Bacterivory in eutrophic Lake Oglethorpe, Georgia, was determined by direct observation of tracer particle uptake by all members of the planktonic community. Heterotrophic flagellates dominated grazing at all times, accounting for 49-8 1% of grazing on an areal basis and up to 98% of grazing at some depths. Pigmented (mixotrophic) flagellates were major grazers during winter and spring blooms, when they contributed up to 45% of community grazing on an areal basis and 79% at depths of maximum abundance. In late spring to early summer, rotifers and ciliates were responsible for as much as 25 and 30% of bacterivory at some depths, but averaged 3 and 11% over the year, respectively. Grazing impact of cladoceran crustaceans was generally < 1% of the total. Bacterivory by copepods was not detected. Total bacterial mortality due to grazing ranged from 11 to 162% of bacterial cell production estimated from the incorporation of [3H]thymidine.Use of 0.57-pm microspheres as tracers gave similar estimates ofingestion to fluorescently labeled bacteria in this system. The use of 4% ice-cold glutaraldehyde mixed 1 : 1 with the water sample was found to be equally effective to an acrolein-tannic acid mixture for reducing egestion of particles by protists. We therefore consider our findings to be representative of bacterivory by the planktonic community in this eutrophic system.

Despite differences in the species compositions and absolute abundances of planktonic microorganisms in fresh-and saltwater, there are broad similarities in microbial food webs across systems. Relative abundances of bacteria and nanoplanktonic protozoa (HNAN, primarily heterotrophic flagellates) are similar in marine and freshwater environments, which suggests analogous trophic relationships. Ranges of microbe abundances in marine and fresh waters overlap, and seasonal changes in abundances within an ecosystem are often as great as differences in abundances between freshwater and marine systems of similar productivities. Densities of bacteria and heterotrophic nanoplankton, therefore, are strongly related to the degree of eutrophication, and not salt per se. Data from the literature is compiled to demonstrate a remarkably consistent numerical relationship (ca 1000 bacteria: l HNAN) between bacterioplankton and HNAN from the euphotic zones of a variety of marine and freshwater systems. Based on the results of a simple food web model involving bacterial growth, bacterial removal by HNAN, predation on HNAN, and the observed relationships between bacterial and HNAN abundances in natural ecosystems, it is possible to demonstrate that bottom-up control (food supply) is more important in regulating bacteiial abundances in oligotrophic environments while top-down control (predation) is more important in eutrophic environments.

Arranging organisms into functional groups aids ecological research by grouping organisms (irrespective of phylogenetic origin) that interact with environmental factors in similar ways. Planktonic protists traditionally have been split between photoautotrophic "phytoplankton" and phagotrophic "microzooplankton". However, there is a growing recognition of the importance of mixotrophy in euphotic aquatic systems, where many protists often combine photoautotrophic and phagotrophic modes of nutrition. Such organisms do not align with the traditional dichotomy of phytoplankton and microzooplankton. To reflect this understanding, we propose a new functional grouping of planktonic protists in an eco-physiological context: (i) phagoheterotrophs lacking phototrophic capacity, (ii) photoautotrophs lacking phagotrophic capacity, (iii) constitutive mixotrophs (CMs) as phagotrophs with an inherent capacity for phototrophy, and (iv) non-constitutive mixotrophs (NCMs) that acquire their phototrophic capacity by ingesting specific (SNCM) or general non-specific (GNCM) prey. For the first time, we incorporate these functional groups within a foodweb structure and show, using model outputs, that there is scope for significant changes in trophic dynamics depending on the protist functional type description. Accordingly, to better reflect the role of mixotrophy, we recommend that as important tools for explanatory and predictive research, aquatic food-web and biogeochemical models need to redefine the protist groups within their frameworks.

Utilization of dissolved organic nitrogen (DON) from natural (forests) and anthropogenic (animal pastures, urban/ suburban storm water runoff) sources (three sites per source) by estuarine plankton communities was examined in spring, summer, and fall. The proportion of DON utilized ranged from 0 to 73%. Overall, urban/suburban storm water runoff had a higher proportion of bioavailable DON (59% Ϯ 11) compared to agricultural pastures (30% Ϯ 14) and forests (23% Ϯ 19). DON bioavailability varied seasonally; however, the seasonal pattern differed for the three sources. Bacterial production increased linearly with the amount of DON utilized across all sources and seasons; the rate of increase was approximately five times greater per micromole of N as DON used relative to dissolved inorganic N (DIN) used. Although phytoplankton production generally increased with DON addition, the increased production was not correlated with the amount of DON utilized, suggesting that a variable portion of dissolved organic matter (DOM)-N was directly or indirectly available to the phytoplankton. This indicates that phytoplankton production is not a good measure of the amount of bioavailable DON, and measurements of the amount of bioavailable DON based on bacterial responses alone might not reflect N available to phytoplankton. Preliminary seasonal budgets of bioavailable N (DIN plus bioavailable DON) as a function of land use suggest that ϳ80% of the total dissolved N (TDN) from urban/suburban runoff is bioavailable, whereas a lower proportion (20-60%) of TDN is bioavailable from forests and pastures. N budgets for aquatic ecosystems based on only DIN loading underestimate bioavailable N loading, whereas total N or TDN budgets overestimate bioavailable N inputs.

The bioavailibility of dissolved organic nitrogen (DON) in river water entering estuaries was examined for the Delaware and Hudson Rivers, USA. Water collected from above the salinity intrusion zone of each river was filtered, brought to a salinity of 15 ppt, and inoculated with estuarine bacteria. Bacterial production rates (8 to 26 X 105 cells ml-' d-') during the initial 2 d in these experiments were within the range measured in these and other estuaries, indicating that riverine dissolved organic matter can contribute to production of estuarine bacteria. Average DON concentrations decreased by 40 to 72 % within the 10 to 15 d time course of the experiments; the decreases in DON were accounted for by increases in microbial biomass plus remineral~zation to inorganic nitrogen. The time scale over which DON was utilized suggests that in estuaries with residence times on the order of weeks to months, such as Delaware Bay, river inputs of the biologically available portion of DON are first utilized within the estuary. In contrast, in estuaries with residences times of less than a week, such as New York Bay, a portion of the biologically available DON may be utilized first within the estuary, with the rema~nder exported and utilized In continental shelf waters The large proportion of the DON that was biolog~cally available in these experiments, coupled with the knowledge that inputs of organic nitrogen can account for 20 to 90% of the total nitrogen loading to estuaries, suggests that organic nitrogen inputs may contribute more to estuarine and shelf eutrophication than was previously suspected. These experiments demonstrate that dissolved inorganic nitrogen (DIN) inputs underestimate, and total nitrogen lnputs likely overestimate, the inputs of biologically available N to estuaries. In order to develop a 'biolog~cally available nitrogen budget' for an ecosystem, DIN inputs, plus that portion of the organlc N that is biologically available must be quantified.