The trophic state of lakes is commonly defined by the concentration of nutrients in the water column. High nutrient concentrations generate high phytoplankton production, and lakes with low nutrient concentrations are considered low-productive. This simplified view of lake productivity ignores the fact that benthic primary producers and heterotrophic bacteria can be important basal producers in lake ecosystems. In this thesis I have studied clear-water and brown-water lakes with respect to primary production, respiration and bacterial production based on allochthonous organic carbon. These processes were quantified in pelagic and benthic habitats on temporal and spatial scales. I also calculated the net ecosystem production of the lakes, defined as the difference between gross primary production (GPP) and respiration (R). The net ecosystem production indicates whether a lake is net heterotrophic (GPP < R), net autotrophic (GPP > R) or in metabolic balance (GPP = R). Net heterotrophic lakes are sources of carbon dioxide (CO2) to the atmosphere since respiration in these lakes, by definition, is subsidized by an external organic carbon source. External organic carbon is transported to lakes from the terrestrial environment via inlets, and can serve as a carbon source for bacteria but it also limits light availability for primary producers by absorbing light. On a seasonal scale, four of the clear-water lakes studied in this thesis were dominated by primary production in the soft-bottom benthic habitat and by respiration in the pelagic habitat. Concentrations of dissolved organic carbon (DOC) were low in the lakes, but still high enough to cause the lakes to be net heterotrophic. However, the lakes were not low-productive due to the high production in the benthic habitat. One of the clear-water lakes was studied also during the winter and much of the respiration under ice was supported by the benthic primary production from the previous summer. This is in contrast to brown-water lakes where winter respiration is suggested to be supported by allochthonous organic carbon. By studying lakes in a DOC gradient (i.e. from clear-water to brown-water lakes) I could draw two major conclusions. The lakes became less productive since benthic primary production decreased with increasing light extinction, and the lakes became larger sources of CO2 to the atmosphere since pelagic respiration was subsidized by allochthonous organic carbon. Thus, lake carbon metabolism can have an important role in the global carbon cycle due to their processing of terrestrial organic carbon and to their possible feedback effects on the climate system.