The development of phytoplankton communities was studied in floating enclosures. The enclosures were supplied with either surface water or water from 40 m depth. Nutrients with or without silicate were added in some of the experiments, while others recieved no artificial fertilization. It is shown that diatom dominance occurred irrespective of season if silicate concentration exceeded a threshold of approximately 2 PM. Flagellate dominance changed to diatom dominance within a few days after nutrient addition resulting in silicate concentrations above this threshold. Dominance of Phaeocystis sp. appeared on several occasions after the bloom of another species, but never at high silicate concentrations. The success of the diatom group seemed to be due to a high inherent growth rate at non-limiting silicate concentrations. Calculations indicated that the inherent growth rate for the diatom group had to be 5 to 50 % higher than for the flagellate group in order to explain the outcome of our experiments. ' Skeletonerna costatum Skeletonema costatum Chaetoceros sp.

A model for nutrient uptake rate (V) in phytoplankton is derived: V = n A v S / ( l + h A v S ) where n is cellular number of uptake sites, A is area of the uptake site, h is time required for handling one nutrient ion, vis the mass transfer coefficient, and S i s substrate concentration. The model is based on 2 time requirements necessary for active uptake of nutrients: (i) time required for realizing encounters with nutrient Ions and (ii) time necessary for handling of ions. The Michaelis-Menten model applied to algal nutrient uptake is a special case of the proposed model, and shortcomings of this model are discussed in view of the more general model. While maximum uptake rate (V,,,,, = n h -l ) may b e considered as a parameter reflecting solely biological constraints, the half-saturation (K, = ( h A v ) -l ) and affinity (a = n AV) depend on both biological and physical constraints. Interrelations between maximum uptake rate, half-saturation constant and affinity are demonstrated and discussed with reference to actual observations. On the basis of the model we hypothesize that maximum uptake rate should increase linearly with the square of cell radius, whlle half-saturation constant and affinity should increase linearly with cell radius. Furthermore, maximum uptake rate should increase exponentlally with temperature, while half-saturation should also increase with temperature, but at a slower rate than maximum uptake rate. Affinity should increase with temperature in the same way as molecular diffusion does.

Sound scattering layers (SSLs) and vertical distribution of zooplankton and micronekton in Masfjorden, western Norway, were studied in January 1988. Two characteristic SSLs were found in the deepest fjord basin (bottom depth 490 m) while single scattering layers were observed in shallower baslns (depths <300 m). The SSLs performed diurnal vertical displacements, exhibiting migrational speeds of up to 90 m h-' The adjustment of the SSLs' day depth was directly proportional to the rate of change in the logarithm of surface light and inversely proportional to the extinction coefficient of the water. Large horizontal differences in biomass in the fjord are related to a combination of advection and changes in migrational behaviour. Feeding activity of the micronekton was higher during light hours. Euphausiids of the deepest basin were mainly feeding on detritus while those in the innermost part of the fjord seemed to be preying upon copepods. The observed vertical distribution of the zooplankton and micronekton were examined in relation to mode of feeding (visual vs tactile), trophic level (omnivory vs carnivory), size (small vs large), and food type and availabihty.

Habitat profitability is analysed with a set of simple equations comprising predation risk, food availability, sex ratio, number of offspring per survivor, generation time and expected reproductive outcome. We show that organisms with fixed generation time should prefer hunger instead of increased feeding if the increase in feeding is associated with the same relative increase in predation risk. On the contrary, feeding is of higher reproductive value for organisms that may canalize increased feeding to a shortening of the generation time. These are more likely to tolerate increased predation risk associated with increased feedlng possibilities. Generation times are often temperature dependent. The andlysis indicates that an increase in environmental temperature may he more beneficial than the same relative decrease in predation risk. The theory is illustrated with data on the copepods Calanus ilnmarchicus and Paracalanus parvus.

Sound scattering layers (SSLs) and vertical distribution of zooplankton and micronekton in Masfjorden, western Norway, were studied in January 1988. Two characteristic SSLs were found in the deepest fjord basin (bottom depth 490 m) while single scattering layers were observed in shallower baslns (depths <300 m). The SSLs performed diurnal vertical displacements, exhibiting migrational speeds of up to 90 m h-' The adjustment of the SSLs' day depth was directly proportional to the rate of change in the logarithm of surface light and inversely proportional to the extinction coefficient of the water. Large horizontal differences in biomass in the fjord are related to a combination of advection and changes in migrational behaviour. Feeding activity of the micronekton was higher during light hours. Euphausiids of the deepest basin were mainly feeding on detritus while those in the innermost part of the fjord seemed to be preying upon copepods. The observed vertical distribution of the zooplankton and micronekton were examined in relation to mode of feeding (visual vs tactile), trophic level (omnivory vs carnivory), size (small vs large), and food type and availabihty. KEY WORDS: Fjord. Light. Micronekton. Migration. Sound scattering layers (SSLs). Zooplankton