Abstract
The critical assumptions of the dilution method for estimating grazing rates of microzooplankton were tested by using a community from the sediment-water interface of Lake Anna, Va. Determination of the appropriate computational model was achieved by regression analysis; the exponential model was appropriate for bacterial growth at Lake Anna. The assumption that the change in grazing pressure is linearly proportional to the dilution factor was tested by analysis of variance with a lack-of-fit test. There was a significant (P < 0.0001) linear (P > 0.05) relationship between the dilution factor and time-dependent change in ln bacterial abundance. The assumption that bacterial growth is not altered by possible substrate enrichment in the dilution treatment was tested by amending diluted water with various amounts of dissolved organic carbon (either yeast extract or extracted carbon from lake sediments). Additions of carbon did not significantly alter bacterial growth rates during the incubation period (24 h). On the basis of these results, the assumptions of the dilution method proved to be valid for the system examined.