1. A new laboratory-based growth model for brown trout (Salmo trutta) was u
sed to explore latitudinal variation in growth among natural populations. T
he model included the effects of differences in ambient temperatures and fi
sh size among populations. Annual growth rates of anadromous brown trout pa
rr from 22. Norwegian populations at 61-70 degreesN were compared with pred
ictions from the growth model. Published field data from one Spanish, 15 Br
itish and four Danish populations at 44-58 degreesN were included in the an
alysis to increase the latitudinal range.
2. Among the Norwegian populations, the ratio between observed and predicte
d growth rates was not significantly different from 1.00 in eight rivers, b
ut was significantly higher in eight, and was significantly lower in six. O
bserved growth was highest, relative to predicted growth, in the coldest ri
vers. In Spanish, British and Danish rivers, observed growth did not exceed
predicted growth.
3. The ratio between observed and predicted annual growth rate decreased si
gnificantly with increasing annual mean temperature. Observed annual growth
was higher than predicted growth only in rivers with an annual mean temper
ature lower than 5.1 degreesC, and this indicates that some kind of thermal
adaptation may occur in trout populations in the coldest rivers.
4. Regression analyses showed that besides the direct effects of temperatur
e and body size predicted by the growth model, annual growth rates were sig
nificantly related to annual mean temperature, densities of juvenile salmon
ids, duration of twilight (average for May-August) and latitude. Adding the
se variables to the original model increased the explanatory power from 73.
9 to 80.6%.