winter. But in spring and early summer, there was a spurt in the growth rate.Laboratory studies provided regression models for the prediction of the rate of food consumption from measurements of growth. The estimates of the consumption rate indicated the effect of the growth in body size and seasonal variations during the first year of life. It was estimated that a fish of mean length in the population consumed 3 150 mg of food in a year in which it grew from 65.8 to 552.0 mg, with an overall gross growth efficiency of 15.4%.The study illustrated the integration of laboratory and field studies to obtain reasonable estimates of the rate of food consumption by fish throughout their first year of life.

No abstract