Microbial communities from the sea ice and adjacent water column at the time of ice melting in the northwestern part of the Weddell Sea SYLVIE MATHOT, SYLVIE BECQUEVORT and CHRISTIANE LANCELOT Mathot, S.. Becquevort. S . & Lancelot. C. 1991: Microbial communities from the sea ice and adjaccnt water column at the time of ice melting in the northwestern part of the Weddcll Sea. Pp. 267-275 in Sakshaug. E., Hopkins. C. C . E. & 0ritsland. N . A. (eds.): Proceedings of the Pro Mare Symposium on Polar Marine Ecology, Trondheim, 12-16 May 1990. Polar Research l O ( 1 ) . Microbial composition - including microalgae, bacteria and protozoans - and potential metabolic activity of its autotrophic compartment were measured in December 1988 in several micro-environments that characterise the North-West Sector of the marginal area of the Weddell Sea; infiltration and band assemblages of ice floes and adjacent waters were investigated. At the time of ice melting, a shift from a diatom dominated population (ice) to a flagellate dominated population (water column) was observed. Nevertheless. this shift was not due to an “inability” of the icc-diatoms to grow in the water colum. Macro- grazing and/or sedimentation arc suggcstcd as possible causes of the disappearance of diatoms during ice melting. The rcmaining small autotrophic forms released by the ice would constitute a significant seeding stock for the growth of ice-edge blooms. Syluie Mathor, Syluie Becqueuortand Christiane Lancelot, Groupe de Microbiologie des Milieux Aquariques, University of Brussels. CP 221, Campus de la Plaine. B-1050 Brussels, Belgium. Introduction The annual sea ice of the Southern Ocean pro- vides suitable microhabitats for microalgae (Whi- taker 1977; Ackley et al. 1979; Palmisano & Sullivan 1983; Horner 1985; Garrison et al. 1986; Garrison & Buck 1989b), heterotrophic flag- ellates and other protozoans (Garrison & Buck 1989a; 1989b; Buck et al. 1990; Garrison & Gowing 1992), as well as bacteria (Sullivan 1985; Kottmeier & Sullivan 1987; Kottmeier et al. 1987). Indeed the sea ice offers a set of physico- chemical conditions for microorganisms living in close association with it, either attached to ice crystals or suspended in the interstitial water between ice crystals (Horner 1985). When released from the ice upon melting, the fate of ice-associated algae may be variable: part of the algae settles down (Schnack et al. 1985; von Bodungen et al. 1986), part is grazed by pelagic herbivores such as krill (Marschall 1988) or copepods (Fransz 1988), and part survives in the water column (Garrison & Buck 1985). The latter part should constitute an inoculum for ice- edge blooms that grow in the shallow, stable water column induced by the melting of pack ice (Garrison et al. 1986, 1987; Fryxell & Kendrick 1988; Smith & Nelson 1985, 1986; Sullivan et al. 1988). Accordingly, many of the nano- and microheterotrophic species are common to both ice and water (Garrison & Gowing 1992), and those which seem to grow in ice after their initial incorporation (Garrison & Buck 1989b) may pro- duce an inoculum in the water column at the time of ice melting (Garrison & Gowing 1992). This study presents data concerning the struc- ture of microbial communities originating from sea ice biota at the time of ice melting during the period of retreat of the ice edge. We have compared microbial inhabitants (microalgae, bac- teria and protozoans) of several ice environments and adjacent water columns in the Weddell Sector of the Antarctic Ocean. The possible genesis of an ice edge bloom through the release of living cells seeded into the water column depends not only on the physical properties of the adjacent water column, but also on the ability of the released ice-algae to be metabolically active. Thus, the potential activity of the primary producers released by the melting process also has been tested in a simulated seeding experiment under controlled conditions in filtered seawater. 268 S . Mathot, S . Becqueuort & C. Lancelot 5 0' 47 4 4* 56' '\ - 60 ey 1' e - 6 2 ' -58' t b on c u - 1 5 0 A S S E Y B L A G E C Material and methods Sampling Samples were collected during the European expedition EPOS leg 2 on board R/V POLARSTERN from 22 November, 1988, to 9 January, 1989. Three sites were investigated in December 1988 in the marginal ice zone of the northwestern Wed- dell Sea in an area extending from 47" to 49" W, between 61" and 62"s (Fig. 1A and B). Referring to the terminology proposed by Hor- ner et al. (1988), results presented here only concern the infiltration assemblage located at the snow-ice interface of floes (Stations 169 and 194), and the band assemblage which appeared as a brown coloured band in the middle (60 t o 80 cm from the top of the ice floe) of cores taken at Station 178 (Fig. 1C). Bottom assemblages were inconspicuous in the study area. To reduce osmotic shock (Garrison & Buck 1986), ice samples were melted in sterile, filtered seawater in the dark prior to fixative addition or experiment start. Ice assemblages were sampled randomly. For comparison between ice and water environ- ments, samples of seawater were also collected directly under the ice floes. Microscopical analysis Qualitative and quantitative analysis of plankton was carried out using two different microscopical methods: 1) Microplankton samples (mainly dia- toms, ciliates, dinoflagellates), preserved with modified Lugol's solution (Thomas pers. comm.), were analysed in settling chambers using the inverted microscope technique of Utermohl (1958); 2) Nanoplankton (mainly autotrophic and heterotrophic flagellates) and bacterioplankton samples fixed respectively with glutardialdehyde (0.5% final conc.) and formalin (2% final conc.) were analysed by epifluorescence microscopy after 4'6 diamidino-2-phenylindole (DAPI) stain- ing according to the procedure of Porter & Feig (1980). In both methods, cell volumes were calculated from measurements of the dimensions and shapes F i g . I . Sampling area and sites: A . Location of the sampling area in the Southern Ocean: B . Cruise track of Nov. 1988- Jan. 1989 EPOS Leg 2 expedition in the Scotia/Weddell Sea area. showing the ice stations sampled; C. Schematic repre- sentation of the sea ice biota sampled. with a vertical scale (in cm) of the thickness of ice floes. Microbial communities in the northwestern part of the Weddell Sea 269 0 0 4 of cells. In the case of diatoms, cell volumes were converted to carbon biomass according to Edler’s (1979) recommendations using a conversion fac- tor of 0.11 pgC pW3. For ciliates, cell volumes were converted to carbon values using the con- version factor of 0.08 pgC pm-’ (Sherr & Sherr 1984). The conversion factor of 0.11 pgC pm-3 (Edler 1979) was used for autotrophic and het- erotrophic flagellates. Concerning the bacteria, biovolumes were estimated on the enlargements of microphotographs. Conversion into carbon biomass was done using the biovolume dependent C/biovolume ratio proposed by Simon & Azam ( 1989). Activity measurements The experimental determination of photosyn- thetic parameters involved short-term I4C incu- bation (Steemann-Nielsen standard method) performed at different light intensities (P-I curves). Bottles (Cel-Cult) tissue culture flasks of 60, 250 and 700ml were incubated in a water bath with running seawater at in situ temperature illuminated by artificial light. Maximum irradiance reached 135 pmol m-*s-I, i.e. very close to the light saturation constant characteristic of antarctic phytoplankton. 14C incubations were conducted at in situ temperature for different fractions of light intensity (0, 1, 4, 6, 15, 20, 40, 60, 100%). Incubation times of 4-6 hours were chosen after a preliminary study of P-I curves for different incubation times. This choice minimises losses by respiration and increases accuracy. After incubation, samples were filtered on GF/F filters. Radioactivity was measured on the filter (photosynthetic carbon fixation) and in the dis- solved organic matter (excretion). However, radioactivity of the latter was never significantly different from that of the background. Excretion was therefore assumed to represent a maximum of 5% of total photosynthesis. Photosynthetic parameters K,,,, a, and p were then statistically estimated by means of the Platt et al. equation (1980). 2A) and band (Fig. 3A) algal assemblages as well as their respective water column phytoplanktonic assemblages (Figs. 2B and 3B). Carbon uptake has been normalised t o “active” carbon biomass calculated from cell counts and biovolume measurements. Thus, only carbon associated with vegetative, healthy autotrophic cells has been taken into account; this was a necessary correction since resting spores accounted for nearly half (45%) of the total autotrophic carbon in the band assemblage. Photosynthetic characteristics of the populations as computed by statistical fitting of experimental data using the Platt et al. equation (1980) are presented in Table 1. From exam- ination of the P-I curves, it is obvious that, apart from the band assemblages (Fig. 3A), which revealed a low maximum specific rate of photo- synthesis (see Table 1, mean K, = 0.012 h-l), the infiltration assemblages exhibited similar values (Fig. 2A, Table 1, mean K, = 0.049 h-I) to those 0 0 5 0 0 4 0 0 3 0 0 2 0 01 0 Infiltration Assemblage K , t i - ’ ,r i 0 200 400 600 moo 1000 A 1, pno1.m-0,s-1 0 0 3 0.02 0 01 u Results Primary producers and potential activities Figs. 2 and 3 show algal uptake of carbon as a function of irradiance for typical infiltration (Fig. 0 200 400 800 BOO 1000 B I , pno1.m-0,s-1 Fig. 2. Photosynthesioirradiance relationship of natural popu- lations sampled (A) in the infiltration assemblage and (B) in the adjacent water column. 270 S. Mathot, S. Becquevort & C . Lancelot Hand Assemblage ~ 1 ~~- I < * ''.I (veprl.l,ra cells) - - O o 5 I 0 0 4 0 03 0 0 2 0 200 400 6W BOO woo A I , p ~ i i d . m - ~ . s - ' Water Column K. h-I 0 0 5 0 04 n 03 00' 0 0 1 0 0 203 4 0 0 OW boo ::?I> B I , pno1.m-2.6" Fig. 3. Photosynthesis-irradlance relalionship of natural popu- lations sampled (A) in the band assemblage and (B) in the adjacent water column. For the band assemblage, only veg- etative cells have been considered. of the water column phytoplankton (Fig. 2B and 3B, Table 1, mean K, = 0.041 h-l). Photo- synthetic efficiencies were in the same order of magnitude for infiltration and water column assemblages, ranging between 0.00066 h-' (prnolm-*s-')-' for the former and 0.00047 h-l(pmol m-' s-')-' for the latter. Ik (index of photoadaptation) values were higher for infiltration and water column assemblages than for the band assemblages, being 75 and 92 p o l m-* s-' for the former, and much lower for the latter (Table 1). Thus photosynthesis- irradiance relationships did not exhibit clear vari- ations between infiltration assemblage and the surrounding water column, and both communities were similarly well adapted to prevailing physico- chemical conditions. "Taxonomic" composition The dominant autotrophic taxons present in the different environments, i.e. ice assemblages and the adjacent water column, are summarised in Table 2A. Results are expressed as percentages of total autotrophic cell number. This table reveals a predominantly algal community with pennate diatoms always dominant in the ice assemblages (mean = 74% for infiltration assemblage; mean = 82% for band assemblage). The diatoms were mainly of the genus Nitzschia (up to 94% of autotroph cell number, e.g. N . closterium, N . cyl- indrus, N. curta, N . kerguelensis), but also genera such as Tropidoneis and Amphiprora, which pre- sented a certain variability within each type of ice assemblage as well as in between the two types studied, were present (Table 2A). Centric diatoms were scarce - if at all present - in the infiltration assemblages, whereas they accounted for 1.4 and 32% of autotroph cell num- bers in the band assemblages, both as vegetative cells and resting spores. Autotrophic flagellates were not abundant in the ice assemblages, and even if they did reach 51% of the autotrophic cell number (Table 2A) in a particular infiltration sample, their con- Tuble 1. Photosynthetic characteristics of the infiltration assemblage and the band assemblage (= vegetative cells only, with exclusion of the resting spores) from ice communities and of phytoplankton from the water column. n a KIT Ik Infiltr.Ass. Band Ass. Phytoplankton 2 2 h O.ooo66 2 O.ooOo7 O.OOO18 f O.ooOo7 0.00042 2 0.00016 0.049 It 0.003 0.012 -C 0.007 0.041 It 0.014 75 It 12.8 62 t 15.7 92 -t 19 n ' number of samples a . photosynthetic efficiency [ h - ' ( p m o l m - * s - ' ) . ' ] . K, : maximal specific rate of photosynthesis ( h - ' ) . Ik : light adaptation parameter (K,,,,' a: pmol m e ' s - ' ) Microbial communities in the northwestern part of the Weddell Sea 271 Table 2. A. Composition of the autotrophic community encountered in the different environments sampled. Results expressed as percentage of total autotrophic cell number. (-) = negligible. Infilt .Assemblage Band Assemblage n = 4 n = 2 Taxon Range, X Range, X Water Column n = 3 Range, X Pennate diatoms Nitzschia sp. 4 1-94 70.7 2 15 57-82 70 ? 12.5 %I3 10.3% 1.8 Tropidoneir sp. 0-7 2.5 2 2.5 1 . 6 2 1.8 f 0.2 (-) Amphiprora sp. 0-1 0.3 2 0.4 6.5-15 10.8 2 4.3 (-) Centric diatoms (-1 1.4-32 16.7 2 15.3 (-) Flagellates Dinoflagellates 0-2 1.0 2 1.0 s2.5 1.3 2 1.3 2-3 2.7 2 0.4 Nanoflagellates &5 1 25 2 13 (-) 85-89 87.3 4 1.6 tribution to the autotrophic total biomass (Table 3A) remained negligible compared to that of dia- toms. Thus, the autotrophic composition of the ice assemblages was highly variable and showed a patchy distribution of microalgae, reflecting the heterogeneity of the ice environment. In contrast, the composition of autotrophic communities of the adjacent water column was constant and homogeneous, without any con- spicuous differences between the different locali- ties. Indeed, Nitzschia sp. was the only diatom present, accounting for only 9 to 13% (mean = 10%) of the autotroph cell numbers (Table 2A), whereas the bulk of the autotrophic population consisted of flagellated cells (8549% autotrophic nanoflagellates including Cryptomonas spp., Pyr- amimonas spp., Phaeocystis pouchetii; I-3% autotrophic dinoflagellates). In terms of biomass (Table 3A), diatoms represented about one-fifth of the total autotrophic biomass (range = 19- 25%, mean = 21%), and autotrophic flagellates occupied the remaining four-fifths, with a net dominance of nanoflagellates (range = 64-72%, mean = 69%) other than dinoflagellates (range = 3-18%, mean = 10%). Concerning the protozooplankton abundance in the three environments considered (Table 2B), there were no differences between infiltration assemblages and water column assemblages in the composition of the groups, with net dominance of heterotrophic nanoflagellates (9599% of heterotrophic cell number) over heterotrophic dinoflagellates (1-5%). The biomass of het- erotrophic dinoflagellates (Table 3B) was higher (mean = 31% of total heterotrophic carbon) than that of the heterotrophic nanoflagellates (mean = 20% of total heterotrophic carbon) in the infil- tration assemblages, whereas in the water column, the non-dinoflagellate heterotrophic nanoflagellates dominated the flagellated fraction (see Table 3B). Ciliates were not significant although this group did contribute to some extent Table 2. 8 . Composition of the protozoan community encountered in the different environments sampled. Results expressed as percentage of total protozoan cell number. 1nfilt.Assemblage n = 4 Taxon Range, i Band Assemblage n = 2 Range, x Water Column n = 3 Range, X Ciliates (-1 ~~ 30-80 552 25 (-) Flagellates Dinoflagellates 1-5 3.5 2 1.3 20-70 45 ? 25 1-4 3 4 1.3 Nanoflagellates 95-99 96.5 2 1.3 (-) 9 6 9 9 97 5 1.3 272 S. Mathot. S . Becquevort & C. Lancelot Table 3. A . Composition of the autotrophic biomass encountered in the various environments sampled. Results expressed as percentage of total autotrophic carhon biomass. ( - ) = negligihle. In611 Assemblage n = l Taxon Rangc. x Band Assemblage n = 2 Range. A Water Column n = 3 Range. X Diatoms 7 1 9 5 84 ? 5.6 9 1 9 9 98.5 2 0.5 1 s 2 5 21.3 2 2.4 Flagellates 10.7 2 5.1 Dinoflagellates 1-14 Nanoflagellates s 1 4 7.5 2 4.5 0.5-1 0.8 ? 0 3 W 7 2 68.7 f 3.1 1 . 5 ? 1 . S ( - ) ( - ) 6.5 2 4 w.5 0.3 2 0.5 S1R Phaeocystic col r Cyanobacteria 0 . s 2 . 2 1 ? 0.6 ( - ) ( - ) TubL 3 B Composition of the heterotrophic biomass encountered in the various environments sampled. Results expressed as percentage of total heterotrophic biomass. lnfilt .Assemblage Band Assemblage Water Column n = 4 n = 2 n = 3 Taxon Range. A Range, x Range. x Protozoa Ciliates 5-13 8 f. 3.6 6.2-69.3 38 f 31.6 5-10 7.2 2 2.1 Dinoflagellates 10-55 31 2 19.7 0-1.3 0.1 2 0.07 9-32 19.3 2 8.3 Nanoflagellates MI 20 f 10.6 ( - ) 19-31 25 -c 4 Bacteria 16-59 41 2 12.6 29-94 62 ? 32.2 3 M 7 48.5 f. 12.2 Table 4 . Composition of microorganisms in the sea ice and adjacent water column samples. Results expressed in percentage of total carbon biomass (autotrophs + heterolrophs). Diatoms R.S. = Resting Spores; Diatoms V.C. = Vegetative Cells. ( - ) = negligible. lnfilt . Assem blage n = 4 Taxon Range. x Band Asscmblage n = 2 Range. x Water Column n = 3 Range, A Autotrophs Diatoms R.S. Diatoms V.C. Phaeocystis col Dinoflagellates Nanoflagellates Cyanobacteria Total Heterotrophs Ciliates Dinoflagellates Nanoflagellates Bacteria ( - ) 65-93 0-3.3 1-1 1 3-13 0.1-1.8 82-97 0 1-2.4 0.1-4.5 0.2-3 1 3 - 1 1 17 f. 8.7 1.2 ? 1.2 5.6 2 3.2 6.9 f. 4.3 0.9 2 0.6 91 -t 4.8 0.9 5 0.7 2.1 -c 1.2 1.6 2 0.7 3.8 r 3.5 12-45 50-63 0-0.3 0.2-1 ( - ) ( - ) 76-96 1.4-2.9 0-0. 1 M . 6 I .%22 29 ? 16.2 56.5 f. 6 0.2 f. 0.2 0.6 f 0.4 ( - ) ( - ) 11-15 2-10 38-50 ( - ) 13.4 ? 1.6 6.7 ? 3.2 43.6 f 4 86 f 10 2 2 2 0.8 0 05 f 0 05 0 3 e 0 3 1 2 2 10.4 6@7 1 I .9-3 3 . 1 9 7-12 %27 64 ? 14.7 2.5 f 0.4 6.5 f. 1.8 9 ? 2.1 1 8 . 2 2 5.9 Total 8.3 2 4.9 4-24 14 ? 10 2%40 36.3 f 4.9 Microbial communities in the northwestern part of the Weddell Sea 273 .... ....... ........ .......... ........... ............ ............. ..... ......... ,:.::::::::::: I.i.’iiiii:iiii ........ ......... ............ diatoms >2Op A Infiltration Assemblage B Band Assemblage C Water Column Legend 1 -1 A ~ t o t r o p h s 2-10 Itm Autolrophs 10-20 pm Autolroplis >20 pin Fig. 4. Composition of the total autotrophic biomass in three size classes of distribution, with distinction between diatoms and flagellates in each size range: A. Infiltration assemblage 169/1; B. band assemblage 178/1; C. water column assemblage 194. to the total heterotrophic biomass (Table 3B), with mean values of 8% for the infiltration assem- blages, and 7% for the water column. In the band assemblages, heterotrophic nano- flagellates were present in negligible numbers compared t o heterotrophic dinoflagellates and ciliates, which showed a great heterogeneity in their respective distributions (Table 2B). Discussion Relative proportions - in terms of biomass - of autotrophs (diatoms, dinoflagellates, nanoflag- ellates and cyanobacteria) as well as hetero- trophs (ciliates, dinoflagellates, nanoflagellates, and bacteria) are shown in Table 4 for both ice and water column communities. Floristic analysis of the ice assemblages showed a clear dominance of diatoms over other auto- trophs (65-95% of total biomass). Moreover, comparison of ice assemblages indicated the pres- ence of resting spores in the band assemblages, suggesting that these might be remnants from a sub-ice algal bloom from the previous year, which were “trapped” in two-year-old ice according to Ackley et al. (1979) and McConville & Wetherbee (1983). In contrast, diatoms in the water column con- stituted a minor fraction never exceeding 15% (Table 4); the bulk of the biomass was contributed by the autotrophic flagellates (45 t o 57%) with a net dominance of the nanoflagellated fraction (78-96%). These results seem to contradict observatons made by Garrison & Buck (1985), Garrison et al. (1986) and Smith &Nelson (1986), all of whom found great similarity among assem- blages from ice floes and from planktonic popu- lations, supporting the seeding hypothesis from the ice to the water column. Our results, however, do not exclude the potential role of seeding, but do indicate that other factors (such as early graz- ing by macrozooplankton) can prevent seeding of the water column assemblages. Note that unlike Fryxell & Kendrick (1988) who suggested that Phaeocystis colonies found in the water column in the same area could have been seeded from the melting ice, Phaeocystis colonies were present in two of our infiltration samples where they accounted for 1 and 3% of the total biomass, but no colonies were observed in the surrounding waters. In the ice assemblages, the autotroph/ 214 S . Mathot. S. Becquevort & C . Lancelot heterotroph biomass ratio was quite different from one sample to another, ranging between 3 and 33. In contrast, water samples presented a lower and remarkably constant ratio (1 5 2 . 5 ) . The relative proportions of heterotrophs were always more important in the water column than in the ice environment, although protozoans (including flagellates and ciliates) and bacteria are regular and abundant components of the ice biota (Garrison et al. 1986). The abundance of heterotrophs in the ice environment indicates an active food web within the ice community (Gar- rison & Buck 1989b). In fact, the relative pro- portions of heterotrophs were greater in the water column since only a part of total autotrophic biomass remained in the water column at the time of ice melting. This accounts for the shift observed from a diatom dominated population (ice) t o a flagellate dominated population (water column) at the time of ice melting. However, as has been shown pre- viously (Table 1, Fig. 2), this shift cannot be explained by an “inability” of the ice com- munities - at least for their autotrophic con- stituents- to grow in the water column, but rather by an effective disapppearance from the water column, either by grazing pressure (macro- zooplankton) or sedimentation processes. Indeed, euphausiids have been shown to follow the retreating ice edge, taking advantage of the elevated food supply when the ice is melting (see Sakshaug & Skjoldal 1989). During the EPOS Leg 2 expedition, Cuzin-Roudy & Schalk (1989) reported an abundance of krill under ice floes, indicating that sea ice can provide a nursery ground for larval krill (Marshall 1988) which feed on particles released by the melting infiltration and band assemblages. Smetacek et al. (1990) even presents a hypothetical annual cycle where krill switch from scraping ice algae to filtering phytoplankton. The disappearance of diatoms but not of flagellates does not seem t o reflect selection of “species” but rather of particle size. As seen in Fig. 4, the distribution of autotrophs in the various size ranges is inverted when going from the ice environment to the water column. This figure shows an example of autotrophic biomass distribution in three size ranges (2-10pm. 10- 20 pm, > 20 pm) for an infiltration assemblage (Fig. 4A), a band assemblage (Fig. 4B) and a water column phytoplanktonic community (Fig. 4C), with separation between diatom and flag- ellate biomass contribution within each size range. Although krill is capable of feeding on very small particles, large cells are taken more efficiently (Segawa et al. 1983; Boyd et al., 1984). Meyer & El-Sayed (1983) also showed the pref- erential feeding on micro-sized (20-200 pm) par- ticles by krill. On the other hand, because of the spatial con- straint of living in the ice environment, the growth of ice algae can occur in aggregated entities in between ice crystals (Tropidoneis oanheurckii often aggregated in our samples). These obser- vations are in accordance with experimental results obtained by Riebesell et al. (1991) col- lected during the same EPOS expedition. Being heavier, these aggregates could, together with larger cells, sink out of the surface mixed layer leaving small cells and flagellates in the adjacent water column. In fact, high sedimentation fol- lowing spring blooms has been reported as being one of the fates (Schnack et al. 1985) if not the dominant one (von Bodungen et al. 1986) of ice- edge blooms. In summary, by comparing the floristic corn- position of the ice environment with that of the water column following ice melting, an obvious shift from a diatom dominated population in the ice environment to a flagellated one in the water column was observed, with an apparently neg- ligible seeding effect of ice algae into the water column. When analysing the photosynthetic capa- bilities of these ice algal communities (infiltration assemblages being largely dominant over band assemblages in the area of the Weddell Sea we visited), there is clear evidence that they could grow at the same rate as the water column assem- blages. Thus. other factors such as grazing by pelagic herbivores or sedimentation at very early stages during and after melting of the ice might significantly modify the structure of ice-associated microbial communities entering the water column. Acknowledgernenrs. - We thank R. Sharek and E-M. Nothig for helping in the initial identification of ice algae. V. Smetacek. E . Sakshaug and two anonymous reviewers provided valuable comments to the manuscript. This work is part of the Belgian Scientific Research Programme on Antartica. funded by the Science Policy Office (Brussels, Belgium), under contract Nr ANTAR/OS. Data presented here were collected during the European Polorsfern Study (EPOS) sponsored by the European Science Foundation and the Alfred Wcgcner Institute for Polar and Marine Research. Bremerhaven. References Ackley. S. F . . Buck. K . R . & Taguchi. S. 1979: Standing crop Microbial communities in the northwestern part of the Weddell Sea 275 of algae in the sea ice of the Weddell Sea region. Deep-sea Res. 26A, 269-282. Boyd, C. M.. Heyraud. M. & Boyd, C . N. 1984: Feeding of the Antarctic krill Euphausia superba. 1. Crustacean Biol. 4 . 12S141. Buck, K. R., Bolt. P. A. & Garrison, D . L. 1990: Phagotrophy and fecal pellet production by an athecate dinoflagellate in Antarctic sea ice. Mar. Ecol. Prog. Ser. 60, 75-84. Cuzin-Roudy, J. & Schalk, P. H. 1989: Macrozooplankton - Biomass, development and activity. Pp. 146-159 in Hempel, I . . Schalk, P. H. & Smetacek, V. (eds.): The Expedition Antarktis V11/3 (EPOS LEG 2) of R/V “Polarstern” in 1988/ 89. Reporfs on Polar Research. Edler, L. 1979: Recommendations for marine biological studies in the Baltic Sea - Phytoplankton and Chlorophyll. Baltic Marine Biologisrc 5 , 1-38. Fransz, H . G . 1988: Vernal abundance, structure and develop- ment of epipelagic copepod populations of the Eastern Wed- dell Sea (Antarctica). Polar Biol. 9, 107-114. Fryxell, G . A . & Kendrick, G . A . 1988: Austral spring micro: algae across the Weddell Sea ice edge: spatial relationships found along a northward transect during AMERIEZ 83. Deep-sea Res. 35, 1-20. Garrison, D . L. & Buck, K. R. 1985: Sea-ice algal communities in the Weddell Sea: species composition in ice and plankton assemblages. Pp. 103-122 in Gray, J . S. & Christiansen, M. E. (eds.): Marine Biology of Polar Regions and Effects of Stress on Marine Organisms. John Wiley and Sons. Garrison, D. L. & Buck, K. R. 1986: Organism losses during ice melting: a serious bias in sea ice community studies. Polar Biol. 6 , 237-239. Garrison, D. L. & Buck, K. R. 1989(a): Protozooplankton in the Weddell sea, Antartica: Abundance and distribution in the ice-edge zone. Polar Biol. 9, 341-351. Garrison, D. L. & Buck, K. R. 1989(b): The biota of antarctic pack ice in the Weddell Sea and Antarctic Peninsula regions. Polar Biol. 10, 211-219. Garrison, D. L. & Gowing, M. M. 1992: Protozooplankton. In Friedmann. E. 1. (ed.): Antarcric Microbiology. In prcss. Garrison, D. L.. Buck. K. R. & Fryxell. G . A . 1987: Algal assemblages in Antarctic pack ice and in ice-edge plankton. 1. Phycol. 23,564-572. Garrison, D. L., Sullivan, C. W. & Ackley, S. F. 1986: Sea ice communities in Antarctica. BioScience 36, 243-249. Horner. R . A. 1985: Ecology of sea ice microalgae. pp. 83-103 in Horner, R . A. (ed.): Sea Ice Biota. CRC Press, Boca Raton. Horner. R. A , , Syvertsen, E . E . , Thomas, D. P. & Lange. C. 1988: Proposed terminology and reporting units for Sea ice Algal Assemblages. Polar Biol. 8, 249-253. Kottmeier, S. T., McGrath Grossi. S. & Sullivan, C. W. 1987: Sea ice microbial communities: 8. Bacterial production in annual sea ice of McMurdo Sound, Antarctica. Mar. Ecol. Prog. Ser. 35, 175-186. Kottmeier. S . T. & Sullivan, C. W. 1987: Late winter primary production and bacterial production in sea ice and seawater west of the Antarctica Peninsula. Mar. Ecol. Prog. Ser. 36, Marschall. H. P. 1988: The overwintering strategy of Antarctic krill under the pack-ice of the Weddell Sea. Polar Biol. 9, 129-135. Meyer. M. A . & El-Sayed. S . Z. 1983: Grazing of Euphausia superba Dana on natural phytoplankton populations. Polar Biol. I , 193-203. 287-298. McConville, M. J . & Wetherbee, R. 1983: The bottom-ice microalgal community from annual ice in the inshore waters of east Antarctica. 1. Phycol. 19, 431. Palmisano, A . C . & Sullivan C . W. 1983: Sea Ice Microbial Communities (SIMCO) 1 . Distribution, Abundance, and Pri- mary Production of Ice Microalgae in McMurdo Sound. Ant- arctica in 1980. Polar Biol. 2, 171-177. Platt, T.. Gallegos, L. L. & Harrison. W. G . 1980: Pholo- inhibition of photosynthesis in natural assemblages or marine phytoplankton. 1. Mar. Res. 38, 687-701. Porter, K. G. & Feig, Y. S. 1980: the use of DAPl for identifying and counting aquatic microflora. Limnol. Oceanogr. Z S ( 5 ) . 943-948. Riebesell. U . , Schloss, I . , & Smetacek, V. 1991: Aggregation of algae released from melting sea ice: Implications for seed- ing and sedimentation. Polar Biol. I!, 239-248. Sakshaug, E . & Skjoldal. H. R. 1989: Life at the ice edge. Ambio 18(1), 60-67. Schnack, S. B.,’ Smetacek, V. S., von Bodungen. B. & Stegmann, P. 1985: Utilization of phytoplankton by copepods in Antarctic Waters during spring. Pp. 6 S 8 1 in Gray, J . S. & Christiansen, M. E . (eds.): Marine Biology of Polar Regions and Effecfs of Stress on Marine Organisms. John Wiley and Sons. Segawa, S., Kato, M. & Murano, M. 1983: Growth, moult and filtering rate of krill in laboratory conditions. Mem. Nod. Inst. Polar Res., Special Issue 27, 9S103. Sherr, B. F. & Sherr, E. B. 1984: RBle of heterotrophic protozoa in carbon and energy flow in aquatic ecosystems. Pp. 412-423 in Klug. M. J. & Reddy, C . A . (eds.): Currenf Perspectives in Microbial Ecology. American Society for Microbiology. Washington, D.C. Simon. M. & AYdm, F. 1989: Protein content and protein synthesis rates of planktonic marine bacteria. Mar. Ecol. Prog. Ser. 51. 201-213. Smetacek, V., Sharek, R., & Nothig E-M. 1990: Seasonal and regional variation in the pelagial and its relationship to the life history cycle of krill. Pp. 103-114 in Kerry, K. R . and Hempel. G . (eds.): Antarctic Ecosystems. Ecological Change and Conseruation. Springer-Verlag, Berlin. Smith, W. 0.. Jr. &Nelson, D. M. 1985: Phytoplankton bloom produced by a receding ice edge in the Ross Sea: spatial coherence with the density field. Science 227, 1 6 f 1 6 6 . Smith, W. O., Jr. & Nelson, D . M. 1986: Importance of ice edge phytoplankton production in the Southern Ocean. BioSctence Sullivan, C. W. 1985: Sea ice bacteria: reciprocal interactions of the organisms and their environment. Pp. 159-171 in Horner, R. A . (ed.): Sea Ice Biota., C R C Press, Boca Raton. Sullivan, C. W., McClain. C. R., Comiso, J. C. & Smith, W. O., Jr. 1988: Phytoplankton standing crops within an Antarctic ice edge assessed by satellite remote sensing. J. Geophys. Res. 93(CIO), 12487-12498. Utermohl. H . 1958: Zur Vervollkommnung der quantitativen Phytoplankton - Methodik. Mitt. Int. Verein. Theor. Angew. Limnol. 9 . 1-38. von Bodungen, B.. Smetacek, V. S.. Tilzer, M. M. & Zeitz- schel, B. 1986: Primary production and sedimentation during spring in the Antarctic Peninsula region. Deep-sea Res. 33, 177-194. Whitaker, T . M. 1977: Sea ice habitats of Signy Island (South Orkneys) and their primarry productivity. Pp. 75-83 in Llano, G. A . (ed.): Adaptations within Antarcric ecosystems. Gulf Publishing. Houston. 36, 251-257.