Planktonic Food Web Structure along the Sau Reservoir (Spain) in Summer 1997

We studied the planktonic food web in eutrophic Sau Reservoir (Catalonia, NE Spain). Along the longitudinal axis from the Ter River downstream to the dam, we characterized a microbial succession of food web dominance of bacteria‐HNF‐ciliates. The Ter River transports a large load of organic material into the reservoir, with a bacterial density of ∼9 · 10⁶ large cells per ml. While at the first lacustrine station of the Reservoir HNF were the dominant bacterial consumers, at the others, an oligotrich ciliate, Halteria grandinella, was the main protozoan bacterivore. Most of the bacterial production in the reservoir epilimnion was consumed by grazing. The spatial succession of the reservoir microbial food webs was followed downstream by maximum densities of their potential predators among zoo‐plankters – rotifers, and early developmental stages of copepods.     

Microbial Food Webs in an Artificially Divided Acidic Bog Lake

The microbial loop of a naturally acidic bog lake, Große Fuchskuhle (Northeastern Germany), that had been artificially divided into 4 basins, was investigated. In the northeast (NE) and southwest (SW) basins, which differ strongly in chemistry and primary production, we conducted intensive studies of the main carbon fluxes through microbial food webs. In the less acidic, NE basin, much higher phytoplankton as well as bacterial biomass and production were found in parallel with negligible numbers of larger zooplankters. Weakly top-down controlled populations of protists were characterized by an exceptionally low numerical proportion of heterotrophic nanoflagellates (HNF) to ciliates (-1.5-3.5). The ciliate community was dominated by a scuticociliate, Cyclidium sp. (>95% of total ciliates), with an estimated grazing rate equal to 46–80% of heterotrophic bacterial production. In contrast, in the more humic, SW basin, both phyto- and bacterioplankton dynamics seemed to be top-down controlled by abundant populations of small fine-filter feeding cladocerans, Ceriodaphnia quadrangula and Diaphanosoma brachyurum. Consequently, ciliates disappeared from the food web structure of the SW basin, HNF dropped to negligible numbers and bacteria showed very uniform morphology, dominated by small cocci or short rods. Our investigations have shown that the division of the lake into separate compartments can lead to very different microbial food web structures with extreme species compositions.     

The annual cycle of heterotrophic freshwater nanoflagellates: role of bottom-up versus top-down control

The population dynamics of heterotrophic nanoflagellates (HNF)were analyzed in pre-alpine Lake Constance over three consecutiveyears. A recurrent seasonal pattern led to the identificationof five seasonal phases: winter, spring, clear-water, summerand autumn. HNF biomass was lowest in winter and highest m latespring several weeks after the phytoplankton spring bloom. Theaverage biomass of HNF was 5–12% of bacterial biomassand 13–34% of ciliate biomass respectively. The largestHNF cells were recorded during the spring phase, whereas theaverage cell size was reduced to one-third during the subsequentclear-water phase. The pronounced differences in the mean cellsize were attributed mainly to varying grazing impact on HNFThroughout most of the year, HNF production was balanced bygrazing of microzooplankton, namely ciliates, within the microbialloop. During the dear-water phase, however, the grazing impactwas mainly due to rotifers and daphnids. Changing grazing impactwas primarily responsible for the observed 2-fold interannualdifference m the mean biomass of HNF Overall, top-down controlby grazing was more important in governing the population dynamicsof HNF than bottom-up control by bacterial food supply.     

The bacterivory of interstitial ciliates in association with bacterial biomass and production in the hyporheic zone of a lowland stream

Rates of bacteria ingestion by interstitial ciliates were estimated and compared to bacterial biomass and production. Investigation was carried out in the hyporheic zone of a lowland stream. FISH was applied to quantitatively determine bacteria within the ciliate's food vacuoles. To estimate bacteria ingestion rates using FISH, we had to strike a new path. When numbers of bacteria in the food vacuoles remains constant with time (bacterial digestion and ingestion are at equilibrium), ingestion rate can be estimated based on the digestion time and the average number of bacteria per cell. Ciliate community was predominantly composed of bacterivorous ciliates. FISH-signals deriving from ingested bacteria were detected in Cinetochilum margaritaceum, 'other small scuticociliates', Pleuronema spp., and Vorticella spp. Ingestion rates for these taxa were 78, 150, 86, and 38 bacteria ind(-1) h(-1), respectively. The grazing impacts on bacterial biomass and carbon production were calculated based on these ingestion rates. Ciliate grazing caused a decrease in bacterial biomass of 0.024% day(-1) and in bacterial carbon production of 1.60%. These findings suggest that interstitial ciliate grazing impact on bacteria biomass and production was too low to represent an important link in the carbon flow of the hyporheic zone under study.     

Changes in the Epilimnetic Bacterial Community Composition,Production, and Protist-Induced Mortality along the Longitudinal Axis of a Highly Eutrophic Reservoir

We studied changes in the epilimnetic bacterial community composition (BCC), bacterial biomass and production, and protistan succession and bacterivory along the longitudinal axis of the canyon-shaped, highly eutrophic Sau Reservoir (NE Spain) during two sampling campaigns, in April and July 1997. Longitudinal changes in BCC from the river inflow to the dam area of the reservoir were detected by using oligonucleotide probes targeted to the kingdom Bacteria, to the alpha, beta, and gamma subclasses (ALFA, BETA, and GAMA) of the class Proteobacteria, and to the Cytophaga/Flavobacterium (CF) cluster. In general, the inflow of the organically loaded Ter river, with highly abundant allochthonous bacterial populations, induced a clearly distinguishable longitudinal succession of the structure of the microbial food web. The most dynamic changes in microbial parameters occurred at the plunge point, the mixing area of river water and the reservoir epilimnion. Changes within members of BETA and CF were the most important in determining changes in BCC, bacterial abundance and biomass. Much less relevant changes occurred within the less abundant ALFA and GAMA bacteria. From the plunge point downstream, we described a significant shift in BCC in the form of decreased proportions of BETA and CF. This shift spatially coincided with the highest values of heterotrophic nanoflagellate bacterivory (roughly doubled the bacterial production). CF numerically dominated throughout the reservoir without any marked longitudinal changes in their mean cell volume. In contrast, very large cells affiliated to BETA clearly dominated in the allochthonous bacterial biomass brought by the river. BETA showed a marked downstream trend of decreasing mean cell volume. We conclude that the observed BCC shift and the longitudinal shift in food web structure (bacteria-heterotrophic nanoflagellates-ciliates) resulted from highly complex interactions brought about by several major factors: varying hydrology, the high localized allochthonous input of organic matter brought by the river, downstream changing substrate availability, and selective protistan bacterivory.     

Structure and functioning of the microbial loop in a boreal reservoir

The abundance and biomass of the main components of the microbial plankton food web (“microbial loop”)—heterotrophic bacteria, phototrophic picoplankton and nanoplankton, heterotrophic nanoflagellates, ciliates and viruses, production of phytoplankton and bacterioplankton, bacterivory of nanoflagellates, bacterial lysis by viruses, and the species composition of protists—have been determined in summer time in the Sheksna Reservoir (the Upper Volga basin). A total of 34 species of heterotrophic nanoflagellates from 15 taxa and 15 species of ciliates from 4 classes are identified. In different parts of the reservoir, the biomass of the microbial community varies from 26.2 to 64.3% (on average 45.5%) of the total plankton biomass. Heterotrophic bacteria are the main component of the microbial community, averaging 63.9% of the total microbial biomass. They are the second (after the phytoplankton) component of the plankton and contribute on average 28.6% to the plankton biomass. The high ratio of the production of heterotrophic bacteria to the production of phytoplankton indicates the important role of bacteria, which transfer carbon of allochthonous dissolved organic substances to a food web of the reservoir.     

Omnivory in the calanoid copepod Temora longicornis: feeding, egg production and egg hatching rates

We measured in laboratory experiments the ingestion, egg production and egg hatching rates of female Temora longicornis as a function of diet. The diets consisted of a diatom (Thalassiosira weissflogii), an autotrophic dinoflagellate (Heterocapsa triquetra), and a bacterivorous ciliate (Uronema sp.) given as sole foods, or combinations of these single-food items: diatom+dinoflagellate, diatom+ciliate, dinoflagellate+ciliate, and diatom+ciliate+dinoflagellate. For the three single-item diets, the functional response was similar; i.e., ingestion rate increased linearly with food concentration (food range: ∼25 to ∼600 μg C l⁻¹). When all diets were considered, maximum daily carbon ration (∼70% of body weight) was independent of food type. However, the maximum daily egg production rate (12% of body carbon) was obtained with the diatom diet. For all diets, both ingestion and egg production rates increased with food concentration. Ingestion and egg production rates were affected differently by the interaction of food concentration and food type: at low food concentrations, ingestion rates were highest on diets containing the diatom. At high food concentrations, egg production rates were highest on the two phytoplankter diets and their combination. The presence of the ciliate in the diet did not enhance ingestion rate or egg production. Mixed-food diets did not enhance egg production relative to single-food diets. Hence, dietary diversity did not appear to be particularly advantageous for reproduction. Carbon-specific egg production efficiency (EPE; egg production/ingestion) was independent of food concentration and type, and equaled 9%. Egg hatching success was low (mean<30%) and independent of food concentration and type, and egg production rates. Our results are consistent with previous observations that egg production in T. longicornis is enhanced during diatom blooms. However, the relatively low EPE and egg hatching success suggest that reproduction and recruitment in this study were severely constrained by the biochemical composition of the diet, or the physiological condition of the females towards the end of their season of growth in Long Island Sound.     

Identification of Bacterial Micropredators Distinctively Active in a Soil Microbial Food Web

The understanding of microbial interactions and trophic networks is a prerequisite for the elucidation of the turnover and transformation of organic materials in soils. To elucidate the incorporation of biomass carbon into a soil microbial food web, we added ¹³C-labeled Escherichia coli biomass to an agricultural soil and identified those indigenous microbes that were specifically active in its mineralization and carbon sequestration. rRNA stable isotope probing (SIP) revealed that uncultivated relatives of distinct groups of gliding bacterial micropredators (Lysobacter spp., Myxococcales, and the Bacteroidetes) lead carbon sequestration and mineralization from the added biomass. In addition, fungal populations within the Microascaceae were shown to respond to the added biomass after only 1 h of incubation and were thus surprisingly reactive to degradable labile carbon. This RNA-SIP study identifies indigenous microbes specifically active in the transformation of a nondefined complex carbon source, bacterial biomass, directly in a soil ecosystem.     

Towards a budget of leaf litter decomposition in a first-order woodland stream

To construct a budget of carbon transformations occurring during leaf decomposition, alder leaves were placed in a woodland stream, later retrieved at weekly intervals, and rates of fungal and bacterial production, microbial respiration, and release of dissolved organic matter (DOM) and fine particulate organic matter (FPOM) were determined during short laboratory incubations. Carbon dioxide was the major decomposition product, explaining 17% of the microbially mediated leaf mass loss. DOM and FPOM were also important products (5 and 3% of total mass loss, respectively), whereas carbon flow to microbial biomass was low (2%). Fungal biomass in leaves always exceeded bacterial biomass (95–99% of total microbial biomass), but production of bacteria and fungi was similar, indicating that both types of microorganisms need to be considered when examining leaf decomposition in streams. Comparison of leaf mass loss in coarse and fine mesh bags revealed, in addition, that the shredder, Gammarus pulex, had a major impact on leaf decomposition in this study.     

Abundance and biomass of heterotrophic microorganisms in Lake Tanganyika

1. This study focused on heterotrophic microorganisms in the two main basins (north and south) of Lake Tanganyika during dry and wet seasons in 2002. Bacteria (81% cocci) were abundant (2.28–5.30 × 10⁶ cells mL^?1). During the dry season, in the south basin, bacterial biomass reached a maximum of 2.27 g C m^?2 and phytoplankton biomass was 3.75 g C m^?2 (integrated over a water column of 100 m). 2. Protozoan abundance was constituted of 99% of heterotrophic nanoflagellates (HNF). Communities of flagellates and bacteria consisted of very small but numerous cells. Flagellates were often the main planktonic compartment, with a biomass of 3.42–4.43 g C m^?2. Flagellate biomass was in the same range and often higher than the total autotrophic biomass (1.60–4.72 g C m^?2). 3. Total autotrophic carbon was partly sustained by the endosymbiotic zoochlorellae Strombidium. These ciliates were present only in the euphotic zone and usually contributed most of the biomass of ciliates. 4. Total heterotrophic ciliate biomass ranged between 0.35 and 0.44 g C m^?2. In 2002, heterotrophic microorganisms consisting of bacteria, flagellates and ciliates represented a large fraction of plankton. These results support the hypothesis that the microbial food web contributes to the high productivity of Lake Tanganyika. 5. As the sole source of carbon in the pelagic zone of this large lake is phytoplankton production, planktonic heterotrophs ultimately depend on autochthonous organic carbon, most probably dissolved organic carbon (DOC) from algal excretion.     

Responses of Bacterial Communities in Arable Soils in a Rice-Wheat Cropping System to Different Fertilizer Regimes and Sampling Times

Soil physicochemical properties, soil microbial biomass and bacterial community structures in a rice-wheat cropping system subjected to different fertilizer regimes were investigated in two seasons (June and October). All fertilizer regimes increased the soil microbial biomass carbon and nitrogen. Both fertilizer regime and time had a significant effect on soil physicochemical properties and bacterial community structure. The combined application of inorganic fertilizer and manure organic-inorganic fertilizer significantly enhanced the bacterial diversity in both seasons. The bacterial communities across all samples were dominated by Proteobacteria, Acidobacteria and Chloroflexi at the phylum level. Permutational multivariate analysis confirmed that both fertilizer treatment and season were significant factors in the variation of the composition of the bacterial community. Hierarchical cluster analysis based on Bray-Curtis distances further revealed that bacterial communities were separated primarily by season. The effect of fertilizer treatment is significant (P = 0.005) and accounts for 7.43% of the total variation in bacterial community. Soil nutrients (e.g., available K, total N, total P and organic matter) rather than pH showed significant correlation with the majority of abundant taxa. In conclusion, both fertilizer treatment and seasonal changes affect soil properties, microbial biomass and bacterial community structure. The application of NPK plus manure organic-inorganic fertilizer may be a sound fertilizer practice for sustainable food production.     

Trophic interactions among heterotrophic microplankton,nanoplankton, and bacteria in Lake Constance

A considerable portion of the pelagic energy flow in Lake Constance (FRG) is channelled through a highly dynamic microbial food web. In-situ experiments using the lake water dilution technique according to Landry & Hasset (1982) revealed that grazing by heterotrophic nanoflagellates (HNF) smaller than 10 µm is the major loss factor of bacterial production. An average flagellate ingests 10 to 100 bacteria per hour. Nano- and micro-ciliates have been identified as the main predators of HNF. If no other food is used between 3 and 40 HNF are consumed per ciliate and hour. Other protozoans and small metazoans such as rotifers are of minor importance in controlling HNF population dynamics.Clearance rates varied between 0.2 and 122.8 nl HNF^–1 h^–1 and between 0.2 and 53.6 µl ciliate^–1 h^–1, respectively.Ingestion and clearance rates measured for HNF and ciliates are in good agreement with results obtained by other investigators from different aquatic environments and from laboratory cultures. Both the abundance of all three major microheterotrophic categories — bacteria, HNF, and ciliates — and the grazing pressure within the microbial loop show pronounced seasonal variations.     

Trophic relationships between bacteria and protozoa in the hypolimnion of a meromictic mesostrophic lake

The development of ciliated protozoan biomass in the hypolimnion of Piburger See, a small subalpine lake, was demonstrated to depend mainly on two factors. Firstly, the availability of oxygen or nitrate as electron acceptors determines the depth profiles of ciliates. Large quantities of ciliates and even maximum numbers were found at depths where no oxygen could be detected. If nitrate also disappeared during the summer stagnation period, the biomass of protozoa was strongly reduced. Nitrite peaks generally corresponded with ciliate peaks. An extension of Finlay's findings (dissimilatory nitrate reductase within the inner mitochondrial membrane) to other ciliate groups is hypothesized.Secondly, the biomass development of hypolimnetic ciliates was strongly correlated with the bacterial biomass registered approx. 2 weeks before (r² = 0.891, n = 14). The biomass of bacteria, on the other hand, was dependent upon the sedimentation rate of organic carbon (r² = 0.850, n = 15), if a time lag of approx. 2 weeks was taken into account. Therefore a total time lag of approx. 4 weeks was assumed to take place between sedimentation of organic substance and the corresponding increase in ciliate biomass (r² = 0.853, n = 14). Bacteria were shown to be an important intermediate link in the food chain of the hypolimnion. They appear to represent the principal energy source for pelagic ciliates. Sedimentation of organic carbon, nitrogen and phosphorus is the driving force for the establishment of the hypolimnetic microbial community.     

Relationship between bacterial and primary production in a newly filled reservoir: temporal variability over 2 consecutive years

Seasonal and spatial variations in bacterial abundance, biomass and production in a recently flooded reservoir were followed for 2 consecutive years, in conjunction with phytoplankton biomass (chlorophyll a) and activity (primary production). Between the 2 years of the study, the mean value of primary production remained constant, while those of the chlorophyll a concentration, bacterial abundance (BA), bacterial biomass (BB) and bacterial production (BP) decreased. The observed trends of the bacterial variables were linked to changes in the relative importance of allochthonous dissolved organic matter. Moreover, this factor would explain discrepancies observed between the slope of the model II regression equations established from results of the present study and those of the predictive models from the literature, relating to bacterial and phytoplankton variables. An estimate of the carbon budget indicated that 22 and 5% of the ambient primary production in the Sep Reservoir might be channeled through the microbial loop via BP during the 1st and 2nd year of the study, respectively. We conclude that heterotrophic BP in the Sep Reservoir may, on occasion, represent a significant source of carbon for higher order consumers.     

Food Web Structure in the Recently Flooded Sep Reservoir as Inferred From Phytoplankton Population Dynamics and Living Microbial Biomass

Phytoplankton dynamics, bacterial standing stocks and living microbial biomass (derived from ATP measurements, 0.7-200 mm size class) were examined in 1996 in the newly flooded (1995) Sep Reservoir ('Massif Central,' France), for evidence of the importance of the microbial food web relative to the traditional food chain. Phosphate concentrations were low, N:P ratios were high, and phosphate losses converted into carbon accounted for <50% of phytoplankton biomass and production, indicating that P was limiting phytoplankton development during the study. The observed low availability of P contrasts with the high release of "directly" assimilable P often reported in newly flooded reservoirs, suggesting that factors determining nutrient dynamics in such ecosystems are complex. The phosphate availability, but also the water column stability, seemed to be among the major factors determining phytoplankton dynamics, as (i) large-size phytoplankton species were prominent during the period of increasing water column stability, whereas small-size species dominated phytoplankton assemblages during the period of decreasing stability, and (ii) a Dinobryon divergens bloom occurred during a period when inorganic P was undetectable, coinciding with the lowest values of bacterial standing stocks. Indication of grazing limitation of bacterial populations by the mixotrophic chrysophyte D. divergens (in late spring) and by other potential grazers (mainly rotifers in summer) seemed to be confirmed by the Model II or functional slopes of the bacterial vs phytoplankton regressions, which were always <0.63. Phytoplankton biomass was not correlated with phosphorus sources and its contribution was remarkably low relative to the living microbial biomass which, in contrast, was positively correlated with total phosphorus in summer. We conclude that planktonic microheterotrophs are strongly implicated in the phosphorus dynamics in the Sep Reservoir, and thus support the idea that an important amount of matter and energy flows through the "microbial loop" and food web, shortly after the flooding of a reservoir.     

Bacterial Production in the Recently Flooded Sep Reservoir: Diel Changes in Relation to Dissolved Carbohydrates and Combined Amino Acids

The spatial distribution of bacterial abundance and production were measured every 4 h in a recently flooded oligo-mesotrophic reservoir (the Sep Reservoir, Puy-De-Dôme, France), in relation to concentrations of dissolved carbohydrates and combined amino acids. The concentration of dissolved organic matter (DOM) components in the recently flooded Sep Reservoir were higher than those measured in other lakes of similar trophic status. Short-term variations in the bacterial production in this new reservoir appeared cyclical and endogenous to bacterial communities. These results highlight the need for the evaluation of diel changes in bacterial production, if estimation of the daily production rate of bacteria is to be done accurately for a reliable model of carbon flow through bacterioplankton and ultimately through aquatic microbial food webs. Bacterial growth, measured over time and space, did not appear exclusively governed by DOM components from phytoplankton primary production.     

Relationship between bacteria, phytoplankton and heterotrophic nanoflagellates along the trophic gradient

Bacterial and heterotrophic nanoflagellates (HNF) abundance, as well as bacterial production and chlorophylla levels, were measured at five sites extending from the coastal zone toward the open Adriatic in the period from March to October 1995. The investigated areas were grouped into trophic categories according to concentrations of chlorophylla. All the biotic-para-meters increased along the trophic gradient, leading to eutrophy, but they did not increase at the same rate. The bacterial biomass: phytoplankton biomass (BB: chla) ratio decreased from about 10 in the very oligotrophic area to 0.8 at the eutrophic site. In contrast, the bacterial abundance: HNF abundance ratio (B: HNF) increased from 1000 bacteria per 1 flagellate in the oligotrophic system to 1700 bacteria flagellate⁴ in the eutrophic area. Decreasing BB: chla and increasing B: HNF ratios along the trophic gradient might reflect the different structures of the microbial food web. Relationships between bacterial abundance and production, and chla and HNF showed that bacterial abundance along the trophic gradient was regulated by the interplay between nutrient supply and grazing pressure. But in the oligotrophic system, bacterial abundance was more closely related to bacterial production and chla than in the eutrophic system, suggesting stronger control of bacterial abundance by substrate supply. On the other hand, the coupling between bacteria and HNF, and uncoupling between bacterial abundance and production in the eutrophic system, showed that the importance of bacteriovory increased in richer systems.     

Identification of bacterial micropredators distinctively active in a soil microbial food web

The understanding of microbial interactions and trophic networks is a prerequisite for the elucidation of the turnover and transformation of organic materials in soils. To elucidate the incorporation of biomass carbon into a soil microbial food web, we added 13C-labeled Escherichia coli biomass to an agricultural soil and identified those indigenous microbes that were specifically active in its mineralization and carbon sequestration. rRNA stable isotope probing (SIP) revealed that uncultivated relatives of distinct groups of gliding bacterial micropredators (Lysobacter spp., Myxococcales, and the Bacteroidetes) lead carbon sequestration and mineralization from the added biomass. In addition, fungal populations within the Microascaceae were shown to respond to the added biomass after only 1 h of incubation and were thus surprisingly reactive to degradable labile carbon. This RNA-SIP study identifies indigenous microbes specifically active in the transformation of a nondefined complex carbon source, bacterial biomass, directly in a soil ecosystem.     

Trophic relationships between planktonic bacteria,heterotrophic nanoflagellates and viruses in a mesoeutrophic reservoir

The abundance, biomass, and production (Р _В) of bacrerioplankton; the taxonomic composition, abundance, biomass of heterotrophic nanoflagellates (HNF) and the rate of consumption of bacteria by HNFs; and the abundance of virioplankton, frequency of visibly infected bacterial cells, virus-induced mortality of bacterioplankton, and viral production were estimated in the mesoeutrophic Rybinsk Reservoir. The rate of bacterial mortality due to viral lysis (7.8–34.1%, on average 17.2 ± 2.0% of daily Р _В) was lower than the consumption of bacteria by the HNF community (15.4–61.3%, on average 32.0 ± 4.2% of daily Р _В). While consuming bacteria, HNFs simultaneously absorbed a significant number of viruses residing on the surface and inside the bacterial cells.     

Qualitative importance of the microbial loop and plankton community structure in a eutrophic lake during a bloom of cyanobacteria

Plankton community structure and major pools and fluxes of carbon were observed before and after culmination of a bloom of cyanobacteria in eutrophic Frederiksborg Slotssø, Denmark. Biomass changes of heterotrophic nanoflagellates, ciliates, microzooplankton (50 to 140 μm), and macrozooplankton (larger than 140 μm) were compared to phytoplankton and bacterial production as well as micro- and macrozooplankton ingestion rates of phytoplankton and bacteria. The carbon budget was used as a means to examine causal relationships in the plankton community. Phytoplankton biomass decreased and algae smaller than 20 μm replacedAphanizomenon after the culmination of cyanobacteria. Bacterial net production peaked shortly after the culmination of the bloom (510 μg C liter^?1 d^?1 and decreased thereafter to a level of approximately 124 μg C liter^?1 d^?1. Phytoplankton extracellular release of organic carbon accounted for only 4–9% of bacterial carbon demand. Cyclopoid copepods and small-sized cladocerans started to grow after the culmination, but food limitation probably controlled the biomass after the collapse of the bloom. Grazing of micro- and macrozooplankton were estimated from in situ experiments using labeled bacteria and algae. Macrozooplankton grazed 22% of bacterial net production during the bloom and 86% after the bloom, while microzooplankton (nauplii, rotifers and ciliates larger than 50 μm) ingested low amounts of bacteria and removed 10–16% of bacterial carbon. Both macro-and microzooplankton grazed algae smaller than 20 μm, although they did not control algal biomass. From calculated clearance rates it was found that heterotrophic nanoflagellates (40–440 ml^?1) grazed 3–4% of the bacterial production, while ciliates smaller than 50 μm removed 19–39% of bacterial production, supporting the idea that ciliates are an important link between bacteria and higher trophic levels. During and after the bloom ofAphanizomenon, major fluxes of carbon between bacteria, ciliates and crustaceans were observed, and heterotrophic nanoflagellates played a minor role in the pelagic food web.