The densities of perch, Perca fluviatilis L., and roach, Rutilus rutilus (L.), were estimated in six and three lakes, respectively, using mark-recapture and ranged from 25-1064 perch ha )1 and 865-2749 roach ha )1 . Effects of fish density, net type and water temperature on catch-per-unit-effort (CPUE) were analysed by generalised linear models (GLM) and generalised additive models (GAM). GAMs were fitted to estimate the simultaneous linear and nonlinear effects of temperature and density on CPUE. These models showed significant nonlinear effects of density on CPUE -mostly for perch, and also for both species at high densities. The models also revealed that both species had distinct density-temperature criteria of expected maximum CPUE and zero CPUE values. Because of the nonlinear relationship between CPUE and density, it was concluded that CPUE should be used with caution as a proxy of density. K E Y W O R D S : catchability, density, GAM, multimesh nets, perch, roach, temperature.

This study was carried out to assess the efficiency of eight fishways constructed in connection with hydroelectric dams in the Glomma river system in south-eastern Norway. The fishways were of the pool and weir type, except for one which was of the counter current (Denil) type, and had a discharge of 0.5-1.0 m3/s. During 1985-8 3077 grayling and 2995 brown trout were captured in traps in the fishways, measured and floy-tagged before being released in the river upstream of the fishways. The number of fish captured in two of the fishways were compared with estimated fish stocks in the areas around the fishways. Migrations began in May or June when the water temperature was > 5-6°C. Spawning migration of grayling occurred in May. During May more than 200 grayling may enter a single fishway within a week. In other fishways grayling occur only sporadically before late June or early July, when they migrate upstream to feed. Brown trout migrate during late spring, summer and autumn. The spawning migration takes place more or less during the whole summer, but mostly in late July and August. Immature brown trout also pass through the fishways, with a peak in October in three fishways. This migration is probably an ascending emigration from recruitment areas located downstream of these fishways. The efficiency of these fishways appears low, as the number of fish ascending was less than 2% of the estimated stock within the stretches where 90% of the recaptures occurred. The discharge in the fishway relative to the total discharge seemed to be of great importance, and to achieve efficient fishways they should be constructed for higher discharges, compared with river discharge, than the fishways in this area. The longest migration of grayling was 100 km, whereas the longest migration of brown trout was 122 km. The fish seemed to exploit shorter river stretches in areas influenced by several dams/fishways than fish in an area influenced by only one dam. KEY WORDS Migration timing Distances Purpose Trends

Vendace and whitefish in Lake Osensjøen (boreal south-east Norway) were studied by means of gillnet test fishing and hydro acoustic acquisition . Vendace increased in number between 1998 and 2009 while growth and size at maturity decreased. The relative density of whitefish decreased in the pelagic habitat, whereas growth and size remained the same. Both species exhibited varying year-class strength. Whereas strong year-classes of both species became less frequent after 1980 than before this, this seemed to change after 2000, especially for vendace. Generalised additive models suggested a strong positive relationship between July/ August mean air temperature and year-class strength of both vendace and whitefish. Whitefish recruitment was also negatively affected by the new regulation regime implemented since 1981, and positively correlated with water level after hatching and by late ice off. The reason for the diverging impacts of environmental factors on the two species may be that vendace spawn in deeper waters that whitefish do. The results indicate that increasing summer temperatures benefit recruitment of both species, whereas low water level and early ice off will harm whitefish recruitment. Both trends are caused by climate warming. It may be speculated that increased density of the specialised plankton feeding vendace may affect the algal community through increased predation on herbivorous zooplankton and potentially affect the trophic state of the lake.

Population structure, density and reproductive potential of perch (Percafluviatilis L.) from 4 acidified lakes in a river system in Southern Norway were investigated. The upper 3 lakes were most affected by acidification and the number of perch caught per unit effort indicate three-fold increase in density from the upper Gjerstadvann to the lower Brsbsrvann. Low density of perch in the upper lakes is explained by; 1) Abnormal mortality occurring in episodes due to acidification, probably caused by aluminium toxicity at pH = 5.2-5.4.2). Acid water (pH 55.0) during spawning and development of eggs and larvae, resulting in recruitment failure.Juvenile mortality of perch may depend on the density of the parental stock. In the 3 upper lakes, the density of adult perch probably was too low to produce significant juvenile mortality, and in these lakes the recruitment probably depended more on the May water quality. Improved water quality increases egg hatching and survival and also benefits zooplankton production, the main food for perch during the first summer.Low population density has reduced competition for food and thus improved growth of perch in the 3 upper lakes compared to the lower lake.The perch in Brabsrvann mature at higher age and have lower individual fecundity than perch in Gjerstadvann, an effect of both slower growth and lower length specific fecundity in Brabsrvann. In Gjerstadvann, the perch therefore have a higher reproductive potential relative to the stock density, and may, when the water quality is good enough, give rise to strong year-classes and more frequent year-class fluctuations than perch in the less acid Brsbsrvann.

This study was based on genotyping eight microsatellite loci of 463 brown trout, Salmo trutta L., sampled in nine differently sized tributaries in three areas on the eastern shore of Lake Mjøsa, south-eastern Norway. The populations were genetically structured, and Mantel's test showed that genetic distance correlated positively with geographical distance. Temporal differentiation F ST over a 2-year period was estimated in five streams and was nonsignificant after Bonferroni correction. Effective population size N e was positively correlated with the habitat length available from the lake (0.3-22 km) and negatively with the number of full sib pairs in the sample. There was no correlation between N e and genetic diversity, and private alleles were recorded in three medium-sized streams, but not in the largest two. The importance of small spawning and nursery streams for the maintenance of genetic diversity of brown trout was demonstrated.