Summary 1. The objectives were: (i) to check the validity of a new growth model; (ii) to examine the relationship between population density and both mean mass and mean growth rate and (iii) to discover if compensatory growth occurred. First (0+) and second (1+) year-old juvenile sea-trout were sampled by electrofishing at the beginning and end of the summer from 1967 to 2000. Additional samples were taken in some years in winter and in the critical period for survival when the fry first emerge from the gravel. The trout left the stream as pre-smolts in May, soon after their second birthday.
2. A growth model (Elliott, Hurley & Fryer, 1995) estimated the mean mass of the trout over the 2 years spent in fresh water. The date and mean mass at the start of the growth period were defined as the median date for fry emerging from the gravel and their mean mass at emergence, both being estimated from individual-based models (Elliott & Hurley, 1998a, b).
3. The variation in mean mass among year-classes was small for newly-emerged fry (CV = 6.2%), maximum at the start of the first summer of the life cycle (CV = 38.1%), and then decreased gradually for successive life-stages to a low value for pre-smolts (CV = 10.8%). Mean mass was not related to population density and, therefore, mean growth rate was density-independent. Growth in the first, but not the second, winter of the life cycle was lower than model prediction, but when it was assumed in the model that there was no first-winter growth, there was good agreement in most year-classes between model estimated values and observed mean mass. Exceptions were that mean masses and growth rates for 0+ trout after four summer droughts were lower than expected, but compensatory growth followed, so that observed and expected masses were similar for 1+ trout.
4. Pre-smolt mean mass on 30 April measured total growth achieved in the freshwater phase of the life cycle. This was significantly related to mean mass at the end of the first and second summers of the life cycle, but not to the emergence date and mean mass of emerging fry.
5. These juvenile sea-trout were growing at their maximum potential in most year-classes but when this was not achieved, compensatory growth soon restored their mass to values expected from the model. This ensured a low variation in the mean mass of pre-smolts just before they migrated to the sea. However, the latter mass was higher in more recent year-classes (1987–98) than in previous ones (1967–86), demonstrating the effect of slightly higher stream temperature. This study has shown the importance of developing realistic growth models in order to detect departure from maximum potential growth, and the more subtle effects of temperature change, possibly due to the effects of climate change.