a test of A > B > C can be made, but it is more informative than the .... New York: John Wiley .... Thistle, D., J. A. Reidenauer, R. H. Findlay and R. Waldo: An.
Marine Biology
Marine Biology 88, 143-148 (1985)
. nll(.mOun.
.ndm..UIW.l.n
Microbial food partitioning by three species of benthic copepods K. R. Carman and D. Thistle Department of Oceanography, Florida State University; Tallahassee, Florida 32306-3048, USA
Abstract Using radioactively labeled bacteria and photoautotrophs in undisturbed sediment cores, we show that three cooccurring species of benthic copepods feed on different microbial food sources in their natural environment. Specifically, Thompsonula hyaenae feeds on photoautotrophs, Halicyclops coulli feeds on bacteria, and Zausodes arenicolus feeds on both photoautotrophs and bacteria. Species of benthic copepods feed differently from one another in the field, and meiofaunal species' distributions could be influenced by distributions of their preferred microbial food.
Introduction Although meiofauna are recognized as major participants in many important processes in marine benthic communities (e.g. Gerlach, 1978) and have been employed in the investigation of general ecological issues (e.g. Bell, 1980; Thistle et al., 1984), much remains unknown about their ecology. In particular, meiofaunal taxa in soft-bottom habitats are usually non-randomly distributed (Gray and Rieger, 1971; Heip, 1976; Thistle, 1978; Findlay, 1981), but our understanding of the processes underlying the nonrandomness is not complete. Meiofaunal distribution~are influenced by gross environmental changes, e.g. in sediment type (e.g. Wieser, 1960, Coull, 1970; McLachlan et al., 1977; Tietjen, 1977), in salinity (Jansson, 1966, 1967), and in oxygen availability (Jansson, 1966, 1967). However, responses to such environmental gradients do not explain small-scale patchiness (Gray, 1966, 1968; Ravenel and Thistle, 1981). From the results of a series of laboratory experiments, Gray (1966, 1967, 1968) concluded that the localization of meiofauna may be controlled, in part, by the localized distribution of particular bacterial species upon which the meiofaunal species feed. A variety of
laboratory feeding experiments (Gray and Johnson, 1970; Brown and Sibert, 1977; Vanden Berghe and Bergmans, 1981; Rieper, 1982; Ustach, 1982) have shown that meiofaunal species have marked preferences among the microbial foods offered. However, the work is difficult to interpret because the cultures used, even when obtained from the relevant field locality, bear no necessary relationship to the naturally occurring microbiota because of the selectivity of the culturing process (Jannasch and Jones, 1959; Alexander, 1971; White et al., 1979). The hypothesis that spatial distributions of meiofaunal species are strongly influenced by the distribution of their preferred microbial food requires testing in an intact microbial community in the field (Hicks and Coull, 1983). In this study, we show that species of benthic copepods feed differently from one another in the field, suggesting that meiofaunal species distributions could be influenced by distributions of their preferred microbial food.
Materials and methods Locality Samples were taken at an intertidal sandy-bottom site in St. George Sound, Florida (Lat. 2g054"40'N, Long. 84031"301W). The bottom slope over the site was approximately zero, and the sediment was composed of wellmixed fine sand (graphic-median diameter = 242 ,urn). Samples were collected from an area approximately 40 m offshore where a 10- to 15-cm deep tide pool occurred at low tide. Species At our site, three copepod species were numerically abundant and easily identifiable under the dissecting microscope. Zausodes arenicolus (a harpacticoid) is a
,
K.R Carman and D. Thistle: Copepod feeding relatively large, dorsoventrally-flattened species that burrows between sand grains. Thompsonula hyaenae (a harpacticoid) and Halicyclops coulli (a cyclopoid) are both large, epibenthic species.
Feeding experiments: general procedure Feeding experiments were designed to measure the feeding preferences of benthic copepods under essentially natural conditions. Undisturbed sediment cores (3.5-cm inner diameter) were taken, and all but 2 cm of the overlying water was removed. Cores were then injected with either 5 pCi of [1,2] "C-acetate (New England Nuclear, 54.7 mCi/mmol) or 15 pCi of NaH-14C05 (New England Nuclear, 10 mCi/mmol) through a self-sealing (silicon) port in the side of the core 0.5 cm below the sediment surface. The amounts of label chosen gave adequate labeling without significantly increasing the concentrations of the substrate above background levels. In particular, the amount of acetate added was less than 10% of the estimated naturally-occurring acetate (R. H. Findlay, personal communication). The amount of bicarbonate added was ca 0.15% of the naturally occurring bicarbonate (Riley, 1971). The label was applied evenly across the core using Findlay's (unpublished) technique of injection during withdrawal of a microliter syringe (Hamilton Co.) that had been inserted horizontally into the sediment. Two equal-volume injections (applied at different angles) per core were used. Five replicate cores were used for each time (4- or 8-h incubation) and label treatment. The incubations were done in the field with the top stoppers removed from the cores to allow gas exchange. The 14C-acetate-labeled cores were intended to measure copepod feeding on bacteria and thus were incubated in the dBrk to prevent photoautotrophs from incorporating respired l4CO9. The NaH-14C0, cores were used to measure copepod feeding on photoautotrophs. These cores were incubated under ambient light. Overheating of cores was prevented by placing the cores in a water-filled tub in the swash zone of the beach during incubation. At the end of an incubation, the 2 cm of overlying water were aspirated off and preserved in 4% formaldehyde along with the top 1 cm of sediment.
Control experiments Poisoned controls. In order to determine the importance of abiotic uptake (adsorption or absorption) of labeled compound by copepods after fixation, we conducted a poisoned-control experiment. In this experiment, sediment cores were taken from the study site, and the overlying water and top 1 cm of sediment were preserved in 4% formaldehyde. Labeled substrate was added to the poisoned samples (the same amounts used in the feeding experiments), and samples were mixed thoroughly to assure an even distribution of the radioactive material. Three
replicates of the poisoned-control experiment were performed for each substrate. Dark controls. To be certain that the photoautotrophfeeding experiment was a ~ t m l l ymeasuring feeding on organisms that photosynthet$ca*ly fix COz, we performed a dark-control experiment. Thme replicate cores were treated in the manner describ94 in the feeding experiments, except that they were incubated in the dark for 8 h.
Laboratory treatment o Experimental animals from all treatmgqts were subjected to 6 d sseparated the same laboratory protocol. ~ o ~ i ~ were from the sediment and concentrated on"a 62-pm sieve using the Barnett (1968) troughing procedure. Copepods were then stained with rose bengal and rinsed on a 62-pm sieve. The copepods of interest were removed under a dissecting microscope, identified to species, counted, and rinsed in distilled water. Ca~epadsjwecesorted by species into liquid-scintillatioq-cou~ter (LSC) vials containing 0.5 ml of distilled water. Before counting, 0.5 ml of Protosol (New England Nuclear) was added, and the vials were heated (60 "C, 24 h) to solubilize the copepod tissue. After solubilization, 10 ml of Aquasol (New England Nuclear) was added, and the vials were placed in the dark for 24 h to allow chemoluminescence resulting from the Aquasol/ Protosol reaction to subside. Radioactivity (counts per minute - CPM) was measured on a Beckman model LS-100C LSC machine. CPM were converted to DPM (disintegrations per minute) using external-standardschannels ratios and a standard-quench curve for 14Ctoluene.
Statistical procedures Friedman's test (Hollander and Wolfe, 1973), a nonparametric analog of two-way analysis .of variance, was used to determine whether an overall difference could be detected in the amounts of radioactivity ingested by the three copepod species. The number of individuals per species per core varied, precluding an: accurate determination of the variance of mean DPM per individual for each species, and thus necessitated the use of non-parametric statistics. Page's test for ordared alternatives (based on Friedman-rank sums) (Hollander and Wolfe, 1973) was used to determine whether a significant hierarchy, in terms of radioactivity ingested, was presbnt for the three copepod species. Page's test examines the significance of the hierarchy A 2 B 2 C (with ona inequality being strict). The test is not as powerful as multiple comparisons, where a test of A > B > C can be made, but it is more informative than the overall ANOVA. For example, if the hierarchy A 2 B 2 C is significant, it can be concluded that A > C, A S B,andB 2 C.
K. R,Carrnan and D. Thistle: Copepod feeding
Results Poisoned controls None of the three copepod species took up radioactively labeled compounds ("C-acetate or NaH-14C03)abiotically in formaldehyde-treated samples, so adsorption or absorption of free, labeled compounds during the period between fixation and troughing did not contribute to the radioactivity found in copepods.
Table 1. NaH-14C0, Dark-control experiments. Data are uptake of radioactivity (DPM/individual) accumulated by copepod species during an 8-h incubation. Incubations were performed in the dark to prevent photosynthetic fixation of 14C0,. Numbers of individuals per species per core are in parentheses Replicate
Thompsonula hyaenae
HalicycZops coufli
Zausodes arenicolus
NaH-14C03experiments
Dark controls. Feeding experiments conducted with NaH14C0,-labeled sediment cores that were incubated in the dark revealed a low uptake of radioactivity by copepods (Table 1). Photoaulotroph-feeding experiment. Results of the lightincubated, photoautotroph-feeding experiment are presented in Table 2 and Fig. 1. The Friedman test (Hollander and Wolfe, 1973) indicates a significant difference overall among species in radioactivity ingested (p Thompsonula hyaenae, and in relation to herbivony Thornpsonula hyaenae > Zausodes arenicolus > H@Bcyclopscoulli. If it is assumed that different species of bacteria and photoautotrophs take up labeled compounds at different rates (or not at all), then radioactivity in copepods is a reflection of their feeding on variously labeled bacteria and photoautotrophs. The assumption of differential uptake by microbes is undoubtedly more accurate than that of uniform uptake (Fenchel and Jsrgensen, 1977; Findlay and White, 1984), making interpretation of the data difficult. The data become less dificult to interpret when the bacterium- and photoautotroph-feeding experiments are considered simultaneously and in conjunction with observations on live animals. The relatively low levels (compared to Halicyclops coulli and Zausodes arenicolus) of radioactivity observed in Thompsonula hyaenae in the bacterium-feeding experiment does not, for the reasons discussed above, preclude the possibility that T. hyaenae feeds on bacteria. However, the relatively high radioactivity observed in II: hyaenue in the photoautotroph-feeding experiment suggests that herbivory plays an important role in its nutrition. Sellner (1976) found that T. hyaenae could be cultured on a diet of diatoms and preferred a mixture of species over any one species of diatom. We observed that most live T , hyaenae had green material in their guts, suggesting they were feeding on plant material. These data indicate a primarily herbivorous mode of nutrition for T. hyaenae. Halicyclops coulli showed relatively high rates of uptake in the bacterium-feeding experiment and low rates of uptake in the photoautotroph-feeding experiment. The guts of live H.coulli did not contain pigmented material,
K.R. Carman and D. Thistle: Copepod feeding
suggesting that herbivory plays a minor role in its nutrition. H.coulli appears to be bactivorous. Zausodes arenicolus ranked second in both bacteriumfeeding and photoautotroph-feeding experiments, suggesting that this species is an omnivore. Ravenel and Thistle (1981) concluded that the field distribution of 2. arenicolus was not influenced by sediment microbes. This result suggests a non-selective feeding habit, which is consistent with the results presented here. Observations on live 2.arenicolus from the field revealed varying amounts of green material in their guts. The three species of co-occurring benthic copepods studied here feed on different microbial food resources in their natural environment. This finding is consistent with the hypothesis that meiofaunal distributions are influenced by heterogeneous food sources (Lee et al., 1977; Findlay, 1981). The idea that food sources are "heterogeneous" is not new, but exactly what is meant by heterogeneity is often poorly defined. It has been shown that heterogeneous distributions of total microbial biomass do not account for the patchy distribution of meiofauna (Dauer et al., 1982; Montagna et al., 1983). Our findings suggest that important food-source heterogeneity is present at the broad level of microbial types, i.e. photoautotrophs and bacteria. Acknowledgements. These people helped us in the field: F. Dobbs, R. Findlay, G. Mitchum, J. Reidenauer, and S. Welsh. J. Reidenauer and T. Bowman advised on copepod identifications. D. C. White and P. A. LaRock provided laboratory space for work with radioactive material. These people commented on the manuscript: F. Dobbs, P. Froelich, J. Guckert, P. LaRock, J. Lawrence, J. Reidenauer, K. Sherman, A. Thistle, R, Varon, and D. White. The work was supported, in part, by grants from the Society of the Sigma Xi and the Department of Oceanography, F.S.U. (K. Carman) and Office of Naval Research contract N00014-82-C-0404 (D. Thistle). We express our thanks for this kind help. This paper is Contribution No. 1024 of the Florida State University Marine Laboratory.
Literature cited Alexander, M.: Microbial ecology, 5 11 pp. New York: John Wiley and Sons 1971 Anderson, J. W. and G. C. Stephens: Uptake of organic material by aquatic invertebrates. VI. Role of epiflora in apparent uptake of glycine by marine invertebrates. Mar. Biol. 4, 243-249 (1969) Barnett, P. R, 0.: Distribution and ecology of harpacticoid copepods of an intertidal mudflat. Int. Revue ges. Hydrobiol. Hydrogr. 53, 177-209 (1968) Bell, S. S.: Meiofauna-macrofauna interactions in a high salt marsh habitat. Ecol. Monogr. 50, 487-505 (1980) Brown, T. J. and J. R. Sibert: Food of some benthic harpacticoid copepods. J. Fish. Res. Bd Can. 34, 1028-1031 (1977) Cadee, G. C. and J. Hegeman: Primary production of the benthic microflora living on tidal flats in the Dutch Wadden Sea. Neth. J. Sea Res. 8,260-291 (1974) Coull, B. C.: Shallow water meiobenthos of the Bermuda Platform. Oecologia 4, 325-357 (1970)
Dauer, D. M., R. M. Ewing, G. H. Tourtellotte, T. W. Harlan, J. W. Sourbeer and H. R. Barker, Jr.: Predation, resource limitation and the structure of benthic infaunal communities of the lower Chesapeake Bay. Int. Revue ges. Hydrobiol. Hydrogr. 67,477-489 (1982) Ehrlich, H. L.: Geomicrobiology, 393 pp. New York: M. Dekker, Inc. 1981 Fenchel, T. M, and B. B, Jerrgensen: Detritus food chains of aquatic ecosystems: the role of bacteria. In: Advances in microbial ecology, pp 1-58. Ed. by M. Alexander. New York: Plenum Press 1977 Findlay, R. H. and D. C. White: In situ determination of metabolic activity in aquatic environments. Microbiol. Sci. I, 90-95 (1984) Findlay, S. E. G.: Small-scale spatial distribution of meiofauna on a mud- and sandflat. Estuar, cstl Shelf Sci. 12,471-484 (1981) Cerlach, S. A.: Food-chain relationships in subtidal silty sand marine sediments and the role of meiofauna in stimulating bacterial productivity. Oecologia 33,55-69 (1978) Gray, J. S.: The attractive factor of intertidal sands to Protodrilus symbioticus Giard. J. mar. biol. Ass. U.K. 46,627-645 (1966) Gray, J. S.: Substrate selection by the archiannelid ProtodriIus hypoleucus Armenante. J. exp. mar. Biol. Ecol, I, 47-54 (1967) Gray, J. S.: An experimental approach to the ecology of the harpacticoid Leptastacus constrictus Lang. J, exp. mar. Biol. Ecol. 2,278-292 (1968) Gray, J. S. and R. M. Johnson: The bacteria of a sandy beach as an ecological factor affecting the interstitial gastrotrich Turbanella hyalina Schullze. J. exp. mar. Biol. Ecol. 4, 119-133 (1970) Gray, J. S. and R. Rieger: A quantitative study of the meiofauna of an exposed sandy beach at Robin Hoods Bay, Yorkshire. J. mar. biol. Ass. U.K. 51, 1-19 (1971) Heip, C.: The spatial pattern of Cyprideis torosa (Jones, 1850) (Crustacea: Ostracoda). J. mar. biol. Ass. U.K. 56, 179-189 (1976) Hicks, G. R. F. and B. C. Coull: The ecology of marine meiobenthic harpacticoid copepods. Oceanogr, mar. Biol. Ann. Rev. 21,67-175 (1983) Hollander, M. and D. A. Wolfe: Nonparametric statistical methods, 503 pp. New York: John Wiley and Sons 1973 Jannasch, H. W, and G. E. Jones: Bacterial populations in seawater as determined by different methods of enumeration, Limnol. Oceanogr. 4, 129-139 (1959) Jansson, B. 0.: Microdistribution of factors and fauna in marine sandy beaches. Veroff. Inst. Meeresforsch. Bremerh. (Sonderbd) 2,77-86 (1966) Jansson, B. 0.: The significance of grain size and pore water content for the interstitial fauna of sandy beaches. Oikos 18, 3 11-322 (1967) Joint, I. R.: Microbial production of an estuarine mndflat. Estuar. cstl mar. Sci. 7, 185-195 (1978) Lee, J. J., J. H. Tietjen, C. Mastropaolo and H. Rubin: Food quality and the heterogeneous spatial distribution of meiofauna. HelgolLnder wiss. Meeresunters. 30, 272-282 (1977) McLachlan, A., P. E. D. Winter and L. Botha: Vertical and horizontal distribution of sub-littoral meiofauna in Algoa Bay, South Africa. Mar. Biol. 40,355-364 (1977) Montagna, P. A.: In situ measurement of meiobenthic grazing rates on sediment bacteria and edaphic diatoms. Mar. Ecol. Prog. Ser. 18, 119-130 (1984) Montagna, P. A,, B. C . Coull, T. L. Herring and B. W. Dudley: The relationship between abundances of meiofauna and their suspected microbial food (diatoms and bacteria). Estuar. cstl Shelf Sci. 17,381-394 (1983) Munro, A. L. S. and T. D. Brock: Distinction between bacterial and algal utilization of soluble substances in the sea. J. gen. Microbiol. 51, 35-42 (1968) Ravenel, W. S, and D. Thistle: The effect of sediment characteristics on the distribution of two subtidal harpacticoid copepod species. J. exp. mar. Biol. Ecol. 50, 289-301 (1981)
K. R. Caman and D. Thistle: Copepod feeding Revsbech, N. P., B. B. Jorgensen and 0. Brix: Primary production of microalgae in sediments measured by oxygen microprofile, H14C0, fixation, and oxygen exchange methods. Limnol. Oceanogr. 26,717-730 (1981) Rieper, M.: Feeding preferences of marine harpacticoid c~pepods for various species of bacteria. Mar. Ecol. Prog. ler. 7, 303-307 (1982) Riley, J. P.: Introduction to marine chemistry, 465 pp. New York: Academic Press 1971 Sellner, B. W.: Survival and metabolism of the harpacticoid copepod Thompsonula hyaenae (Thompson) fed different diatoms. Hydrobiologia 50,233-238 (1976) Steemann Nielsen, E.: The use of radioactive carbon (C14) for measuring organic production in the sea. J. Cons. perm. int. Explor. Mer 18, 117-140 (1952) Thistle, D.: Harpacticoid dispersion patterns: implications for deep-sea diversity maintenance. J. mar. Res. 36, 377-397 (1978) Thistle, D., J. A. Reidenauer, R. H. Findlay and R. Waldo: An experimental investigation of enhanced harpacticoid (Copepoda) abundances around isolated seagrass shoots. Oecologia 63,295-299 (1984) Tietjen, J. H.: Population distribution and structure of the freeliving nematodes of Long Island Sound. Mar. Biol. 43, 123-136 (1977)
Ustach, J. F.; Algae, bacteria and detritus as food for the pseudonunni Coull and harpacticoid wpe;pod, HeVer~p~yllu~ Palmer. J. exp. qar.,Bipl. Eaol. 64, 203-214 (1982) Vanden Berghe, W. and h4. Bergmans: Differential food preferences in lhrbe co.'&bcml.ing species of Tisbe (Copepoda, Harpacticoida). MM. &ol. Ptog. Ser. 4, 213-219 (1981) Van Raalte, C., W. C. S.tqvart I. VaPela: A 14Ctechniaue for measuring algJ prod~cthvityiq,rwllt marsh muds. Bot. dar. 17, 186-188 (1974) White, D. C., R. S. Bobbie, J. h Herron. 5. D. Kinn and S. J. Morrison: Biochemical rmelosu$6&h?s bf microbid mass and pp69-81. American activity from environmental s~n~lpk@s, Society for Testing and Materia18,BulpJlcqioq 695, 1979 Wieser, W.: Benthic studies in Buzzards Bay*11: the meiofauna. Limnol. Oceanogr. 5, 121- 137 (1960) Wright, R. T. and J. E. Hobbie: Use 04 gfucose a d ' acetate by bacteria and algae in aquatic ecosystems. Eoology 4@,447-464 (1966)
Date of final manuscript acceptance: May 15, 1985. Communicated by J. M. Lawrence, Tampa
