Phytoplankton-bacteria relationship in the Antarctic marine ecosystem GILLES B l L L E N and SYLVIE BECQUEVORT Billen, G . & Becquevort, S. 1991: Phytoplankton-bacteria relationship in the Antarctic marine ecosystem. Pp. 245-253 in Sakshaug. E . . Hopkins, C. C. E. & Britsland. N. A . (eds.): Proceedings of the Pro Mare Symposium on Polar Marine Ecology, Trondheim, 12-16 May 1990. Polar Research l O ( 1 ) . Time series data of phytoplanktonic and bacterial biomass during the ice retreat period at different latitudes in the Weddell Sea and Prydz Bay areas show a distinct delay in the development of bacteria with respect to phytoplankton. Use of a general ecophysiological model of bacterial growth, along with direct in situ measurements of growth and mortality rates, allowed the simulation of the observed timing of bacterial development. I t is suggested that the uncoupling between phytoplanktonic and bacterial development at the earliest stage of the spring ice-edge related algal bloom is not the result of the low temperatures occurring in the Southern Ocean but rather is due to the macromolecular nature of thc dissolved organic matter released from phytoplankton. Gilles Billen and Syluie Becqueuort. Groupe de Microbiologie des Millieux Aquatiques, University of Brussels, C P 221. Campus de la Plaine, B-1050 Brussels, Belgium. Introduction That planktonic bacteria directly depend on phy- toplankton in aquatic systems devoid of alloch- tonous input of organic matter is now widely recognized. Evidence for this statement is pro- vided by the correlation which exists between bacterial biomass and chlorophyll u concentration (Linley et al. 1983; Bird & Kalff 1984) or between bacterial production and both chlorophyll u con- centration and planktonic primary production (Cole et a]. 1988), over a large range of temperate and tropical aquatic environments. However, the exact nature of the trophic relationship between phytoplankton and bacteria is still a matter of controversy. Exudation by algae of low molecular weight photosynthates has long been considered as the dominant dissolved organic carbon (DOC) flux to heterotrophic bacteria (Larsson & Hag- strom 1979; Wolter 1982; M d l e r Jensen 1983). Recently, however, BjGrnsen (1989) suggested that exudation should be interpreted as a passive leakage of small metabolites across cellular mem- branes and should represent no more than 1- 5% of phytoplankton biomass per day (0.0005- 0.002 h-l). Other processes producing DOC therefore probably account for an additional and often dominant part of the organic matter transfer to the bacterial compartment, as e.g. direct lysis of phytoplankton cells (Jassby & Goldman 1974; Billen 1984) or rapid diffusion from incompletely digested fecal material of macro- or micrograzers (the so-called sloppy feeding of Lampert 1978) (Jumars et al. 1989). These processes mainly pro- duce dissolved organic substances with high molecular weight which require extracellular hydrolysis before they can be taken up by bac- terioplankton. For this reason, a looser coupling between bacteria and phytoplankton can be expected in situations where the latter processes of DOC-supply are dominant rather than in situa- tions where bacteria live mainly on algal exu- dation products (Billen 1990). How these ideas apply to polar systems at low temperatures is the subject of this paper. Con- tradictory observations have been published in previous literature concerning the role of bacteria in Arctic and Antarctic marine ecosystems. Some authors (Hodson et al. 1981; Azam et al. 1981; Hanson & Lowery 1983; Hanson et al. 1983; Sullivan et ai. 1990) reported measurements of bacterial activities in the Antarctic ocean of the same order of magnitude as those observed i n temperate seas. They concluded that the microbial loop performs a similarly important role in the Antarctic marine system as in tem- perate and tropical systems. On the other hand, situations with high phytoplanktonic biomass and low bacterial numbers and activities have been reported in both the Arctic and Antarctic systems (Kriss et al. 1969; Sorokin 1971; Mullins & Priddle 1987; Davidson 1985; Pomeroy et al. 1990). Pomeroy & Deibel (1986) explained this by a dramatic decrease of bacterial activity below 2”C, while algae continued to be active at these tem- peratures. This could explain the preservation of DOC, which has sometimes been found in high 246 G . Billen & S. Becqitevort concentrations in Antarctic waters (Bolter & Dawson 1982). This organic matter would then be exported to lower latitudes by deep oceanic circulation (Sorokin 1971). As observations of low bacterial activity most often concern early spring situations, a possible explanation reconciling these contradictory observations may be that a time lag exists in the response of bacterioplankton t o algal develop- ment. Only rather fine time-series observations could confirm this hypothesis. Such time series are difficult to obtain, however, for obvious logistic reasons. In this paper we present such data on phytoplankton and bacterioplankton develop- ment obtained during the period of ice retreat in two sectors of the Antarctic Ocean. We discuss them in the light of a simplified ecophysiological model of bacterioplankton development. Biotopes and methods The two data sets used in this paper result from the participation of Belgian teams, in an Aus- tralian cruise to Prydz Bay area (Marine Science Voyage 7 of the Australian Antarctic Division on board of the M / V N E L L A D A N ) and in the European EPOS cruise to the Weddell Sea, on board the R/V POLARSTERN Fig. 1). The Prydz Bay area was first visited in mid-January 1987 by Joiris et al. ing the French Indigo 111 cruise. At that time, the position of the ice edge was about 65"s. From mid- February to end-March 1987, the area between 62 and 66"s was free of ice. In t h e Weddell Sea area, several successive transects along the meridian 49"W were sampled from mid-November 1988 to the beginning of January 1989. The ice edge location varied during this period from about 60"s to 61.5"s. Chlorophyll a determinations were carried out according to Lorenzen (1967). Conversion into carbon biomass was made by using a constant C/ Chlorophyll a ratio of 30. This represents the median value derived from comparison of chloro- phyll a determination and algal carbon content calculated from microscopic cell size deter- mination using Edler's (1979) formulas (Mathot pers. conim.). At most stations, samples were taken at 10,20, 40,60. and 120 m depth. The results reported here are the means of the duplicate determinations performed at the depths located in the upper (1987) on board the R / V M A R I O N D U F R E S N E dur- krn Fig. 1. The two areas in the Southern Ocean where the data sets discussed in this paper were collected. homogeneous layer, as determined from the CTD vertical profile (range 15-5011-1). For only a few stations, samples were only taken at 10 m depth. In these cases the results reported are the means of the duplicate determinations. Bacteria were enumerated by epifluorescence microscopy after either acridine orange (Hobbie et al. 1977) (Prydz Bay data) or DAPI staining (Porter & Feigh 1980) (Weddell Sea data). Bio- volumes were estimated on enlargements of microphotographs. Conversion into carbon bio- mass was done using the biovolume dependent C/biovolume ratio proposed by Simon & Azam (1989). Differences in the bacterial biomass val- ues reported here for the Prydz Bay area with previously reported values (Lancelot et al. 1989) originate from the use of this new conversion factor. Bacterial production rates were measured according to the 3H-thymidine incorporation method of Fuhrman & Azam (1982). 3H-thy- midine was added at a final concentration of 10 nmol I - ' , a concentration shown saturating for the process of incorporation. The uptake of radioactivity in the cold TCA fraction was deter- mined after 3-4 hours incubation. Empirical cali- bration with cell number increase in 0.2ym filtered seawater reinoculated with 2 p n filtered water (Riemann et al. 1987) was carried out sev- eral times at both sites. The conversion factors Phytoplankton-bacteria relationship 241 Da. iaa.31 were 5.0.10' and 1.25. lo9 cells/nmol thymidine incorporated in the Prydz Bay area and the Wed- dell Sea, respectively. Maximum growth rate of bacteria was measured after addition of a mixture of amino acids and monosaccharides in the following final concentrations: glucose (4 mg I - I ) , galactose (4 mg I-'), casamino acids (15 mg I - [ ) , am- monium acetate (1.5 mg ]-'), ammonium hydro- genophosphate (2.5 mg 1-I). Bacterial mortality was estimated according to the method developed by Servais et al. (1985, 1989). The part of grazing in overall mortality rate was estimated by comparing the mortality rate of an untreated sample with a sample filtered through a 2 p m membrane and treated with a colchicine and cycloheximide (concentration 100 and 200 mg I-', respectively) (Becquevort et al. unpublished data). Results and discussion Relationship between bacterial and phytoplanktonic biomass When plotted against each other, no distinct relationship is apparent between bacterial and phytoplankton biomass (Fig. 2a) as would be expected in the case of a direct response of bac- teria to phytoplankton. Data from successive situations in the same area, however, describe counterclockwise trajectories, indicating that phytoplankton development precedes bacterial response by about 15-30 days (Fig. 2b). This "1 0 % 1 0 i chlorophyll a, pg.1-l Fig. 2 A . Bacterial biomass plotted against chlorophyll a values in the Weddell Sea (0) and Prydz Bay ( 0 ) area during the period of ice retreat. Weddell Sea 590503 49oW Dec. 1-20 Nor. 26-34 1 2 chlorophyll a, pg.1-I Weddell Sea 5 9 6 49oW 7 k. 1-10 & - Da. 1-10 Nor. -0 s .j 10. E 3 f . Pry& Bay 7 chlorophyll a, pg.1-I Fig. 2 B . Counterclockwise trajectories of bacterial and phy- toplankton biomass values at 3 sites in the Southern Ocean. The counterclockwise trajectories indicate a delayed response of bacteria to algal development. 248 delay in the bacterial response to phytoplankton is quite clear in the Prydz Bay data set, which shows a clear chlorophyll maximum in mid-Janu- ary, while bacterial biomass peaks only about o n e month later (Fig. 3 ) . In the Weddell Sea data s e t , a delay is also quite apparent in the response of bacteria to the algal bloom which follows the southward retreat of the ice edge (Fig. 4). G . Billen & S . Becquevort Growth and mortality rates of bacterioplankton Specific growth rate of bacteria, as estimated by the ratio between thymidine incorporation measurements during 4 hours incubation at in situ temperature and bacterial biomass, varied between 0.05 and 0.001 h - l . A n immediate increase of this growth rate was observed when amending the samples with a mixture of amino acids and monosaccharides at a final concen- trationof 10 mgC I - ' and a C : N : P ratio of 1 6 : 4 : 1 by weight. W e considered that the value of the specific growth rate measured with this amend- ment represents the maximum growth rate of the natural bacterial assemblage. Use of short incubation times ( 4 hours) minimises the risk of modifications in the bacterial populations. Fig. 5 shows the value of this maximum growth rate measured at different incubation tempera- ture for samples from the Weddell Sea and the Prydz Bay area. Also shown a r e the results of similar measurements carried o u t in the Belgian coastal zone of the North Sea (Billen 1990). Clearly, the temperature range of the Southern PrydzBay 63-666 0 0 2 0 -z 0 0 0 0 January February March Fig. 3. Temporal variations of chlorophyll a ( 0 ) and bacterial biomass (0) during the 1987 ice-retreat period in an area off Prydz Bay between 62 and 64"s. 5 3, (i 1 0 4 , Ol Weddell Sea (490W) Nov 2&30,1988 & - c 0 Weddell Sea (490W) Nov 2&30,1988 h 15 0 Fig 4 . N-S Iransects of chlorophyll a ( 0 ) and bacterial biomass (0)valuesobserved at different tirnesduringthe 1989ice retreat period in the Weddell Sea area (meridian 49"W). Ocean ( - 2 t o + 2 T ) is suboptimal for bacterial growth, in contrast t o what is observed for phy- toplankton growth (see e.g. Lancelot et al. 1989). Nevertheless, when compared with the tem- perature response of bacterial communities at temperate latitudes, a remarkable adaptation t o lower temperatures is found in Antarctic bacteria, with an optimum temperature 18°C lower with respect to the former. Antarctic bacteria a t nega- tive temperatures a r e able to grow as fast as North Sea bacteria in early spring conditions (8-10°C). Measured rates of bacterial mortality in the two areas studied a r e remarkably constant, v'irying between 0.002 and 0.005 h - ' in the range - 2 t o +2"C. A clear effect of temperature o n mortality rate has been observed (Fig. 6). Grazing by proto- Phytoplankton-bacteria relationship 249 temperature, 0C Fig. 5. Maximum specific growth rate of natural assemblages of antarctic bacteria as determined by short term thymidine incorporation measurement after amendment with direct mono- meric substrates. as a function of incubation temperature. 0 = Prydz Bay; W = Weddell Sea. Also shown are the results of similar determinations carried out on bacterial assemblages from the Southern Bight of the North Sea ( A ) . zoans contributed to 22 t o 100% of the total bacterial mortality. Modelling bacterial response to phytoplankton development In order to better understand the causes of the delay observed in the bacterial response to phytoplanktonic development in the Antarctic ecosystem, we used an idealised model of bac- terioplankton dynamics, described and justified in details elsewhere (Servais 1986; Billen & Fon- tigny 1987; Billen & Servais 1989; Billen 1991). The basic assumption of this model is that bac- terial growth rate is directly dependent on the concentration of small, monomeric substrates, while most of the DOC is under the form of macromolecules requiring extracellular hydroly- sis. Uptake of direct substrates is assumed to obey an overall Michaelis-Menten kinetics (Parsons & Strickland 1962; Wright & Hobbie 1965). A constant fraction Y of the amount of substrate taken up is used for biomass production, the remaining part being respired (Servais 1986). Extracellular hydrolysis of biopolymers was also shown to obey a Michaelis-Menten kinetics (Som- ville & Billen 1983; Somville 1984). Bacterial exoenzymes are mostly attached to bacterial envelopes and are present in constant amount with respect to biomass (Fontigny et al. 1987). Servais (1986) showed that it is possible to des- cribe the bacterial utilisation of phytoplanktonic derived organic matter by assuming that it is made of two fractions ( H I , H Z ) with different sus- ceptibilities to extracellular hydrolysis, hence dif- ferent parameters of their Michaelis-Menten degradation kinetics. The process of bacterial mortality can be represented, as a first approxi- mation, by a first order kinetics (Servais et al. 1985). The following equations can therefore be writ- ten for describing the dynamics of bacterial growth: B dS _ - HI + KHI dt -elmax dB S - = Yb,,, - B - k d B dt S + K, (4) where elmax, eZmax are the maximum rates of polymers hydrolysis per unit bacterial biomass; K H I , KH2 are the half-saturation constants of polymer hydrolysis; pH1, pH2 are the rate of production of poly- mers through phytoplankton lysis; bmax is the maximum rate of substrate uptake by bacteria; Ks is the half-saturation constant of sub- strate uptake; Y is the growth yield; kd is the mortality constant; Pex is the rate of production of direct monomeric substrate through phy- toplankton exudation. The value of most of the parameters involved in these equations has been determined experi- mentally. From the data shown in Fig. 5, the maximum growth rate of bacteria (pmaX), and its temperature dependency can be estimated as follows: pmax(T) = 0.18 h - ' * 0.1 + 0.9* { TOPI = 12°C exp [ - (s)2]} Ti - Topt with Ti = 5°C 250 G . Billen & S. Becquevorr b 0 0 m 20 ternperature,'C Fig. 6. Grazing rate of natural bacterial assemblages of the Weddell Sea (0) and the Prydz Bay (D) as a function of temperature. T h e value of Y . in the absence of nutrient limitation, is generally close to 0.3 (Servais et al. 1987). b,,, is calculated as h a x / Y . K, is taken arbitrarily as 0.01 mgC I - I , as this parameter does not significantly influence t h e results of the cal- culations. T h e parameters characterising the rate of exoenzymatic hydrolysis of macromolecules from phytoplanktonic origin have been estimated as follows, o n the basis of experiments in which the kinetics of bacterial degradation of a sonicated and filtered algal culture is measured after inocu- lation with a natural assemblage of bacteria (Ser- vais 1986; Billen 1990): e l m a x (Top,) = 0.75 h - ' , K H , = 0.1 mgC.1-I ermax(Topt) = 0.25 h - ' , KH2 = 2 . 5 mgC.1-I The same dependence t o temperature ( T ) as described above for pmax has been considered. O n the basis of experimental determinations of the total mortality rate constant kd, the following relationship with temperature has been considered: k d = 0.004 h - ' + 0.0005 h - l . T. O n the other hand, it has been assumed that organic matter is produced by phytoplankton through two distinct processes: (i) exudation, pro- ducing low molecular weight substrates ( S ) a t a rate proportional t o its biomass (kc,); and (ii) lysis and/or sloppy feeding, producing high molecular weight biopolymers of two classes of utilisability ( H I , H z ) in a ratio 1 : 1. In t h e basic simulation, k,, has been maintained a t t h e lowest value sug- gested by B j ~ r n s e n (1989) (0.0005 h-l). T h e rate of D O C production through lysis or sloppy feed- ing is t h e only adjustable parameter of the model. Forcing functions for the simulation a r e the observed variation of water temperature and phytoplankton biomass. Figure 7 shows that t h e general trends of the observed bacterioplanktonic response to phyto- plankton development both in Prydz Bay and in the Weddell Sea can be satisfactorily simulated with a single value of the first o r d e r rate constant of macromolecular DOC production by phyto- plankton (kly, = 0.004 h - l ) . In particular t h e model predicts a lag of about 10 t o 30 days between the peaks of algal and bacterial biomass, as observed. T h e rates of bacterial production calculated by the model a r e compared in Fig. 8 with the measured values. Agreement is generally within a factor of 2, and again, the major trends of geographical and seasonal variations a r e correctly simulated. A s t h e measurements of bacterial bio- mass and production rates are quite independent of each other, this overall agreement constitutes a further validation of the model. In order to obtain further insight into the factor responsible for the time lag between phyto- plankton and bacterioplankton, the effect of vary- ing some parameters or forcing variables in the model were tested. Fig. 9 shows the effect of increasing t h e temperature to a constant value of 5°C: the timing of the bacterial development is hardly modified. O n t h e o t h e r hand, if the rate k,, of algal exudation of low molecular weight D O C is increased (at t h e expense of the pro- duction rate of macromolecular material, klYt), the model predicts a much more rapid bacterial response. Conclusion Our observations in the Prydz Bay and Weddell Sea areas have shown clearly that a significant time lag exists in the response of bacterioplankton to the early spring phytoplankton development. Application of our model of bacterioplankton Phytoplankton-bacteria relationship 251 . Weddell Sea(l9aW) "1 WeddellSea(49.W) 1% Jan Feb Mar 5o 59diJS 25 0 4 61405 [25 \ . / 0 , , o 5 6 2 6 10 0 , * . 63-666 a j , , Sa, 0 0 Nov Dec Jan Feb M a r Apr Fig. 7. Simulation of the seasonal variations of bacterial biomass in response to observed phytoplankton development (dotted curve) at different latitudes. according to the HSB model, for a single value of the rate of phytoplankton macromolecular DOC release (klyr = 0.004 h - ' ) and a rate of phytoplankton exudation of 0.0005 h - ' . The black dots represent the observed values of bacterial biomass. 0- Nov Dec Jan Feb Mar . 59030s . A 6 0 6 1 4 B 'l1 fi0030S 6 1 6 .05 61030s . 0 . .05 6 2 6 0 , '1 PrydzBay 63-666 0 I , , ,A, Nov Dec Jan Feb Mar Fig. 8. Simulation of the seasonal variations of bacterial pro- duction rate in response t o observed phytoplankton dev- elopment (dotted curve) at different latitudes, according to the HSB model, for the same values of the parameters as in Fig. 7. The black dots represent the observed values of bacterial production rates. dynamics, with measured values of the kinetic parameters involved, suggests that this time lag is not the result of a reduction of bacterial activity due to low temperatures. The fact that DOC produced by phytoplankton is mostly under the form of macromolecular material requiring extra- cellular hydrolysis before being taken up by bac- terioplankton is a more likely explanation. 252 G. Billen & S. Becqueoort I 1 I November December January The occurrence of a delay between phyto- plankton and bacterioplankton development does not, therefore, appear to be a specific feature of polar systems. Accordingly. a similar, though shorter delay in bacterioplankton response to the spring phytoplankton bloom has been also described in temperate marine ecosystems (Billen 1990; Billen et al. 1990). Acknowledgements. -This work is part of the Belgian Scientific Research Programme on Antarctica. funded by the Science Policy Office ( B N S S ~ S . Belgium). under contract nr ANTAR/ 05. The authors are indebted to the Australian Antarctic Div- ision for invitation to take part in Marine Science Voyage 7 to Prydz Bay (1987). Data from the Weddell Sea were collected during the European Polarsfern Study (EPOS) sponsored by the European Science Foundation and the Alfred Wcgener Institute for Polar and Marine Research. References Azam. F.. Ammerman. J W . & Cooper. N. 1981: Bacterio- plankton distrihutional patterns and metabolic activities in the Scotia Sea. A n f . 1. U.S. 16. 16&165. Billen. G . 1984: Heterotrophic utilization and regeneration of nitrogen. Pp. 31?-355 in Hobble. J . E. & Williams. P. J . le B. (eds.): Heferofrophic Acfiuify I n rhe Sea. Plenum Press. Billen, G . 1990: Delayed debelopment of bacterioplankton with respect to phytoplankton: a clue for understanding their trophic relationships. Arch. Hydrobiol. Beih. Ergebn. Limnol. 1 1 4 . 415-429. FIR. 9. Simulation of the effect of temperature b) and nature of thc organic matter released by phytoplankton c) on the dynamics of bacterioplankton development in thc Weddell Sea. a) basic simulation. as in Fig. 7. k,,, = 0.004 h - ' ; k,, = 0.0005 h - ' observed in situ temperature. b) simulation of the effect of increasing the temperature up to 5°C. c) simulation of increased proportion of algal exudation in D O C release: k,,, = O.OLM5 h - ' ; k,, = 0.oM h- observed i n situ temperature. Billen. G . 1991: Protein degradation in aquatic environments. Chap. 7. pp. 122-142 in Chrost. R . (ed.): Enzymes in Aquafic Enoironmenrs. Springer Verlag. Billen. G . & Fontigny. A. 1987: Dynamics of Phaeocysris- dominated spring bloom In Belgian coastal waters. 11. Bac- terioplankton dynamics. Mar. Ecol. Prog. Ser. 37, 24'+257. Billen, G. & Servais. P. 1989: Modelisation des processus de degradation de la matierc organique en milieu aquatique. Pp. 21%245 in Bianchi. M. et al. (eds.): Micro-organtimes dans les ecosysremes oceaniques. Masson. Paris. Billen. G . . Joiris. CI.. Meyer-Reil. L. & Lindebloom. M . 1990: RBle of bacteria in the North Sea ecosystem. Nerh. J. Sea Res. 26. 265-293. Bird. D. E. & Kalff, J . 1984: Empirical relationship between bacterial abundance and chlorophyll concentration in fresh and marine waters. Can. J . Fish Aq. Sci. 41, 15-1023. Bjarnsen. P. K. 1989: Phytoplankton exudation of organic matter: Why d o healthy cells d o it? Limnol. Oceanogr 33. 151-154. Bolter. M. & Dawson, R. 1982: Heterotrophic utilization of biochemical compounds in Antarctic waters. Nefh. J . Sea R e s . 16. 315-332. Cole. J . C.. Findlay. S. & Pace. M. L. 1988: Bacterial pro- duction i n fresh and saltwater ecosystems: a cross-system overview. Mar. E d . Prog. Ser. 43. 1-10. Davidson. A . T . 1985: Aspects of the biology of Phaeorysris pouchefii. Thesis, University of Tasmania. Hobart, Australia. 231 pp. Edler. L. 1979: Recommendations for marine biological studies in the Baltic Sea. Phytoplankton and Chlorophyll. Bdrzc Marine Biologisfs 5 . 1-38. Fontigny. A , , Billen. G . & Vives Rego. J. 1987: Some kinetics characteristics of exoprotcolytic activity in coastal sea water. Coasf Esf. Mar. Sci. 2 5 . 127-134. Phytoplankton-bacteria relationship 253 Fuhrman. J . A . & Azam. F. 1982: Thymidine incorporation area measure of heterotrophic bacterioplankton evaluation in marine surface waters: cvaluation and field result. Mar. Biol. 66, 109-120. Hanson. R . B . & Lowry, H . K. 1983: Nucleic acid synthesis in oceanic microplankton for the Drake Passage. Antarctica. Evaluation of steady-state growth. Mar. Biol. 73. 79-89. Hanson. R . 8.. Shafer. D., Ryan, T . , Pope, D. H. & Lowry. H. K. 1983: Bacterioplankton in Antarctic Ocean waters during late austral winter: abundance, frequency of dividing cells and estimates of production. Appl. Enuiron. Microbiol Hobbie, J . E . , Daley, R. J . &Jasper, S. 1977: Use ofNuclepore filters for counting bacteria by fluorescence microscopy. Appl. Enuiron. Microbiol. 33, 1225-1228. Hodson, R . E., Azam, F., Carlucci, A. F., Fuhrman. J. A , . Karl, D . M. & Holm-Hansen. 0. 1981: Microbial uptake of disolved organic matter in McMurdo Sound, Antarctica. Mar. Biol. 61, 89-94. Jassby. A. D. & Goldman. C. R. 1974: Loss rates from a lake phytoplankton community. Limnol. Oceanogr. 19. 618-627. Joiris, C . , Overloop. W., Frankignoulle. M. & Bouquegneau, J . M. 1987: Preliminary discussion of the results obtained in Antarctica during the austral summe; 1986-1987: plankton ecology and ecotoxicology. In Proceedings of fhe Belgian Colloquium on Antarcfic Research, Brussels, Ocf. 20. 1987. Prime Minister's Services. Science Policy Office. Jumars, P. A.. Penry, D. L.. Baross, J. A.. Perry. M. J. & Frost, B. W. 1989: Closing the microbial loop: dissolved carbon pathway to heterotrophic bacteria from incomplete ingestion. digestion and absorption in animals. Deep-sea Res. 36, 48S495. Kriss. A . E . , Mishustina. 1. E.. Lebedeva. M. N. 1969: Bacterial population densities (heterotrophus) in the water column of the Southern and Indian Oceans. Mikrobiologyia 38. 51 1- 517. Lampert, W. 1978: Release of dissolved organic carbon by grazing zooplankton. Limnol. Oceanogr. 3 , 831-834. Lancelot. C., Billen. G. & Mathot. S. 1989: Ecophysiology of phyto- and bacterioplankton growth in the Southern Ocean, pp. 1-93 in Caschetto. S . (ed.): Belgian Scientific Research Programme on Antarcfica. Scientific Results of Phase 1 (Ocf 85-Jan 89). Vol. 1 . Science Policy Office, Brussels. Belgium. Larsson, U . & Hagstrom. A . 1979: Phytoplankton exudate release as an energy source for the growth of pelagic bacteria. Mar. B i d . 52. 199-206. Linley, E. A. S.. Newell, R. C. & Lucas. M. I . 1983: Quan- titative relationship between phytoplankton, bacteria and heterotrophic microflagellates in shelf waters. Mar. Ecol. Prog. Ser. 12. 77-89. Lorenzen, C. J . 1967: Determination of chlorophyll and phaeoigments: spectro-photometric equations. Limnof. Oce- anogr. 12. 34S346. M ~ l l e r Jensen, L. 1983: Phytoplankton release of extracellular organic carbon molecular weight composition and bacterial assimilation. Mar. Ecol. Prog. Ser. 11. 39-48. Mullins, B. W . a Priddle, J . 1987: Relationships between 45, 1622-1632. bacteria and phytoplankton in the Bransfield Strait and Sou- thern Drake Passage. Br. A n f . Suru. Bull. 76. 51-64. Parsons. T . R . & Strickland, J . D. H. 1962: On the production of particulate organic carbon by heterotrophic proccsscs in seawater. Deep-sea Res. 8, 211-222. Pomcroy. L. R . & Deibel, D . 1986: Temperature regulation of bacterial activity during the spring bloom in Newfoundland coastal waters. Science 233. 359-361. Porter. K. G. & Feigh, Y. S. 1980: Use of DAPI for identifying and counting aquatic microflora. Limnol. Oceanogr. 25,94> 948. Pomeroy, L. R . , Macko, S. A,, Ostrom. P. H . & Dunphy, 1. 1990: The microbial food web in Arctic seawater: con- centration of dissolved free amino acids and bacterial abun- dance and activity in the Arctic Ocean and in Resolute Passage. Mar. Ecol. Prog. Ser. 61. 31-40. Riemann. B., Bjernsen. P. K., Newell. S. Y. & Fallon, R. D. 1987: Calculation of bacterioplankton production from measurements of 'H-Thymidine incorporation. Limnol. Oce- anogr. 3 2 . 471-476. Servais. P. 1986: Etude de la dbgradation de la matiere organ- ique par les bactiries hbtbrotrophes en riviere. Divel- oppement d'une demarche mithodologique et application a la Meuse belge. Universitk Libre de Bruxelles. Faculti des Sciences. Thtse. 271 pp. Servais, P.. Billen, G . & Vives-Rego. J . 1985: Rate of bacterial mortality in aquatic environments. Appl. Enuiron. Microbiol. 49. 1448-1455. Servais. P., Billen, G. & Hascoet, M. C . 1987: Determination of the biodegradable fraction of dissolved organic matters in waters. Wafer Res. 21. 445-450. Servais, P.. Billen. G . . Martinez, J . & Vives-Rego. J . 1989: Estimating bacterial mortality by the dissapearance of 3H- labelled intracellular DNA. Technical validation and field measurements. FEMS Microbiol. Ecology 62, 119-126. Simon, M. & Azam. F. 1989: Protein content and protein synthesis rates of planktonic marine bacteria. Mar. Ecol. Prog. Ser. 51. 201-213. Somville. M. 1984: Measurement and study of substrate speci- ficity of exoglucosidase activity in eutrophic water. Appl. Enuiron. Microbiol. 48, 1181-1185. Somville. M. & Billen, G. 1983: A method for determining exoproteolytic activity in natural waters. Limnol. Oceanogr. 28. 190-193. Sorokin, Y. 1. 1971: On the r61e of bacteria in the productivity of tropical oceanic waters. Int. Reo. Ges. Hydrobiol. 56, 1- 48. Sullivan. C. W.. Cota. G . F.. Krempin. D. W. & Smith. W. 0.. Jr. 1990: Distribution and activity of bacterioplankton in the marginal ice zone of the Weddell-Scotia Sea during austral spring. Mar. Ecol. Prog. Ser. 63, 239-252. Wolter, K. 1982: Bacterial incorporation of organic substances released by natural phytoplankton populations. Mar. Ecol. Prog. Ser. 7 , 287-295. Wright, R . T. & Hobbie, J. E. 1965: The uptake of organic solutes in lake water. Limnol. Oceanogr. 10, 22-28.