UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation of the growth and feeding responses of oligotrich ciliates to food types and concentrations… Montagnes, David J. S. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1993_fall_phd_montagnes_david.pdf [ 8.99MB ]
Metadata
JSON: 831-1.0053328.json
JSON-LD: 831-1.0053328-ld.json
RDF/XML (Pretty): 831-1.0053328-rdf.xml
RDF/JSON: 831-1.0053328-rdf.json
Turtle: 831-1.0053328-turtle.txt
N-Triples: 831-1.0053328-rdf-ntriples.txt
Original Record: 831-1.0053328-source.json
Full Text
831-1.0053328-fulltext.txt
Citation
831-1.0053328.ris

Full Text

AN INVESTIGATION OF THE GROWTH AND FEEDING RESPONSES OFOLIGOTRICH CILIATES TO FOOD TYPES AND CONCENTRATIONS: AN APPROACHTO ASSESSING THE POTENTIAL OF MARINE PLANKTONIC CILIATE BLOOMSbyDAVID JOSEPH SELWYN MONTAGNESB.Sc., The University of Guelph, 1983M.Sc., The University of Guelph, 1986A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Department of Oceanography)We accept this thesis as conformingto the required standardTHE UNIVERSITY 0 BRITISH COLUMBIASeptember 1993©David Joseph Selwyn Montagnes, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of OceanographyThe University of British ColumbiaVancouver, CanadaDateDE-6 (2/88)iiABSTRACTPlanktonic ciliates consume small phytoplankton and can be important in the transfer of carbonthrough food webs. This study examined the impact of clonal ciliate populations on short term algalblooms. Numerical and functional responses (growth and grazing rates with varied food concentration)were established for 5 marine planktonic ciliates and were used in a model to examine predator-preydynamics of ciliates and 8 Am algae. The ciliates Strombidinopsis acuminatum, Strobilidium spiralis,Strobilidium sp., Strombiclium acuminatum, and Strombiclium capitatum were isolated from BritishColumbian waters and maintained in culture. Ciliates were fed the flagellates Isochrysis galbana,Chroomonas sauna, Rhodomonas lens, and the diatom 7halassiosira pseudonana, individually or incombinations. Numerical responses were obtained by keeping ciliates in semi-continuous culture,measuring growth rates and fitting them to a modified Michaelis-Menten function; this provided bothgrowth and mortality rates. Species specific differences existed in numerical response parameters.Functional responses were measured by observing the uptake of fluorescently labeled C. sauna or 5itm beads. This was a poor method: when measured grazing rates were compared to those predictedby a bioenergetic formula, the measured rates either over or underestimated predicted rates by severalfold. Functional responses were determined using 1) the bioenergetic formula Ingestion =(growth+respiration)/assimilation efficiency, and 2) volume specific respiration rates.Ciliate functional and numerical responses from this and other studies were compared, and 3responses were established. These were used in a model which simulated ciliate-algal populationdynamics in a non-steady state, where ciliates and copepods encountered a patch of water with adefined initial algal concentration. The model indicated: 1) ciliates bloom over 10-20 d, whencopepods are rare (<1 L4) and algae are initially abundant (>103 mL4); 2) ciliate blooms canprovide 40-50% of the carbon available to copepods, but when copepods are abundant and initial algallevels low, ciliates are not an important carbon source; 3) under "typical" conditions (103 algae mL4,1 copepod L4), ciliates are a link to copepods, but primary production is low; 4) bloom dynamics andcarbon flow through the food web are dependent on the ciliate species present. In general, ciliates maybe, under transient conditions, important as both links and sinks of carbon, but under "typical" coastalconditions, ciliates are not important components of food webs.iiiTABLE OF CONTENTSABSTRACT^ iiTABLE OF CONTENTS^ iiiLIST OF TABLES viiLIST OF FIGURES^ viiiACKNOWLEDGMENTS xiiCHAPTER 1^INTRODUCTION: OLIGOTRICH CILIATES AS FEAST ANDFAMINE ORGANISMS^ 1CHAPTER 2^THE SALIENT FEATURES OF ONE UNDESCRIBED SPECIES ANDREDESCRIPTION OF FOUR SPECIES IN THE CLASSSPIROTRICHEA (CILIOPHORA, OLIGOTRICHIA) WITHNOTES ON THE CULTURING AND BEHAVIOUR OFTHESE SPECIES^ 5Introduction^ 5CHAPTER 31.0 Materials and Methods^ 61.1 Isolation and General Culturing^ 61.2 Fixation and Staining 71.3 Measurements 72.0 Results and Discussion^ 72.1 General Taxonomy 72.2 Description of Species 82.2.1 Strobilidiid Ciliates^ 82.2.2 Strombidiid Ciliates 122.2.3 Strombidinopsid Ciliates 14GROWTH AND GRAZING RATES OF Strombidinopsis acuminatumSTRAIN SPJSC AS A FUNCTION OF FOODCONCENTRATION^ 26Introduction^26PART I GROWTH RATES: Estimating the Numerical Response 261.0 General Methods^ 261.1 First Set of Experiments^ 281.1.1 Methods 281.1.2 Results and Discussion 291.1.2.1 Consistency of Growth Rate^291.1.2.2 Consistency of Prey Concentration^291.1.2.3 Numerical Response 291.2 Second Set of Experiments^ 301.2.1 Methods^ 301.2.2 Results and Discussion 301.3 Third Set of Experiments 311.3.1 Methods 311.3.2 Results and Discussion^ 312.0 Conjugation of Strain SPJSC 333.0 An Estimate of the Inheritance of Growth Rate^343.1 Methods^ 34iv3.2 Results and Discussion^ 344.0 Sources of Bias and Explanations of Methods^374.1 Mortality^ 374.2 Handling of Ciliates 374.3 Food Quality 385.0 Aspects of the Growth Rates and their Ecological Relevance^386.0 A Numerical Response Model^ 39PART II GRAZING RATES: Estimating the Functional Response 407.0 General Methods 407.1 Results and Discussion^ 417.2 Modeling the Response 42CONCLUDING REMARKS^ 42CHAPTER 4^GROWTH AND GRAZING RATES OF Strobilidium spiralis STRAINIA AS A FUNCTION OF FOOD CONCENTRATION^52Introduction^ 52PART I GROWTH RATES: Estimating the Numerical Response 521.0 General Methods^ 522.0 Results^ 533.0 Change in Culture Growth Rate Over Time^544.0 Examining the Growth Estimates and Developing a NumericalResponse Model^ 565.0 Aspects of the Growth Rates and Their Ecological Relevance 57PART II GRAZING RATES: Estimating the Functional Response 576.0 General Methods^ 577.0 Results, Discussion and Modeling the Response^59CONCLUDING REMARKS 59CHAPTER 5^GROWTH AND GRAZING RATES OF Strobilidium sp. STRAINJERC AS A FUNCTION OF FOOD CONCENTRATION 67CHAPTER 6Introduction^ 67PART I GROWTH RATES: Estimating the Numerical Response 671.0 General Methods^ 672.0 Results and Discussion 682.1 General 682.2 Numerical Response^ 693.0 Aspects of the Growth Rates and Their Ecological Relevance 70PART II GRAZING RATES: Estimates of the Functional Response 704.0 General Methods^ 705.0 Results and Discussion 72CONCLUDING REMARKS 73GROWTH AND GRAZING RATES OF Strombidium acuminatumStrain BJSCH AS A FUNCTION OF FOODCONCENTRATION^78Introduction^ 78PART I GROWTH RATES: Estimating the Numerical Response 781.0 General Methods 782.0 Results and Discussion 802.1 General 802.2 Numerical Response 803.0 Aspects of the Growth Rates and Their Ecological Relevance 80PART H GRAZING RATES: Estimating the Functional Response 814.0 General Methods 815.0 Results and Discussion 826.0 Swimming Behaviour 82CONCLUDING REMARKS 83CHAPTER 7^GROWTH RATES OF Strombidium capitatum STRAIN APAG AS AFUNCTION OF FOOD CONCENTRATION^89Introduction^ 89PART I GROWTH RATES: Estimating the Numerical Response 891.0 General Methods^ 892.0 Results and Discussion 903.0 Aspects of the Growth Rates and Their ecological Relevance^90PART II GRAZING RATES^ 914.0 A Discussion of Grazing Experiments.^ 91CHAPTER 8^AN EXAMINATION OF THE FUNCTIONAL AND NUMERICALRESPONSES OF OLIGOTRICHS 95Introduction^ 951.0 Reasons for Response Variation^ 951.1 Methodological Biases and the Use of Steady StateMeasurements^ 951.1.1 Methodological biases 951.1.2 The Use of Steady State Measurements^961.2 Use of Surrogate Particles to Determine Grazing Rates^981.3 Physiological-Behavioural Variables^1001.4 Species Differences^ 1032.0 Ciliate Growth and Grazing Responses for Food Web Modeling 103CHAPTER 9^A MODEL OF CILIATE AND PHYTOPLANKTON POPULATIONDYNAMICS^ 116Introduction 1161.0 The Model (see Appendix 4)^ 1171.1 General Description 1171.2 Constraints^ 1181.3 Parameters and Rate Equations 1181.3.1 Ciliates 1181.3.2 Copepod 1181.4 Variables^ 1201.5 A Summary of Assumptions^ 1201.6 Equations 120vi2.0 Results of the Model and Discussion of their Implications^1222.1 Ciliate Blooms Control Blooms of Small Phytoplankton 1222.2 Ciliates can Bloom Over 10 to 20 Day Periods 1242.3 Ciliates Are Both a Link and Sink, and Ciliates Can CompeteWith Copepods^ 1252.4 Species Differences Are Important^1282.5 How Ciliate Blooms Influence Carbon Flow to Copepods 129CONCLUDING REMARKS^ 130LITERATURE CITED^ 143APPENDIX 1^THE GROWTH MEDIUM USED IN THIS STUDY TO GROW BOTHCILIATES AND PHYTOPLANKTON^153APPENDIX 2^CONSUMPTION AND FILTRATION: Their Derivation andApplication.^ 155APPENDIX 3^THE ANNUAL CYCLE OF OLIGOTRICHS AND <10 /AMPHYTOPLANKTON IN SECHELT INLET, A BRITISHCOLUMBIAN FJORDAL SYSTEM^158APPENDIX 4^THE CODE FOR A BASIC MODEL TO INVESTIGATE THE ROLEOF CILIATES AS GRAZERS OF SMALLPHYTOPLANKTON^ 172vi iLIST OF TABLESTable 3.1. Growth and grazing data for Strombidinopsis acuminatumstrain SPJSC; parameters and estimates of error of the numericaland functional response equations presented in Figs. 3.3-3.6.^44Table 3.2. The increase in cell numbers and survival of cell lines ofStrombidinopsisacuminatum strain SPJSC.^ 45Table 4.1. The relative growth rate of Strobilidium spiralis strain IAover a 2 and a 4 day period when fed on all the combinations ofthe three prey: Isochlysis galbana, Chroomonas sauna and,Rhodomonas lens.^ 61Table 4.2. Growth and grazing parameters for Strobilidium spiralisstrain IA; parameters and estimates of error of the numerical andfunctional response curves presented in Figs. 4.1 and 4.4.^62Table 5.1. The relative growth rate of Strobilidium sp. strain JERC overa 5 day period when fed on all the combinations of the threeprey: Isochiysis galbana, Chroomonas salina, and Rhodomonas lens.^74Table 5.2. Growth and grazing parameters for Strobilidium sp. strainJERC; parameters and estimates of error of the numerical andfunctional response curves presented in Figs. 5.1 and 5.2.^75Table 6.1. The relative growth rate of Strombidium acuminatum strainBJLSC over a 4 day period when fed on all the combinations ofthe three prey: Isochrysis galbana, Chroomonas salina,and Rhodomonas lens.^84Table 6.2. Growth and grazing parameters for Strombidium acuminatumstrain BJLSC; parameters and estimates of error of the numericaland functional response curve presented in Figs. 6.1 and 6.2.^85Table 7.1. The relative growth rate of Strobilidium capitatum strainAPAG over a 7 day period when fed on all the combinations ofthe three prey: Isochlysis galbana, Chroomonas sauna, andRhodomonas lens.^ 92Table 7.2. Growth parameters for Strombidium capitatum strain APAG;parameters and estimates of error of the numericalresponse curve presented in Fig. 7.1.^ 93Table 8.1. A comparison of numerical response constants for 10 oligotrichs.^105Table 8.2. A comparison of functional response constants for 8 oligotrichs.^106Table 8.3. A comparison of error estimates of numerical and functionalresponses (coefficients of variation) from data of the five speciespresented in Chapters 3-7.^ 107Table 8.4. Parameters used to establish ciliate numerical responsecurves for use in the model of ciliate-prey population dynamics in Chapter 9.^108Table 9.1. Parameters and variables used in the model of ciliate-phytoplankton interactions.^ 133viiiLIST OF FIGURESFig. 2.1. Schematic diagram of Strobilidium spiralis (Leegaard, 1915)Lynn and Montagnes, 1988 strain IA from protargol stains andaugmented by electron microscope observations.^ 17Fig. 2.2. Electron micrographs of Strobilidium spiralis (Leegaard,1915) Lynn and Montagnes, 1988 strain IA. 18Fig. 2.3. Schematic diagram of Strobilidium sp. strain JERC fromprotargol stains and augmented by electron microscope observations.^19Fig. 2.4. Electron micrographs of Strobilidium sp. strain JERC.^20Fig. 2.5. Schematic diagram of Strombidium capitatum (Leegaard,1915) Montagnes et al. 1988.^ 21Fig. 2.6. Schematic diagram of Strombidium acuminatum, (Leegaard,1915) Kahl, 1932 strain BJSC from protargol stains andaugmented by electron microscope observations.^ 22Fig. 2.7. Electron micrographs of Strombidium acuminatum, (Leegaard,1915) Kahl, 1932 strain BJSC.^ 23Fig. 2.8. Schematic diagram of Strombidinopsis acuminatum Faure-Fremiet, 1924 strain SPJSC from protargol stains and augmentedby electron microscope observations.^ 24Fig. 2.9. Electron micrographs of Strombidinopsis acuminatumFaure-Fremiet, 1924 strain SPJSC. 25Fig. 3.1. The variability of growth rate of Strombidinopsis acuminatumstrain SPJSC at nine prey concentrations over a nine day period.^46Fig. 3.2. An indication of the variability of food concentration duringgrowth experiments on Strombidinopsis acuminatum strainSPJSC, over a five day period.^ 47Fig. 3.3. The numerical response of Strombidinopsis acuminatum strainSPJSC from three separate experiments.^ 48Fig. 3.4. The change in growth rate with number of generations forStrombidinopsis acuminatum strain SPJSC grown on saturatingfood concentrations.^ 49Fig. 3.5. The numerical response of Strombidinopsis acuminatum strainSPJSC from a combination of three experiments; the distributionof the residuals for the Michaelis-Menten fit to all the data inFig. 3.5a; and the distribution of the residuals for the Michaelis-Menten fit to a subset of the data in Fig. 3.5a.^ 50Fig. 3.6. The change in number of fluorescent beads ingested byStrombidinopsis acuminatum strain SPJSC over time (at two beadconcentrations), and the functional response of stain SPJSCingesting fluorescent beads.^ 51ixFig. 4.1. The numerical response of Strobilidium spiralis strain IAgrown on the prey flagellate Chroomonas sauna.^ 63Fig. 4.2. The change in culture growth rate of Strobilidiumspiralis strain IA over time.^ 64Fig. 4.3. The number of beads ingested by Strobilidium spiralisstrain IA over time.^ 65Fig. 4.4. The functional response of Strobilidium spiralis strain IAingesting a 1:1 ratio of 5p.m fluorescent beads and Chroomonas sauna.^66Fig. 5.1. The numerical response of Strobilidium sp. strain JERCgrowing on a 1:1 ratio of the prey flagellates Chroomonas saunaand Isocluysis galbana.^ 76Fig. 5.2. The functional response of Strobilidium sp. strain JERCingesting a 1:1 ratio of stained and live Chroomonas sauna.^77Fig. 6.1. The numerical response of Strombidium acuminatum strainBJLSC growing on two different diets: 1) the diatomThalassiosira pseudonana and 2) a 1:1:1 combination of the threeflagellates Isochtysis galbana, Chroomonas sauna, and Rhodomonas lens.^86Fig. 6.2. The functional response of Strombidium acuminatum strainBJLSC grazing on a 1:1 mixture of the diatom Thalassiosirapseudonana and 5 gm fluorescent beads.^ 87Fig. 6.3. The change in position (in the water column vs. on a surface)of Strombidium acuminatum strain BJLSC at different prey(Thalassiosira pseudonana) concentrations.^ 88Fig. 7.1. The numerical response of Strombidium capitatum strainAPAG growing on a 1:1 ratio of the two flagellates Isocluysisgalbana and Chroomonas salina.^ 94Fig. 8.1. The growth rate (numerical response) of 10 oligotrichs inresponse to varied prey concentration, as prey numbers and prey carbon.^109Fig. 8.2. The grazing rate (functional response) of eight oligotrichs inresponse to varied prey concentration, as numbers and prey carbon.^110Fig. 8.3. A comparison of observed grazing rates and those predictedfrom equation 3 (Chapter 8), using growth rates predicted byvalues presented in Chapters 3-6, equation 1 (Chapter 3), andequation 3 (Chapter 8).^ 111Fig. 8.4. A comparison of gross growth efficiencies based on observedgrowth and grazing data and those based on observed growth andpredicted grazing rates (predicted grazing rates were determinedusing equation 3 in Chapter 8 and growth rates predicted byvalues presented in Chapters 3-6 and equation 1, Chapter 3).^112Fig. 8.5. The growth rate of 10 oligotrichs in response to varied preyconcentration, corrected for differences in cell volume andambient temperature.^ 113Fig. 8.6. The grazing rate of eight oligotrichs in response to varied preyconcentration, corrected for differences in cellsize and ambient temperature.^ 114Fig. 8.7. Three growth responses used to represent different ciliatetypes (A-C). These responses were used to model ciliate-nanoplankton prey interactions in Chapter 9.^ 115Fig. 9.1. A schematic diagram of the food web model described in Appendix 4.^134Fig. 9.2. Output from the model of ciliate-phytoplankton dynamics(Appendix 4): the effect of variations in initial algal concentrationand constant copepod abundance on the formation and peakmagnitude of ciliate blooms.^ 135Fig. 9.3. Output from the model of ciliate-phytoplankton dynamics(Appendix 4): five examples of bloom development over 20 daysand the flow of carbon during those 20 days.^ 136Fig. 9.4. Output from the model of ciliate-phytoplanIcton dynamics(Appendix 4): the effect of variations of initial algalconcentration and constant copepod abundance on gross ciliateproduction and gross primary production over a 20 day simulation.^137Fig. 9.5. Output from the model of ciliate-phytoplankton dynamics(Appendix 4): the effect of variations in initial algal concentrationand constant copepod abundance on four bloom parametersmeasured over the 20 day simulation: 1) (algae consumed bycopepods)/(algae produced); 2) (algae consumed byciliates)/(algae produced); 3) (ciliates consumed bycopepods)/(ciliates produced); and 4) (ciliates consumed bycopepods)/(ciliates consumed by copepods + algae consumed bycopepods).^ 138Fig. 9.6. Output from the model of ciliate-phytoplanIcton dynamics(Appendix 4): the effect of variations in initial algal concentrationand constant copepod abundance on algae carbon eaten bycopepods and algae carbon eaten by ciliates over the 20 daysimulation.^ 139Fig. 9.7. Output from the model of ciliates-phytoplankton dynamics(Appendix 4): the effect of variations in initial algal concentrationand constant copepod abundance on the formation and peakmagnitude of ciliate blooms when only one species of ciliatewas included in the model.^ 140Fig. 9.8. Output from the model of ciliate-phytoplanlcton dynamics(Appendix 4): the separate effects of the three ciliate species onbloom development over 20 days and the flow of carbon duringthose 20 days.^ 141Fig. 9.9. Output from the model of ciliate-phytoplankton dynamics(Appendix 4): the effect of initial algal concentration and constantcopepod abundance on four bloom parameters: 1) total (algae +ciliates) carbon eaten by copepods over the 20 day simulation; 2)total (algae + ciliates) carbon eaten by an individual copepodover the 20 day simulation; 3) ciliate carbon eaten by copepodsover the 20 day simulation; and 4) ciliate carbon eaten by anindividual copepod over the 20 day simulation.xi142xiACKNOWLEDGEMENTSOver the last five years there have been a myriad of people who have helped me. Ifyou are missed in these acknowledgments I apologize; accredit it to my poor memory and notmy lack of appreciation. Thank you: M. Adamson, W. Coats, T. de Haan, P. Franks, L.Greenway, R. Haigh, N. Haigh, R. Johnson, D. Jones, E. Lessard, D. Lynn, C. Mewis, D.McPhail, T. Parsons, T. Pedersen, E. Simons, D. Stoecker, T. Sutherland, A. Wait, M.Watt, and M. Weis. I also thank the committee of professors who aided me in conducting thiswork: J. D. Berger, P. J. Harrison, A. G. Lewis, W. Neil, F. J. R. Taylor, and C. F.Walters and my external examiner P. G. Verity.There were several grad' students and post-docs' without whom I would not havesurvived this study. They were always there to listen, encourage, and criticize. I can't thankyou enough: Peter Arthur, John Berges, Philip Boyd, Rob Golblatt, Katriona Hurd and PeterThompson.Finally, there are three people who I would like to dedicate this work to. The first isRamona who takes me to see Shakespeare. The other two are Cheryl Jerome and Alan Martin,who gave me the chance to learn and teach.This study and my living expenses were partially funded by an NSERC graduatefellowship, a University of British Columbia Graduate Fellowship, and NSERC fundingawarded to F. J. R. Taylor.1CHAPTER ONEINTRODUCTION: OLIGOTRICH CILIATES AS FEAST AND FAMINE ORGANISMSBlooms of planktonic ciliates are primarily the result of asexual reproduction and arelargely clonal populations. It is therefore important to understand the population dynamics ofclones, and "to follow the history of a clone accurately, it is essential to study isolatedindividuals in small volumes of medium; this was Maupas' fundamental technicalachievement" (Bell 1988). The purpose of this dissertation was to investigate small scaleplanktonic ciliate blooms and determine their potential impact on carbon flow in food webs, byexamining ciliates in culture.If we know ciliate and prey biomass and growth rates and the grazing rates of theciliates, then we can determine the impact of ciliates on a prey population over time. Bothgrowth and grazing rates change with food concentration, typically as rectangular hyperbolicfunctions (referred to as numerical and functional responses, respectively, see Appendix 2). Ihave investigated the growth rate of ciliates under steady state conditions at a number of preyconcentrations (Chapters 3-7). These experiments provided numerical response curves whichindicated not only the growth rates of the ciliates but also the mortality rates at sub-thresholdconcentrations. I also measured grazing rates at a number of prey concentrations, byobserving the uptake of fluorescently labeled beads and/or flagellates (Chapters 3-7). Thismethod proved unsatisfactory, as it either under or over estimated feeding rates (see Chapter8). I therefore used the numerical response data and a bioenergetic formula to determinefunctional responses (Chapter 8). However, I have presented the grazing rate data, measuredby labeled prey uptake, to indicate that this method, which is often used to estimate ciliategrazing rates, is inaccurate. The numerical and functional responses were used to determinethe impact of ciliates on prey populations (Chapter 9).In the last 150 years it has been established that ciliates are a dominant component ofthe microzooplankton (20-200 Am) and may shunt a substantial portion of energy throughplanktonic food webs (Lynn and Montagnes 1991). Studies have provided information on a2number of aspects of marine planktonic ciliate biology, for example: biomass (Lynn andMontagnes 1991); grazing (Scott 1985, Gifford 1988); growth rate (Rassoulzadegan 1982,Smetacek 1984, Verity 1991a); mixotrophy (Blackbourn et al. 1973, Stoecker et. al. 1988,Putt 1990); behaviour (Buskey and Stoecker 1988, 1989, Fenchel and Jonsson 1988); nutrientflux (Johannes 1965, Stout 1980, Taylor 1982, Verity 1985); distribution (Lynn andMontagnes 1991); migration (Jonsson 1989); food for zooplankton (Jonsson and Tiselius1990, Gifford and Dagg 1991); and taxonomy (Laval-Peuto and Brownlee 1986, Montagnesand Lynn 1991). Many of these studies indicate that ciliates are an important component ofplanktonic trophodynamics.However, Banse (1982) suggested that, in the open ocean, on average ciliates consumelittle food relative to copepods because the concentrations of suitable food particles tend to betoo low. A similar situation may also arise in coastal waters if food concentrations are lowthere too. Thus, ciliates may not normally be important in the flow of carbon in some marinefood webs. Banse (1982) argued that the maximum growth and grazing rates of ciliates are sohigh that, if they grew and fed at these rates, they would rapidly deplete their prey. Thus, thepotential impact of ciliates on phytoplanIcton would be rarely realized, and ciliate ingestion andspecific growth rates would be typically low. Under these conditions, ciliates could only reachtheir potential rates when encountering a short term bloom of small algae, as long term blooms(e.g. the spring bloom) would be exploited by mesozooplankton. During short term algalblooms, ciliates could graze down the algae unless the ciliates are grazed down bymesozooplanIcton.Others have also indicated that planktonic ciliates act as feast and famine organismsover a few days, as Banse suggested (Blackbourn 1974, Ibanez and Rassoulzadegan 1977,Grice et al. 1980, Smetacek 1984, Andersen and Sorensen 1986, Fenchel 1987, Lynn andMontagnes 1991), and that under some conditions ciliates are not an important carbon sourceto upper trophic levels (Montagnes et al. 1988a). Thus, ciliates may only be important forbrief periods in some planktonic ecosystems.3Many non-marine ciliates exhibit a feast-famine existence related to temporal andspatial patches (Fenchel 1987). They can remain dormant in cysts (Corliss and Esser 1974), atreduced metabolic rates (Fenchel 1987), or at very low numbers (Taylor and Shuter 1981),during a famine. Then, at times when food is abundant, ciliates rapidly exploit the resource.Presumably, marine planktonic ciliates also act in this fashion, as some species make cysts(Reid and John 1978, 1983, Paranjape 1980, Reid 1987).In the plankton, food for ciliates may become abundant in small patches, asphytoplanlcton bloom in localized regions. Such phytoplankton blooms appear as rapidincreases in numbers or biomass, visible as transient peaks, and can be stimulated by a numberof factors: tidal or wind mixing events, changes in irradiance, or by allochthonous inputs fromterrestrial run off (Mackas et al. 1985, Legendre 1990). These blooms can be meters tokilometers in size and exist for 10-20 days, when mixing processes (e.g. wind and tides)dissipate them (Haury et al. 1978, Mackas et al. 1985). In contrast, during large persistentblooms of phytoplankton, mesozooplankton populations will respond. Thus, short termphytoplankton blooms represent conditions where ciliates might bloom and where ciliateswould have a selective advantage over mesozooplanIcton, due to their rapid growth rates.Unfortunately, the logistics of sampling marine systems prevent most studies fromdetecting the genesis and following the development of short term blooms. Therefore, ourknowledge of such bloom-dynamics in the field is limited, and most food web analyses havenot acknowledged that short term phytoplankton blooms may be grazed down by ciliates. Oneway to circumvent this problem is to use laboratory data to model blooms. This approachprovides a means to estimate the potential occurrence of such blooms. Further, if we canestablish the conditions under which blooms may occur, then we may reduce our effortsearching for them.Lynn and Montagnes (1991) developed a simple model of the ciliate Strobilidiumspiralis grazing on small flagellates. Their study was based on functional and numericalresponse data obtained from work on S. spiralis (Jonsson 1986). This model indicated thatciliate blooms occur over <20 days and can graze down prey populations.4My goal was to improve the data-base used to model planktonic ciliate blooms, andthen employ it to: 1) further substantiate that field observations of these ciliate blooms are dueto the rapid growth rates of ciliates and 2) show that small scale phytoplankton blooms may begrazed down by ciliate blooms. To do this I have: 1) cultured and identified 5 differentciliates (Chapter 2); 2) obtained functional and numerical responses from these ciliates(Chapters 3-7); 3) compared my data to that obtained in other studies and established 3"typical" ciliate responses (Chapter 8); and 4) used the functional and numerical response datato developed a simple model which illustrates that short term ciliate blooms can occur andindicates conditions under which blooms may exist (Chapter 9). Finally, I have investigatedthe dynamics of these blooms in terms of carbon flow through a phytoplankton-ciliate-copepodfood web (Chapter 9).5CHAPTER 2THE SALIENT FEATURES OF ONE UNDESCRIBED SPECIES ANDREDESCRIPTION OF FOUR SPECIES IN THE CLASS SPIROTRICHEA(CILIOPHORA, OLIGOTRICHIA) WITH NOTES ON THE CULTURING ANDBEHAVIOUR OF THESE SPECIESIntroductionThe "oligotrichs" have been recognized as a distinct assemblage of ciliates, beginningwith Butschli's classification scheme of the 1880's and continuing up to the present (Small andLynn 1985, Montagnes and Lynn 1991, Lynn and Corliss 1991, Petz and Foissner 1992).Corliss (1979) noted in his chapter on the "often neglected oligotrichs" that the literature onthis group, excluding the monographic works on the tintinnines (e.g. Kofoid and Campbell1939), is not very extensive. He commented that a major modern treatise is long overdue.Recently, there have been a number of works that have begun this arduous task (Maeda andCarey 1985, Maeda 1986, Grim 1987, Foissner et al. 1988, Lynn and Montagnes 1988, Lynneta!. 1988, Montagnes eta!. 1988b, Krainer 1991, Montagnes and Lynn 1991, Petz andFoissner 1992, Martin and Montagnes 1993). Still, it is typical that in routine sampling ofmarine waters new ciliate species are found (e.g. Martin and Montagnes 1993). This suggeststhat work on the oligotrichs is far from finished.Most of the taxonomic work conducted on the oligotrichs has used field samples todescribe morpho-species and has assumed that variation in these samples would be sufficient toappraise the variability within a species. Further, field based taxonomy assumes that a speciescan be identified (morphologically) from a single cell. Consequently, species descriptionsfrom field samples may on one hand underestimate the variability of a species and on theother, lump species together. Thus, taxonomic studies on cultured isolates will help to indicatethe variability within and between species.In this chapter, I present the salient features of five species (my experimentalorganisms) obtained from examining protargol stained, scanning electron micrographed and6live specimens. With the salient features, I have included some remarks on the culturing andbehaviour of the ciliates.1.0 Materials and Methods1.1 Isolation and General CulturingOligotrichs have been cultured during the past 20 years (e.g. Gold 1970) and in the last10 years explicit methods have been developed (see Gifford 1985). I have roughly followedthe procedures of these authors (see Chapters 3-7).Strains of marine planktonic oligotrichs were collected in bottles from coastalsubsurface waters of British Columbia. Then, 5-10 mL of the sample water was placed in 20-mL plastic culture plates (Falcon, 3046, Becton Dickinson & Co., Lincolin Park, NJ, USA).Samples were enriched with natural seawater (Appendix 1) and the putative prey: Chroomonassauna NEPCC 275, Isochlysis galbana NEPCC 633, Rhodomonas lens NEPCC 588, andsometimes Thalassiosira pseudonana NEPCC 58 (species were obtained from the NortheastPacific Culture Collection, NEPCC, Department of Oceanography, University of BritishColumbia, Vancouver, British Columbia). Ciliate and phytoplankton cultures were maintainedat 16-17°C on a 14:10 light:dark cycle at 30-70 Amol photons m-2 s-1.Following enrichment, if ciliate numbers increased, some ciliates were removed (usingfinely drawn Pasteur pipettes), transferred through 3-4 washes of sterile culture medium andplaced in medium containing defined prey species in tissue culture plates. After severalgenerations, the above process was repeated to ensure there were no contaminating eukaryotes(cultures were never made bacteria free). Then, cultures were maintained by serial transfers,either to 50 mL of medium and defined prey in 125-mL flasks or to 10 mL of medium in 20mL tissue plates.At a later date, after cultures had been well established, clonal cultures were made byisolating single ciliates. The single cells were allowed to divide several times and then asecond single-cell isolation was repeated; this procedure ensured the cultures were clonal.71.2 Fixation and StainingFor light microscopy, the ciliates were fixed in 5-10% Bouin's solution(volume/volume) (Lee et al. 1985) and protargol silver stained (Montagnes and Lynn 1987,1993). For scanning electron microscopy, the ciliates were fixed in 5% Bouin's or 2% acidLugol's iodine, subsequently dehydrated in ethanol, critical point dried in a Balzers CPD 020,mounted on aluminum stubs, gold coated (25 nm) in a Nanotech SEMPREP 2 sputter coater,and viewed with a Cambridge 250T SEM.1.3 MeasurementsExamination of cells followed the recommendations of Montagnes and Lynn (1991).Longitudinally oriented, protargol stained specimens were examined. The following featureswere measured: somatic length as the maximum longitudinal linear distance, excluding cilia;somatic width as the maximum linear distance (diameter) across the cell at right angles to thelongitudinal axis; macronuclear diameter; number of polykinetids in the anterior or externalpolykinetid zone; and number of polykinetids in the ventral or internal polykinetid one. Allanterior, external and internal polykinetid counts were made on polar orientated cells exceptfor Strombidium acuminatum strain BJSCH in which polar orientation was extremely rare.Counts of the ventral and internal polykinetids may be highly variable due to the orientation ofcells, which often obscured the smaller polylcinetids in the cytostomal region. Further, thedimensions given are based on Bouin's-fixed and protargol-stained material, which are -0.85of live measurements (Jerome et al. 1993). Also, fixation, by inducing some contraction inthe ciliates, may also distort the natural path of structures.2.0 Results and Discussion2.1 General taxonomyThe five ciliate species presented below belong to the class Spirotrichea, subclassOligotrichia (following Lynn and Corliss 1991), and are in the Order Oligotrichida, familyStrombidiidae and Order Choreotrichida, families Strobilidiidae and Strombidinopsidae (Smalland Lynn 1985).8Oligotrichs are often the dominant ciliates in the marine plankton (Montagnes and Lynn1991). This subclass is typically conical or ovoid in shape with specialized oral polyldnetidsused mainly for locomotion and feeding. The somatic ciliature is specialized and may bereduced to non-ciliated kinetids. The Oligotrichida are distinguished by the oral polykinetidsbeing divided into anterior polykinetid (APZ) and ventral polylcinetid zones (VPZ), while theChoreotrichida are distinguished by the oral external polyldnetids (EPZ) forming a completecircle that encloses the internal polykinetid zones (IPZ) (Faure-Fremiet 1969, Small and Lynn1985). Note that different diagnostic features have been used to characterize the oligotrichs(see Petz and Foissner 1992).I have been conservative in the identification of these strains of oligotrichs from theeastern north Pacific, attributing to them the names of species from the north Pacific, easternnorth Atlantic and North Sea, whenever appropriate.2.2 Description of species2.2.1 Strobilidiid ciliatesStrobilidium spiralis (Leegaard, 1915) Lynn and Montagnes, 1988 strain IA(Choreotrichida, Strobilidiidae)(Figs. 2.1, 2.2)Salient features.Cell shape, subspherical, although flattened on one side due to the characteristicstructure of ldnety 2. Cell length, 40 Am (range, 33-57) and width, 45 Am (38-55). Externalpolylcinetidal zone (EPZ) composed of 38 (40-36) external polylcinetids (EPk). Internalpolykinetidal zone (IPZ) composed of 13 (9-20) internal polykinetids (IPk). IPk's and EPk'susually contiguous but those further from the cytostome may be continuous. The IPZ sitabove an acentric concavity that leads to the cytostome. Cytopharyngeal fibers present.Paroral ldnety begins parallel to the last IPk (near the cytostome) and lies internal to theEPk's. Cilia of the paroral ldnety lie on the oral surface inside the region defined by the EPK.Five somatic kineties (K) present, whose cilia (2-3 Am long) are directed to the right (whenviewed aborally). The somatic lcineties are partially covered by a cytoplasmic flap. K1 is9slightly dexterally spiralled and extends from the posterior of the cell to just below the EPZ.K2 describes an arc that encloses a space between K1 and K2 by extending from the posteriorof the cell to a position half way along Kl; at this point K2 reflects away from K1 andproduces a small (2-4 pm) "crook". K3 and K4 are simple; they originate perpendicular toand near the aboral third of K2 and extend anteriorly to just below the EPZ. K5 originatesone-third of the cell length from the aboral end and runs anteriorly, paralleling K 1 . One C-shaped macronucleus, positioned anteriorly around the oral cavity and below the EPZ, has itsopening near the cytostome. Two micronuclei (not always visible) are indented into themacronucleus 1800 around from the cytostome. Cell surface often covered by 2-4 Am rods(possibly extrusomes) which stain darkly with protargol.Time and locality of isolationEarly March, Indian Arm, British Columbia, Canada, (122°53'W, 49°21'N,) at adepth of 2 m, temperature of 8 °C, and salinity of 16 %..Discussion of speciesBefore discussing strain IA specifically, I will provide a brief definition of the genusStrobilidium. The family Strobilidiidae sensu stricto exhibit the following characters: somatickineties with short cilia overlain by a cytoplasmic flap and a closed circle of externalpolylcinetids; the genus Strobilidium possesses these same characters (Lynn and Montagnes1988). Following these criteria Lynn and Montagnes (1988) made Lohmanniella spiralis,Leegaard, 1915 (a commonly referred to marine ciliate) a junior synonym for Strobilidiumspiralis. More recently, Petz and Foissner (1992) have placed a number of species ofStrobilidium in the genus Rimostrombidium Jankowski, 1978. Their arguments for doing thishowever are in conflict with the criteria proposed by Lynn and Montagnes (1988) andresurrection of the "desk genus" Rimostombidium for this group creates unnecessaryetymological confusion. I therefore have maintained the genus Strobilidium for myidentifications.Strain IA is identical to Strobilidium spiralis (Leegaard, 1915) Lynn and Montagnes,1988 except for two features: 1) S. spiralis has only one micronucleus indented into the10macronucleus (1800 around from the cytostome), while strain IA has two, and 2) K5 on S.spiralis originates anteriorly, extends posteriorly, and then curves anteriorly around to describea partial circle; but K5 on strain IA originates one-third of the cell length from the aboral endand runs anteriorly, paralleling Kl. These two differences may denote a new species.However, I have been conservative and considered stain IA to be Strobilidium spiralis.Remarks on culturing and behaviourStrain IA was grown on three flagellates in culture: Chroomonas sauna, Isochrysisgalbana, and Rhodomonas lens. However, it first was grown on a mixed culture of naturalprey; ciliates isolated from these cultures had ingested 30-40 /Am pennate diatoms (likelyNitzschia). Strain IA was also used to estimate the effect of Bouin's and Lugol's fixatives andProtargol staining on cell size (see Jerome et al. 1993).Strobilidium sp. strain JERC(Choreotrichida, Strobilidiidae)(Figs. 2.3, 2.4)Salient featuresCell subspherical with flat anterior and round posterior, 14-40 m long and 15-45 Amwide. External polykinetid zone (EPZ) and internal polylcinetid zone (IPZ) not completelyseparate. EPZ comprised of 22-23 polyldnetids of 35 Am long cilia, surrounding anterior end.Inner portion of each external polylcinetid ends with 4 ciliated lcinetosomes; the cilia are 3-4Am long and directed inward. Oral cavity acentricly placed within the circle of externalpolykinetids (EPK). 1-3 inner polykinetids (IPK) lie in oral cavity; 1-2 IPK are completelyseparate, the others are extensions of EPK. Ten (range, 6-12) somatic kineties, equally spacedaround cell, extend from 5-10 Am below oral region to near posterior pole. Each somatickinety composed of continuous row of cilia directed to the right when viewed from the aboralend. A rudimentary flap covers the base of the somatic cilia. Macronucleus, typically C-shaped but often fragmented, lies below EPZ, with arms of "C" near cytostome. Faintlystaining micronucleus positioned in a depression of macronucleus opposite the cytostome.11Time and locality of isolationLate April, from surface waters (top meter, temperatue of 10 °C, salinity 22 %.) atJericho Pier, English Bay, Vancouver, British Columbia, Canada, (123°10'W, 49°17'N).Discussion of speciesIn shape, strain JERC is superficially similar to four Strobilidium species: S.multinucleatum Lynn and Montagnes, 1988, S. sphaericum Lynn and Montagnes, 1988, S.spiralis (Leegaard, 1915) Lynn and Montagnes, 1988, and S. undinum Martin and Montagnes,1993. However, there are differences between strain JERC and all four of these species: 1) S.multinucleatum has 5 somatic kineties, 18-20 EPK and 11 spherical macronuclei; 2) S.sphaericum has no IPK, 24-30 EPK and many oral fibers that spiral into a central cytostome;3) S. spiralis has a characteristic asymmetrical distribution of somatic ldneties, 33-39 EPK and8-20 IPK; and S. undinum has 6 somatic ldneties, 21-24 EPK, and 4-6 IPK. In contrast, strainJERC has 10 to 12 symmetrically arranged somatic ldneties (although these stain poorly attimes and may appear as 6-9 asymmetrically placed lcineties); 22 EPK; 1-3 IPK; one C-shapedmacronucleus and an acentric cytostome not heavily supported by fibers. Further, strain JERCis 14-40 pm long and 15-45 gm wide which is smaller than S. sphearicum (40-70 gm long and40-60 gm wide) and S. spiralis (40-60 gm long and 40-52 gm wide) but in the same range asS. undinum (16-29 gm long and 15-23 gm wide). Considering the above differences, strainJERC likely represents a distinct and undescribed species.Remarks on culturing and behaviourStrain JERC was maintained in culture for 2-3 months on a mixture of Isochlysisgalbana, Chroomonas sauna, and Rhodomonas lens. Observations of this species in 10 mL ofculture medium in 20 mL tissue plates indicated that the ciliates typically remained in thewater column (i.e. they were planktonic). When disturbed either by motion of the container orwhen hitting a suspended particle, they very rapidly "jumped" approximately 1-3 cell lengthsand then reoriented themselves and continued swimming.122.2.2 Strombidiid ciliatesStrombidium capitatum (Leegaard, 1915) Montagnes et al., 1988 strain APAG(Oligotrichida, Strombidiidae)(Fig. 2.5)Discussion of speciesSeveral isolates of Strombidium capitatum have been used to reclescribe this species(Montagnes et al. 1988b); strain APAG fits the criteria for the species.Time and locality of isolationApril, from surface waters (temperature of 8 °C, salinity of 28 %.) at AgamemnonChannel, British Columbia, Canada, (124°5'W, 49°40'N).Strombidium acuminatum, (Leegaard, 1915) Kahl, 1932strain BJSC(Oligotrichida, Strombidiidae)(Figs. 2.6, 2.7)Salient featuresCell conical, 80-138 Am long and 13-31 kan wide. The posterior tapers to a fine pointthat is curved in fixed material. Anterior polykinetid zone (APZ) and ventral polylcinetid zone(VPZ) not distinctly separate. The inner 2-3 polyldnetids of the VPZ are continuous orcontiguous with the first 2-3 of the APZ. Anterior polykinetid zone comprised of 20 (range,18-26) polylcinetids, surrounding anterior end. Ventral polykinetid zone comprised of 7 (6-11)polykinetids which lie in a shallow oral groove. One ciliated paroral lcinety on right side oforal groove. The girdle begins just below the oral groove, spirals dexterally (when viewedfrom anterior) to dorsal surface, continues around to ventral surface where it anglesposteriorly, spirals to the dorsal surface and extends to -10 Am from posterior. The "ventral"lcinety, situated on the dorsal surface, begins 10-20 Am from the posterior, runs adjacent to,and on the right (dorsal view) of, the descending girdle for 5-10 /Am and then continues to thecell posterior. Descending portion of the girdle and ventral lcinety possess 2-3 /Am long cilia13spaced 2-3 I,lm apart. One ovoid, anteriorly positioned, macronucleus, 17 (13-24) Am long, 6(4-8) Am wide. Lightly staining "trichites" insert along and perpendicular to girdle and extendinternally -10 Am into cell. Dark staining (tear shaped) extrusomes lie below girdle.Extruded, moniliform, extrusomes often intertwined with cilia extend 10-15 Am.Time and locality of isolationJuly, from Sechelt Inlet, British Columbia, Canada, 1 m depth, temperature of 20 °C,and salinity of 20 %.. (123°45' W, 49°40'N).Discussion of speciesThere are four species of Strombidium which, like strain BJLSC, are elongate, conicaland -100 Am long: S. cornucopiae Wailes, 1929; S. pulchrum Leegaard, 1915; S. rhyticollareCorliss and Snyder, 1986; and S. acuminatum Leegaard, 1915.Strombidium corrtucopiae is larger than the typical range of stain BJLSC (97-200 vs.80-138 Am) and has prominent striations which strain BJLSC lacks. Strombidium pulchrurnhas a spiraling ventral lcinety like strain BJLSC, but it is slightly larger (167 Am long) and hasa more extensive zone of oral polykinetids than strain BJLSC. Strombidium rhyticollare issimilar in size to strain BJLSC but differs from it in three ways: S. rhyticollare has 32 oralpolyldnetids while strain BJLSC has 27 (20+7); S. rhyticollare has a groove around itsanterior end (posterior to the oral ciliature) which strain BJLSC lacks; and S. rhyticollare has apair of ventral ldneties that run one-third the cell length (from the posterior) while the ventralkinety and the extension of the girdle of strain BJLSC run together for one- quarter or less ofthe cell.Of the above species, strain BJLSC is most similar to S. acuminatum, although S.acuminatum is smaller (60-88 Am). Leegaard (1915) illustrated three views of S. acuminatum,two of which (her Fig. 12 a and b) are similar to strain BJLSC while the third (Fig. 12c) isnot. Figures 12a and b (Leegaard 1915) depict a ciliate with indications of a girdle similar tothat of strain BJLSC and a similar number of oral polylcinetids. Further, the nuclear shape andposition of S. acuminatum are similar to that of strain BJLSC.14Any of the older descriptions by Wailes or Leegaard could be used to describe strainBJLSC, but the description of S. acuminatum appears most fitting. Further, Wailes (1943)observed S. acuminatum in British Columbian waters. Therefore, I consider strain BJLSC tobe Strombidium acuminatum.Remarks on culturing and behaviourObservations of this species, in 10 mL of culture medium and the prey Thalassiosirapseudonana in 20 mL tissue plates, indicated that the ciliate is not strictly planktonic: at T.pseudonana concentrations ranging from 0 to 5x104 cells mL-1, -15% of the ciliates were inthe water column while the others swam along the surfaces. After maintaining strain BJSC forprolonged periods in culture (>15 days) the prey, T. pseudonana, became clumped. Whenthis occurred, ciliates were typically associated with these clumps. This species may bepseudo-planktonic and associated with suspended particles (marine snow).2.2.3 Strombidinopsid ciliatesStrombidinopsis acuminatum Faure-Fremiet, 1924 strain SPJSC(Choreotrichida, Strombidinopsidae)(Figs. 2.8, 2.9)Salient featuresCell round or cylindrical with tapering posterior, 35-82 Am long and 29-49 gm wide.External polyldnetid zone and internal polykinetid zone not completely separate. Externalpolykinetid zone comprised of 15-16 polylcinetids with -25 Am long cilia, surrounding anteriorend. Deep oral cavity acentricly placed within the circle of external polykinetids. Internalpolyldnetid zone comprised of 6-10 polylcinetids which lie in oral cavity; 4-5 internalpolykinetids are completely separate; the rest are extensions of external polyldnetids. Somaticlcineties, 13-19, equally spaced around cell, composed of dikinetids (both with 3-5 Am cilia),extend from oral region to posterior pole. Macronuclei, typically 2 (range, 1-6) semi-spherical(10 Am in diameter).15Time and locality of isolationJune from 1 m depth, temperature of 20 °C, and salinity of 20 %. in Sechelt Inlet,British Columbia, Canada. (123°45' W, 49°40' N).Discussion of speciesLynn et al. (1991) have rediagnosed the Strombidinopsidae as: free swimming aloricateciliates with many (usually >10) ciliated somatic lcineties, extending the entire length of thecell, composed of dilcinetids. They also typically have two similarly-shaped spherical to ovoidmacronuclei; the genus Strombidinopsis has characters of the family. Strain SPJSC possessthese characters and consequently has been placed in the genus Strombidinopsis.Strain SPJSC is similar to two previously described species: Strombidinopsis spiniferum(L,eegaard, 1915) Lynn et al., 1991 and Strombidinopsis acuminatum Faure-Fremiet, 1924.All three species exhibit characters with ranges that overlap; these are size, oral cilia length,number of external polylcinetids, number of somatic kineties, number of kinetids in somatickineties and macronuclear number. However, the one character that distinguishes strainSPJSC from the two described species is inner polykinetid number; the new isolate has 4-5independent polykinetids while the other two species have only 3.Strombidinopsis spiniferum has a thin tapering posterior which strain SPJCH lacks.The differences in shape between strain SPJSC and Strombidinopsis acuminatum are marginalsince Lynn et al. (1991) indicate that the latter species may be long and thin or short and fat;the latter shape is similar to strain SPJSC.It is possible that strain SPJSC is a new species since it has one more IPK thanStrombidinopsis acuminatum. Rather than diagnosing a new species based on a single innerpolylcinetid, I have been conservative and identified strain SPJSC as Strombidinopsisacuminatum.Remarks on culturing and behaviourStrombidinopsis SPJSC was initially cultured on a mixture of prey: ChroomonasIsochlysis galbana, and Rhodomonas lens. It grew well (-1 division c1-1 at concentrations>104 cells mL-1) on Chroomonas sauna, or Rhodomonas lens as monocultures. It was later16cultured on Thalassiosira pseudonana on which it also grew at a maximum rate (see Chapter3).Observations of this species in 10 mL of culture medium in 20 mL tissue platesindicated that the ciliates typically remained in the water column (i.e. it was planktonic).However some ciliates "grazed" along the surface. The ciliate typically swam in a straight linewith a helical movement, analogous to that described for a species of Strombidium (Fencheland Jonsson 1988). Fenchel and Jonsson (1988) however, attributed this motion to theasymmetric positioning of oral polylcinetids, and such an arraignment does not exist forStrombidinopsis; possibly another mechanism controls its swimming behaviour.PO EPk of EPZ K1^K2^K3MICS17MAC K5 K4 EX50 pmFig. 2.1. Schematic diagram of Srrobilidium spiralis (Leegaard, 1915) Lynn and Montagnes, 1988 strain IA fromprotargol stains and augmented by electron microscope observations. Paroral cilia, PO; external polylcinetid ofexternal polykinetidal zone, EPk of EPZ; internal polykinetid of internal polylcinetidal zone, IPk of IPZ; kineties1-5, K1-K5; macronucleus, MAC; micronuclei, MICS; extrusomes (only shown for a small portion of the cell),EX.18Fig. 2.2. Electron micrographs of Strobilidiurn spiralis (Leegaard, 1915) Lynn and Montagnes, 1988 strain IA.a, anterior view; b, detail of oral region with cytostome (C) and paroral cilia (PO); c, posterior view showingsomatic lcineties (K1-K5); d, lateral view; e, lateral view showing the curved second somatic lcinety (K2).EPk of !Pk ofEPZ^IPZFig. 2.3. Schematic diagram of Strobilidiurn sp. strain JERC from protargol stains and augmented by electronmicroscope observations. Four ciliated lcinetosomes internal to each external polykinetid, 4CK; externalpolykinetid of external polykinetidal zone, EPIC of EPZ; internal polylcinetid of internal polykinetidal zone, IPk ofIPZ; one of 10 somatic Icineties, SK; macronucleus, MAC; micronucleus MIC.192 0Fig. 2.4. Electron micrographs of Strobilidium sp. strain JERC. a, lateral view; b, detail of a somatic kinetyshowing somatic cilia (arrow); c, posterior view showing position of somatic lcineties; d, lateral and anteriorviews of two ciliates.21PCAPk of APZPOVPk of VPZVKMACCS50 pmFig. 2.5. Schematic diagram of Strombidium capitatum (Leegaard, 1915) Montagnes et al., 1988 modified fromMontagnes et al. (1988b). Peristomial collar, PC; anterior polylcinetid of anterior polykinetidal zone, APk ofAPZ; ventral polyldnetid of ventral polykinetidal zone, VPk of VPZ; paroral cilia, PO; girdle cilia, G; trichites,T; ventral kinety, VK; macronucleus, MAC; distended cell surface, CS.OEXAPk ofAPZAfirAt•:::•POVPk ofVPZMACCSVK50 pm22Fig. 2.6. Schematic diagram of Strombidium acuminatum, (Leegaard, 1915) Kahl, 1932 strain BJSC fromprotargol stains and augmented by electron microscope observations. Anterior polykinetid of anteriorpolykinetidal zone, APk of APZ; ventral polykinetid of ventral polykinetidal zone, VPk of VPZ; paroral cilia,PO; girdle, G; ventral kinety, VK; cilia of ventral kinety, C of VK; macronucleus, MAC; distended cell surface,CS; oral trichites (discharged), OEX; somatic trichites, SEX.2 3Fig. 2.7. Electron micrographs of Strombidium acuminatum, (Le-egaard, 1915) Kahl, 1932 strain BJSC. a,anterio-ventral view showing oral ciliature and girdle (G); b, ventral view showing ventral and oral ciliature andgirdle (G) algal prey, Thalassiosira pseudonana (TP); c, lateral view showing inclined orientation of oralciliature; d, vento-lateral view showing curved posterior and cilia of ventral kinety (arrow).EPk ofEPZIPk ofIPZDKSKMACt.i50 pm24Fig. 2.8. Schematic diagram of Strombidinopsis acuminatum Faure-Fremiet, 1924 strain SPJSC from protargolstains and augmented by electron microscope observations. External polykinetid of external polykinetidal zone,EPk of EPZ; internal polykinetid of internal polykinetidal zone, IPk of IPZ; a somatic kinety, SK; macronucleus,MAC; a somatic dikinetid DK.2 5Fig. 2.9. Electron micrographs of Strombidinopsis acuminatum Faure-Fremiet, 1924 strain SPJSC. a, detail ofsomatic dikinetids with cilia (arrow); b, anterior view; c, lateral view.9.5etv2 6CHAPTER 3GROWTH AND GRAZING RATES OF Strombidinopsis acuminatum STRAIN SPJSC ASA FUNCTION OF FOOD CONCENTRATIONIntroductionIn this chapter I examine Strombidinopsis acuminatum strain SPJSC, the first ciliatethat I cultured for an extended period. This ciliate became the "test" organism for a number ofmethods that were used for ciliates in later chapters. These methods are presented in detailhere and only referred to later.Strain SPJSC was unusual in a number of ways. Most remarkably, it was kept inculture for almost a year while the other ciliates I studied were maintained only for weeks tomonths. Thus, a detailed study of SPJSC was possible, but problems arose. In the process ofdetermining growth and grazing response curves, it became apparent that sub-clones haddifferent growth and grazing responses. Several experiments were devised to determine theeffect and nature of these differences. Consequently, the studies on this ciliate becamediversified beyond the original focus on growth and grazing rates.PART I GROWTH RATES: Estimating the Numerical Response1.0 General MethodsA series of experiments was conducted to determine the numerical response of strainSPJSC grazing on the diatom Thalassiosira pseudonana at food concentrations ranging from 0to 105 cells mL-1. The ciliates and prey were maintained in a semi-continuous culture bytransferring the ciliates to new prey and medium on a daily basis. The details of this methodfollow.Prey cultures of T. pseudonana were maintained at a constant nutritional quality (seeHarrison et al. 1990), in a 500 mL glass turbidostat at 2-5x105 cells mL-1 at 16-17°C withcontinuous irradiance of 100 Amol photons m-2 s-1. Prior to the experiments, strain SPJSCwas maintained in enriched natural seawater with T. pseudonana (medium modified from thatof Harrison et al. 1980, see Appendix 1) in 150 mL glass flasks at 16-17°C on a 14:10 hlight:dark cycle at 20 pmol photons m-2 s-1. Ciliates were collected during log or stationary2 7phase; prey concentrations, which the ciliates were collected from, were not controlled andranged from 103-105 prey mL-1.During the experiments, ciliates were grown in the dark (to prevent mixotrophicautotrophy) in 6-well, 20 mL plastic tissue culture plates containing 10 mL of prey andmedium. Prey concentrations were made by diluting prey and medium, collected from theturbidostat, with unenriched, pasteurized natural seawater (heated at 80°C for >12 h andfiltered with Whatman GF/A glass fiber filters, 1.6 Am retention). Prey concentrations weremeasured using a Coulter Counter (Model TAII). Typically, each concentration was replicatedthree times.For any one prey concentration (treatment), the semi-continuous culturing method wasas follows: ciliates were collected from stock cultures and 10 ciliates were placed in each of 3wells in a 6-well tissue plate (thus, n=3), using a finely drawn Pasteur pipette (the 10 mLvolume of water was altered by << 0.01% by this procedure). The tissue plate was thenplaced in the dark for 24 h at 16-17°C. After 24 h, the prey concentration was remade (asdescribed above) and allocated to 3 wells in a new tissue plate. Then, from an old well, 10randomly chosen ciliates were removed and transferred to a new well. If more than 10 ciliateswere in an old well (i.e. population growth was positive), the excess was counted andremoved. If less than 10 ciliates were in a well (i.e. population growth was negative), all theciliates were transferred. It was thus possible to determine a daily change in ciliate numbersand establish a growth rate at a defined prey density. Growth rate (A) was calculated as p. =(in cellst+i/ In cellst), where t = time in days (i.e. exponential growth was assumed over 24h).The numerical response data were fitted to a modified Michaelis-Menten model (Eq.1). This model was chosen since it is a good predictor of both functional and numericalresponses and is based on theoretically sound mechanisms (see Holling 1959, Spain 1982,Fenchel 1986, Taniguchi and Takeda 1988, Appendix 2).p. = {Amax * ([13] - x1)}/ {k + ([P] - x')}^(1)where,28= growth rate (d-1) = (in cellst+1/ in cellst), t = 24 h!max = the maximum growth rate (d-1)[11 = prey concentration (# mL-1)x' = the x intercept (the prey concentration where g = 0) (# mL-1)k = the half saturation constant (i.e. [11 at 0.5 gmax) (# mL-1)Curves were fitted to the data using the Marquardt-Levenberg algorithm (Sigmaplot,Jandel Scientific, CA.). This is an iterative fit which minimizes the sum of squares ofdifferences between the dependent variables in the equation and the observed data. At veryhigh prey concentrations growth rate of the ciliate was often negative. Curves were not fittedto data above 7x104 prey mL-1.The methodology, described above, was followed for three separate sets ofexperiments. As experiments were performed, modifications were made in the procedure todetermine the cause of unanticipated variation in the growth response. The methods, resultsand discussion for each of the three sets of experiments are outlined in three sections below.1.1 First Set of Experiments1.1.1 MethodsA total of 25 treatments (prey concentrations) were examined in 5 experiments, since itwas not logistically possible to run 25 treatments simultaneously. Treatments ranged from2x103-105 prey rnL4. The 5 experiments were similar in their execution, but they were rundays to weeks apart, on different "batches" of prey and on different "batches" of ciliates(ciliates from the same original clone but different sub-clones). At the time, it was assumedthat combining these experiments would not bias the results.Two tests of methodology were made: 1) One of the experiments was run for 9 dayswhile the others were run for 5 days. The 9-day experiment was used to determine thevariation of growth rate over the 9 days. 2) One of the 5-day experiments was used todetermine the stability of the prey concentrations; the prey concentrations were measured priorto adding the ciliates and after the ciliates had been removed (24 h later). Otherwise, the2 9experimental procedure of the 5 experiments followed the semi-continuous culturemethodology described in section 1.0.1.1.2 Results and Discussion1.1.2.1 Consistency of Growth RateFor all prey concentrations, of the 9-day experiment, there was an increase in growthrate of strain SPJSC over the first 1-2 days. This was followed by a relatively constant growthrate between days 3-5. After day 5, in several treatments, growth rate decreased (Fig. 3.1a-i).Typically, a replicate (10 cells) which began to decline in growth rate continued to do so, butother replicates at the same treatment concentration often maintained a constant rate (e.g. Fig.3.1e).Thus, I decided that growth rate did not stabilize until after day 2, and that after day 5there was a potential bias, possibly caused by the experimental procedure or by stochasticextinction (see sections 3.2 and 4.1). Growth rates reported for all experiments use the meangrowth rate from days 3-5 exclusively.1.1.2.2 Consistency of Prey ConcentrationData from one experiment were used to assess the day-to-day precision of maintainingprey stocks at a predetermined concentration. The T. pseudonana concentration typicallyincreased over the 24 h dark period. Higher concentrations remained relatively constant butlower concentrations (<103 cells mL-1) were, at times, 50% higher than the desired treatmentlevel (Fig. 3.2). It is likely that "background noise" added to the imprecision of lower counts.Prey concentrations, used for the numerical response, were the average concentrationsdetermined from the tissue plate wells over days 3-5.1.1.2.3 Numerical ResponseThe growth rate followed a rectangular hyperbolic response from 2x103 to 7x104 preymL-1 (Fig. 3.3a). At concentrations above 7x104 prey mL -1 , the response was lesspredictable: some replicates lay near the asymptote predicted by the hyperbolic function, butothers lay well below it. The variance of the treatment response increased as prey3 0concentration increased. At prey concentrations below 3x103 mL-1, ciliate net mortality(henceforth referred to as mortality) occurred, presumably due to starvation.Equation 1 was fitted to the data between 2x103-7x104 prey mL-1 (Fig. 3.3a). Theparameters and estimates of error of the equation are presented in Table 3.1. The data werenot evenly scattered around the predicted curve. Variation between treatments existed and wasrelated to the different experimental runs (i.e. over a range of similar prey concentrations,experimental runs yielded different values from each other). I speculated that this was due tochanging experimental conditions (e.g. food quality); this was examined next.1.2 Second Set of Experiments1.2.1 MethodsIn the first set of experiments there was variation in growth response within andbetween treatments (see section 1.1.2.3). I hypothesized that the variation between treatmentswas due to combining the 5 separate experiments. To test this, a single experiment was runwith 20 treatments using only one "batch" of prey and one "batch" of ciliates. If inter-treatment variation was due to combining experiments, this procedure would reduce theamount of variation. Because the growth response was asymptotic by 4x104 prey mL-1 (Fig.3.3a), this experiment was conducted over a smaller prey range than the previous experiment:from 0-4x104 mL-1.1.2.2 Results and DiscussionThe growth rate followed a rectangular hyperbolic response between 0-3.7x104 preymL-1 (Fig. 3.3b). There was a decrease in the variance of the treatment response as the preyconcentration increased (Fig. 3.3b). At concentrations below 103, prey mL-1 mortalityoccurred, presumably due to starvation. Equation 1 was fitted to all the data (Fig. 3.3b). At7-8x102 prey mL-1 growth rate was 0. The parameters and estimates of error for the curvefitted to this data are presented in Table 3.1.There was less variation between treatments than in the previous experiment (section1.1) but variation existed within treatments (Fig. 3.3b). This suggested that the variation seenin the previous experiment was not caused by changes in culturing but was due to some31intrinsic factor of the ciliate culture. I speculated that the variation was due to different ciliatestrains within the harvested "batch" culture. This prediction was supported by what appearedto be three distinct numerical response curves formed by the data (Fig. 3.3b); these may haverepresented three genotypes (or phenotypes) which expressed different growth rates. The nextexperiment examined the differences in growth rate between cell lines.1.3 Third Set of Experiments1.3.1 MethodsThe modifications made in the second set of experiments did not reduce the variationwithin treatments. Thus, a third set of experiments was performed to test if the variation wasdue to differences in ciliate cell lines (see section 2.0 and 3.2 for an explanation of howdifferent genetic lines could occur).A single ciliate was placed in each of 24 tissue plate wells containing 10 mL of mediumwith saturating prey (104 mL-1) and allowed to grow (at -1 division d-1, this providedsufficient numbers of progeny, from a single cell, to perform experiments within 8-10 days).In the first experiment of this type, two clones were grown in 6 treatments rangingbetween 4x102-1.8x104 prey mL-1. Some ciliates in one of these clones were conjugating onday 10 of the isolation period. Only non-conjugating individuals were used from this culture,but these may have been exconjugants.In the second run of this experiment, the progeny of one of the isolated cells produced>360 cells within 8 days (sufficient numbers to perform 6 replicates at 6 concentrations).Note: an increase from 1 to 360 cells indicated that this clone had an average growth rate (A)of 0.74 d4 for the first 8 days. For this run, the 6 replicate clones were grown in 6treatments ranging between 3.7x102-1.9x104 prey mu.1.3.2 Results and DiscussionOnly 2 of 24 cells isolated for the first run of this experiment and 1 of 24 cells isolatedfor the second run survived and divided over the 8-10 day isolation period. This suggests asurvival rate of only 4-8% for ciliates in the culture flasks. Since these experiments were run3 2well into the first year of the life of this culture it may be that these low survival rates weredue to ageing of the culture (see section 3.0).Of the three cell lines examined, two exhibited positive growth at some of the preyconcentrations. The other exhibited negative growth at all treatment levels. In the cell linethat exhibited maximum growth (open circles, Fig. 3.3c) -20% of cells were conjugating whenthe cells were harvested. This alone is remarkable since it indicated that strain SPJSC wascapable of selfing conjugation.Equation 1 was fitted to all the data in the two data sets that exhibited positive growth(open circles and triangles, Fig. 3.3c, Table 3.1). For both growth curves, the growth ratefollowed a rectangular hyperbolic response between 4x102-2x104 prey mL-1. For the fastgrowing clone (open circles, Fig. 3.3c), the variance within replicates was considerably lowerthan that observed in previous experiments. Below 1.8x103 prey mL-1 mortality occurred,presumably due to starvation. For the slower growing clone (triangles, Fig. 3.3c), thevariance within replicates was greater than that of the fast growing clone, but lower than thatobserved in previous experiments. For this clone, mortality occurred below 1x104 preymL-1.The clone that was conjugating prior to the experiment produced cells that grew as fastas the maximum rates determined from previous experiments and the three replicates showed ahigher level of precision than replicates from the other clones (cf. Figs. 3.3a, b and c).Assuming that the isolated cells were exconjugants, these data suggest there was a geneticinfluence on growth rate.During clonal growth there are a number of mechanisms which reduce the quality ofthe macronucleus and reduce clonal vitality (see Smith-Sonneborn 1981, Bell 1988 and section3.0 below). Conjugation in ciliates results in the generation of a new macronucleus from themicronucleus. Since the macronucleus mediates cell growth, and thus fission rate, it followsthat conjugation, which "rejuvenates" the macronucleus should both increase the fission rateand decrease the variability in fission rates in a culture; this is precisely what was observed.3 3These data suggest some of the variation presented in sections 1.1 and 1.2 was due to geneticvariation in clonal lines.The data from this experiment also suggest that over short incubation periods (5-10days) experimental manipulation was not the cause of high variation of growth rates within asingle treatment. Instead, intrinsic features, likely genetic infirmity (loaded with mutations),caused the variation. If reduced growth rates were due to genetic infirmity (caused bymaintaining small populations in culture), then the maximum growth rates (observed inprevious experiments) may have been representative of natural populations (assuming thatnatural populations regularly conjugate).These findings are important in analyzing the data of the previous experiments sincethey provide criteria for "editing" the data, prior to constructing a population model (seeChapters 8 and 9). If natural populations grow at near maximal rates then, much of the dataobtained in previous experiments may be ignored (however, see section 3.0).2.0 Conjugation of Strain SPJSCOccasionally, during routine culturing of strain SPJSC, conjugating pairs wereobserved. Since all the sub-cultures were the descendents of a single cell, this strain must havebeen capable of selfing conjugation. Routinely, when conjugating pairs were observed incultures, they were isolated. All attempts to maintain and isolate exconjugants were fruitless;all cells died while joined, or possibly shortly after separating.Why did all these cells die? Possibly the number of heterozygotic mutations in theciliate was sufficient to result in mortality of most exconjugants. However, this type ofmortality typically occurs after several post-conjugation divisions due to phenotypic(cytoplasmically or genetically induced) lag (Berger 1976). Alternatively, it may be that theobserved cells were genetically infirm or that culturing conditions were poor, and the processof conjugation caused mortality.3 43.0 An Estimate of the Inheritance of Growth RateGrowth rate varied between different sub-clones raised for -10 generations, and thesedisparate growth rates may have been caused by genetic differences. If this were so, then theprogeny of a fast growing clone should also be fast growing. Further, if a fast growing clonewas "genetically healthy" (i.e. lacked mutations) then, fast growth should be constant for many(>50) generations (Nanny 1980, Bell 1988, and discussion below). The following experimentwas conducted to test this hypothesis.3.1 MethodsOne hundred and two cells of strain SPJSC were isolated from stock cultures andplaced in separate 20 mL wells containing 10 mL of culture medium (as described in section1.0 above) and saturating levels of T. pseudonana (3x104 mL-1). The ciliates were left inthese containers at 16-17°C on a 14:10 light:dark cycle at 10 pmol photons m-2 5-1 for 9days. On day 4 and day 9 the number of cells in each well was estimated.After 9 days only 17 of the 102 wells had ciliates in them (i.e. -17%). Three of theseclones were used to repeat the experiment: from each of the three clones 34 cells (a total of102) were isolated and maintained for 10 days under the conditions described above. Again,on day 4 and day 10 the cell number in each well was estimated. Finally, after 10 days only 1of the 102 wells had ciliates in it (i.e. <1%). Sixty-six cells were isolated from this sub-clone, maintained for 9 days under the conditions described above, and on day 4 and day 9 thenumber of cells in each well was determined.3.2 Results and DiscussionIf growth rate was strictly an inheritable trait for stain SPJSC, then the progeny of thefast growing cells would also be fast growing, but this was not so. The mean and maximumgrowth rate of sub-clones decreased over 20 generations (Fig. 3.4, Table 3.2). This was alsoobserved in section 1.3 where the growth rate of one clone decreased from 0.74 to <0.5 d-1over 10-15 days. The fast growing cells typically represent 1-15% of the population (Table3.2). This agrees with the 4-8% survival rate observed in section 1.3.2. Thus, when these3 5experiments were performed, culture growth may have been by a small percentage of thepopulation surviving and growing relatively rapidly.The reductions in growth rate may have been the result of experimental manipulation(external sources) but also could be due to intrinsic (genetic) factors. Both empirical andtheoretical evidence suggests that if growth rate was inheritable then the progeny of a fastgrowing clone should also be fast growing and this trait should be constant for >100generations after conjugation.Bell (1988) has reviewed the literature and emphasized what others have shown:typically, isolated cells will maintain a constant growth rate for >100 generations, but by-200 generations after conjugation, most ciliate clones are extinct. However, the older theparent at the time of conjugation, the shorter the life span of the progeny line (Smith-Sonneborn 1981). Thus, in a population, a series of delays in the onset of conjugation wouldreduce the number of viable generations, post conjugation (i.e. clonal lines would maintain aconstant growth rate for <<200 generations).There are 5 reasons for the decrease in vitality in small populations: 1) randomassortment in the macronucleus causes a dilution and/or change of frequency of functionallyhaploid genetic fragments (i.e. heterozygosity verges to homozygosity or a loss of a gene); 2)repeated automixis or setfing conjugation fractions a heterozygotic population intohomozygotic lines; 3) recessive mutations in the micro- and macronuclei become expressedthrough random assortment in the macronucleus and/or repeated automixis and seWng (1 and2 above); 4) Muller' s ratchet (the accumulation of deleterious mutations) works on the mico-and macronuclei in small populations; and 5) probability of extinction from stochasticprocesses occurs in small cultures (i.e. for a finite population growing exponentially there isalways a probability that the population will go extinct; this probability increases exponentiallywith decreasing population size).These processes increase the likelihood of: 1) expression of lethal or sublethal recessivegenes; 2) homozygosity, which may reduce hybrid vigor; 3) the accumulation of mutations;and 4) dilution out of healthy stock from cultures. All of these would increase the likelihood3 6of a culture becoming extinct and help to explain why so many ciliates do not survive well inculture (see Nanney 1980, Smith-Sonneborn 1981, and Bell 1988 for reviews of theseconcepts).The biological solution to these degenerative processes is sex. Conjugation in ciliatesgenerally replaces the macronucleus which controls somatic growth (Lynn and Corliss 1991).Thus, the progeny after conjugation will grow faster than their parents. However, if theparental lines are old, the number of transmittable mutations increases and the progeny willstill have a retarded growth rate. This may be what was happening in the cultures of strainSPJSC.Strain SPJSC was maintained in culture for >250 days. If the ciliate grew near themaximum rate in stock cultures (1-2 division d4), then it exceeded the 200 generations typicalof viable lab cultures. Since conjugation occurred in cultures, the life of the culture may havebeen prolonged. However, conjugation may not have entirely rejuvenated the culture. Whenthe experiment described in this section was conducted, strain SPJSC was near the end of itslife; a month later it was dead. Possibly, the reason that 1-5% of the population grew rapidlywhile the rest died was that, a few cells underwent undetected conjugation which stimulatedrapid growth for several generations.Since the results of these experiments were conducted on "old" cultures it would beimprudent to directly apply them to natural populations. However, the data can be used tointerpret variation in earlier numerical response experiments. Previous results suggested thatgrowth rate was inheritable: 1) groups of 10 ciliates maintained a constant growth rate for 5-8generations (Fig. 3.1); 2) some isolated cells grew to form large numbers (up to 360) in 8 dayswhile others died off; and 3) cells that were likely post conjugates grew at maximum rates.The data from this section do not support the notion that growth rate was strictly inheritableunless I invoke the notion that undetected conjugation occurred. Possibly, both inheritable andnon-inheritable factors affected the growth rate.3 74.0 Sources of Bias and Explanations of Methods4.1 MortalityEquation 1 was fitted to all prey concentrations, including lower prey concentrationswhere mortality occurred. However, this procedure may be incorrect since mortality rate infood deprived cultures may not be constant (Jackson and Berger 1985). In the semi-continuousculturing method a healthy, non-starving, population should (and typically did, Fig. 3.1)maintain a constant growth rate over the experimental period. This is an assumption for theexponential model. However, mortality rate was likely not constant over the same period.Two reasons this may be so are: 1) If the ciliates were dying due to starvation (low food) orthe accumulation of "toxic" material (possibly the cause of mortality at high food conditions),mortality would be time dependent. Since, for negative growth rates, the same cells wereobserved each day, cumulative effects could change the mortality rate; thus, it would not beconstant. 2) The negative growth rate would not be constant for strictly stochastic reasons.Since ciliates were not added to wells where numbers dropped below 10 (see section 1.0), thelikelihood of the population becoming extinct would increase as the cell number decreased (i.e.the probability (P) of a population of n cells becoming extinct is: P(0,n) = 1, if b^d, andP(0,n) = (d/b)'1, if b > d, where d = death rate and b = birth rate, Bell 1988).With these biases in mind, I have still used the exponential growth model to estimatemortality rates and have used the data range where the ciliates were likely starving to establishnumerical response curves; this is similar to methods adopted by others (e.g. Jackson andBerger 1984, Turley et al. 1986). The negative growth rates, established in the abovesections, may be interpreted as 3-day averages. When modeling the response to food for thisciliate and others (in Chapter 8 and 9), mortality rate was estimated with these biases in mind;both biases overestimate mortality rate.4.2 Handling of CiliatesDaily transfer of ciliates with finely drawn pipettes may have delayed division andbiased the growth rate estimates. Others have found similar transfers to cause 10 min delaysin fission rate (e.g. Adl and Berger 1991, for Paramecium tetraurelia). Such delays in fission3 8would cause a small (<1%) reduction in generation time, assuming that similar effectsoccurred for strain SPJSC, and they were ignored in the analysis.4.3 Food QualityFor the growth experiments on strain SPJSC, all the prey were grown in a turbidostat.This was done to provide food of a constant biochemical quality, thus removing one source ofpotential variation. In retrospect, there was considerable variation due to the ciliate and notthe food. In subsequent work, on other ciliates, I did not use a turbidostat to grow the foodbut kept the food in exponential growth phase by serial dilutions of batch cultures.5.0 Aspects of the Growth Rates and their Ecological RelevanceAlthough there was variation in the data, both within and between treatments, 4 piecesof information can be procured from the numerical response data: 1) The thresholdconcentration of T. pseudonana for strain SPJSC (i.e. the prey concentration below which thecilates cannot survive) was -103 mL-1; 2) the major increase in growth rate occurred between103-104 prey mL-1; 3) the maximum growth rate was -1.4 divisions d-1 (Fig. 3.3c, Fig.3.5a, broken line); and 4) strain SPJSC was capable of selfing.During the summer in the Sechelt fjordal system (Appendix 3), where strain SPJSCwas collected, average numbers of phytoplankton of similar size to T. pseudonana (5-10 Am)ranged from 200-44000 cells mL-1 (Appendix 3) and more relevantly, small diatoms (4-5 Am)ranged from 0-2000 cells mL-1 (data not shown). This means that on average natural levels offood could be sufficient to support strain SPJSC, and for short periods this ciliate could begrowing at its maximum rate, potentially forming blooms.The occurrence of selfing in strain SPJSC indirectly supports the notion that it formsmonoclonal blooms. The following scenario may occur if: 1) selfing conjugation is a traitacquired to allow conjugation when ciliates are unable to find other clones, and 2) starvationstimulates conjugation (Nanney 1980): a bloom develops from a single cell finding itself inoptimum-food conditions, the progeny of this ciliate then deplete the food and starvation3 9ensues. This in turn stimulates conjugation. However, the bloom would be monoclonal andselfing would result.This scenario requires that both ciliate abundance and prey patch size are such that onlyone mating type of a species will find a prey bloom. In Indian Arm (British Columbia), in thelate winter, Strombidinopsis abundance was rare (Martin and Montagnes 1993). It was foundat only 1 of 6 sites that were 1-5 km apart. Although this does not indicate Strombidinopsisforms monoclonal blooms, it does suggest a patchy distribution or that this species is rare inthe winter.6.0 A Numerical Response ModelInitially, the goal of these studies was to establish a numerical response curve for strainSPJSC with which to model blooms. But, the expression and evaluation of variance in growthresponse retarded the development of this curve. The variance suggested that many of the lowgrowth rate data resulted from genetically unhealthy lines. Assuming that natural populationsgrow at near maximal rates, much of the lower growth rate data may be ignored.Accordingly, some editing was conducted.Equation 1 was fitted to a compilation of the data presented in sections 1.3-3.3, (Fig.3.5a, solid line). The residual sum of squares (calculated by the Sigmaplot curve fittingfunction) were plotted against prey concentration (Fig. 3.5b, all data). The residuals were notevenly distributed around zero. To adjust for this, points below - 0.4 were omitted (Fig. 3.5b,open circles); these points were assumed to represent genetically unhealthy lines. Equation 1was then fitted to the remaining data (Fig. 3.5a solid circles, broken line; 3.5b, solid circles);the residuals of this fit appeared (by eye) to be evenly distributed around zero (Fig. 3.5c),suggesting a better fit. The second curve (Fig. 3.5a, broken line) provided a higher numericalresponse with a more rapid response to food concentration than the fit to all the data. Thisupper response has been used in Chapters 8 and 9 to predict the growth response of a"genetically healthy" population of strain SPJSC. The parameters and estimates of error forthe solid and broken curves in Fig. 3.5 are presented in Table 3.1.4 0PART II GRAZING RATES: Estimating the Functional Response7.0 General MethodsA series of experiments (n = 16) was conducted to determine the effect of preyconcentration on the grazing rate of strain SPJSC. Fluorescently labeled, 5 Am, latex beads(Seradyn, Particle Tech. Div., Indianapolis IN, USA) were used to simulate the diatomThalassiosira pseudonana, on which the ciliate was cultured. The experiments were run at"food" concentrations ranging from 1.8x103 to 1.1x105 beads mL-1.The beads, which were suspended in a "toxic" surfactant, were rinsed with distilledwater using a 3 Am filter, suspended in a 5% bovine serum albumen-distilled water solutionfor >24 h (up to a month) to reduce clumping (Pace and Bailiff 1987) and refrigerated (5°C)for storage. Prior to use, the beads were rinsed with prefiltered seawater with a 3 AmMillipore filter and resuspended by sonication (50/60 Hz bath sonicator). Beads were alsosonicatal prior to counting with a Coulter Counter (Model TA II).Prior to the experiment, ciliates were maintained as described in section 1.0. Tostandardize the nutritional history of the ciliates prior to experimentation, ciliates were placedin filtered seawater to starve for 4-5 h. Then, 4-5x102 ciliates were transferred in a drop ofseawater (0.1 mL) to a 20 mL plastic tissue culture well containing 10 mL of treatmentsolution (enriched seawater containing 5 Am fluorescent beads at a known concentration). Thebead solution was previously sonicateti for 30 s to ensure mono-dispersion. The experimentswere run at 16-18°C.Every 5 min, for 30 min, after the ciliates were added to the bead solution (time 0) 30-60 ciliates were removed from the container, using a finely drawn pipette and fixed in 5%(volume/volume) Bouin's fluid on coverslips with petroleum jelly on the edges. These werethen placed on a microscope slide, the petroleum jelly sealing the edges. Some ciliates werelost during this manipulation; for each time interval, 20-40 ciliates were observed. Theciliates were examined using fluorescent microscopy (100-200 X; Zeiss filters blue excitation,BP 450-490, FT 510, LP 520). To estimate bead ingestion rate, the mean number of beads4 1ciliated was visually determined and plotted against time. The grazing rates were fitted to aMichaelis-Menten function (Eq. 2, see Appendix 2).G = (Gmax x [B])/ (k + [B])^ (2)where,G = grazing rate (beads ciliated hd)Gmax = the maximum grazing rate (beads ciliated hd)[B] = beads concentration (# mLd)k = the half saturation constant (i.e. [B] at 0.5Gmax)7.1 Results and DiscussionFor all treatments (bead concentrations) the mean number of beads ciliated increasedlinearly with time (Fig. 3.6a). Grazing rate ranged from 0.48 to 27 bead ciliated hd andincreased with bead concentration, but an asymptotic response, as predicted by Eq. 2, did notexist (Fig. 3.6b). If feeding and growth were linearly correlated, this response wouldasymptote after 2x104 prey mLd (see Fig. 3.5).Why did grazing not follow a hyperbolic response? Here are four possibilities: 1) Thenutritional history of the ciliates may have influenced their ingestion rate (i.e. the 4-5 hstarvation period was an inadequate method of standardizing nutritional history). The ciliatesused for the experiment at 7.5x104 beads mLd were isolated from a dense culture of prey(>105 prey mLd), and the points at this concentration appear to be anomalous (open circles,Fig. 3.6b). Further, the starvation period may have caused elevated grazing rates. 2) Thevariation in feeding may have been due to clonal variation (as seen in Part I). 3) Highervariation of feeding data occurred >7x104 beads mLd. Possibly there was grazing inhibitionat high concentrations; this parallels the growth response (Fig. 3.5). 4) The ciliates selectedagainst the beads. This last reason is the simplest and may be the most reasonable, as severalciliates did not eat beads (see Chapter 8).Since the 3-4 h starvation period and a diet of only beads may have biased grazingrates, I modified grazing experiments (on other ciliates, see following chapters) by 1) using a4 21:1 combination of live prey with fluorescent surrogates and 2) acclimating the ciliates to theprey concentrations used during the grazing experiments.7.2 Modeling the ResponseAlthough there was no asymptotic response in the grazing rate with food concentration,there was a good fit of the data to Eq. 2 (Fig. 3.6b, solid line). However, given that thepoints at 7.5x104 beads mL-1 (Fig. 3.6b open circles) may be poor estimates of grazing rate, Irefit the response without them (Fig. 3.6b, broken line). For the reasons discussed above, thedata may be biased, but I have still used the upper curve in Fig. 3.6b to compare grazing ratesin Chapter 8.CONCLUDING REMARKSThe genus Strombidinopsis is found in many coastal marine waters (Faure-Fremiet1924, Gifford 1985, Lynn eta!. 1991, Snyder and Ohman 1991, Putt 1991, Sime-Ngando etal. 1992). This genus has never been reported at high densities, suggesting that it rarely, ifever, forms blooms. However, the data from this study suggest that Strombidinopsis has thepotential to bloom over short periods, as it can grow rapidly when exposed to high preyconcentrations. The growth response of this species has been used in Chapters 8 and 9 tomodel one potential ciliate response to changing prey concentrations.Gifford (1985) cultured Strombidinopsis (cf. acurninatum) on several prey andmaintained it on a combination of Heterocapsa triquetra and Chroomonas sauna for almost ayear. Although her Strombidinopsis culture was not monoclonal, it was the only one of 21isolated oligotrichs that conjugated. Possibly these cultures were selfing or were at leastpredisposed to conjugating in culture. Thus, Strombidinopsis may be a useful organism toculture for future research on oligotrich mating.Gifford (1985) also supported Blackbourn's (1974) assertion that oligotrichs, andspecifically Strombidinopsis, do not grow well on Thalassiosira species, arguing that B-chitanthreads on the frustule render them inaccessible to ciliates. B-chitan threads also inhibited4 3growth of tintinnids fed T. pseudonana, but when stirred or shaken the diatoms lost theirthreads and supported ciliate growth (Verity and Villareal 1986).My cultures of T. pseudonana, which supported Strombidinopsis, were stirred and weretherefore likely "thread-free". This study may therefore represent an "unnatural" situation assome Thalassiosira species possess B-chitan threads in situ. However, the presence of threadsis variable in field samples (F. J. R. Taylor, pers. comm.), and field observations indicate thatciliates do eat small centric diatoms (unpublished data). Further, natural populations of ciliateswere attracted to T. pseudonana, suggesting that it may be a beneficial prey species (Verity1991b). Thalassiosira pseudonana could stimulate ciliate blooms, as it does bloom in somecoastal waters (Guillard and Ryther 1962). However, blooms of T. pseudonana have not beenreported in British Columbian waters. Even if T. pseudonana is not a typical food for ciliates,my data will be useful for modeling potential responses. Assuming that there is a relationbetween organic carbon and growth rate, these data can be used to estimate a general growthresponse of this ciliate (see Chapter 9).44Table 3.1. Growth and grazingerror of the numerical and funcchapter for the equations used.growth rate 02max).data for Strombidinopsis acuminatum strain SPJSC; parameters and estimates oftional response equations presented in Figs. 3.3-3.6. See section 1.0 in thisValues are in units of numbers ml.:1 for concentrations (k^), and c1-1 forvalue^standarddeviationMichaelis-Menten fit from Fig. 3.3acoefficientof variationAmax^0.9769^0.0765 7.840k 4993 1628 32.60x' 3026 426.6 14.10Michaelis-Menten fit from Fig. 3.3bAmax^0.731^0.071 9.664k 1674 505.5 30.20x' 7739 201.1 25.99Michaelis-Menten fit from open circles in Fig. 3.3cAmax^1.179^0.167 14.13k 4063 1208 29.73x' 1753 265.0 15.12Michaelis-Menten fit from triangles in Fig. 3.3cAmax^0.1992^0.1642 82.45k 1.349E4 4521 33.52x' 1.067E4 4266 39.99Michaelis-Menten fit from solid and open circles in Fig. 3.5aAmax 0.7286 0.07747 10.63k 4843 1255 25.92x' 2531 4515 17.84Michaelis-Menten fit from solid circles in Fig. 3.5aAmax^0.9845^0.04613 4.686k 4091 614.5 15.02x' 1567 157.4 10.04Michaelis-Menten fit from all circles in Fig. 3.6bGmax^28670^286900 10011.669E8 1.671E9 1001Michaelis-Menten fit from solid circles in Fig. 3.6bGmax^101^184.8 182.9426300 924600 216945Table 3.2. The increase in cell numbers and survival of cell lines of Strombidinopsis acuminatum strain SPJSC.Single cells were initially isolated in wells (day-0, d0); after day-4 (d4) and day 9 (d9) or 10 (d10), the total-number of cells well1 was determined. Data are presented as the percent of total wells containing a range ofciliates (#). The number of wells in each "RUN" were: RUN 1 = 102, RUN 2a-c = 34 (each), RUN 3 = 66.RUN 2a-c were inoculated from the progeny of three clone with >70 cells in RUN 1. RUN 3 was inoculatedfrom the progeny of the one clone with 60-69 cells in RUN 2a (see section 3.1 for details).# RUN 1dO^d4 d9RUN 2adO^d4RUN 2bd10 dO^d4RUN 2cd10 dO^d4 d10RUN 3dO^d4 d90 52 83 78 97 . 83 100 . 89 100 67 971 100 12 2 100 6 . 100 17 . 100 11 . 100 152-3 3 . 6 • • • • 124-5 7 1 3 • • • • • 3 26-7 6 3 . • • . 3 28-9 3 1 6 • • • • .10-11 5 1 . 3 • • .12-13 5 • • • • . • •14-15 2 1 • • • • . .16-17 2 . • • • • . • •18-19 1 . • • . • •20-24 1 • • • • • .25-29 1 . • • • • • . .30-34 1 1 • • • • • .35-39 . . • • . . .40-44 . • • . . .45-49 . 1 • • . . . .50-59 . . . .60-69 . 2 . 3 . . .70-79 . 3 • • • • • .80-89 . • • • • .90-100 . 1 • • • • • •210 -- 1 $-.* s^--2 -3.0 x 10 mL i4^—1 •-5.0 x 10 mL210- 1- 2210 •-•-•_2 9.0 x 10 mL—113ti210- 1- 21 2 3^4 5^6^7 8 9211 1 11 1 1crit:t-1-20 11111112 3 4^5^8^7 8 9Time (d)46Fig. 3.1. The variability of growth rate of Strombidinopsis acuminatum strain SPJSC at nine prey concentrations(cells mL4 as depicted in panels a-i), over a 9 day period (for 1 of the 5 experiments presented in section 1.1).Data represented by circles and connected by a broken line are the growth rate of each replicate (n = 3). Thesolid lines represent the average growth rate.475.04.8.........,,,N-14.6I3.43.21 42^3Time (d)i-4.4;--1(1)rCD 4.24.0r—Ieml0 3.8C.)W) 3.60rl4.003.903.783.603.305C/14.9004.78 ''''—1—)cri—1—)0CDC.)04.30 0C..)—4.60Fig. 3.2. An indication of the variability of food concentration during growth experiments on Strombidinopsisacuminatum strain SPJSC, over a 5 day period (for 1 of the 5 experiments presented in section 1.1). Datarepresented by circles and connected by solid lines are the log concentration of prey (7halassiosira pseudonanamL-1 ). The broken lines represent the desired concentration of cells.482.01.5..-•` 1.01I- 0.5a)4-, 0.0MIL-4.. 3-0.50;-4-1.0C..115000 30000 45000 60000 75000 90000^0^15000 30000 0^15000-1Thalassiosira pseudonarta, (# mL )Fig. 3.3. The numerical response of Strombidinopsis acurninatum strain SPJSC from 3 experiments (for detailssee section 1.1.2.3 for panel a; 1.2.2 for panel b; and 1.3.2 for panel c). All data points represent growth rates.The open circles, triangles and solid circles in panel c represent different sub-clones (see section 1.3 for details).The curves in all figures are the modified Michaelis-Menten fit (Eq. 1 in text) to the data (see Table 3.1 for theparameters associated with these curves).0.70—, 0.6.--iI'MI 0.5......4^0.4-4.0.3ci.)-i-) 0.2cr5;--1^0.04-I-) 0.0o —0.1;--4C —0.2—0.3490^5^10^15^20^25Number of generationsFig. 3.4. The change in growth rate (c1-1) with number of generations for Strombidinopsis acuminatum strainSPJSC grown on saturating food concentrations (see section 3.0 for details). The solid circles, connected by asolid line, represent the maximum growth rate, which was exhibited by 1-6% of the population. The triangles,connected by a broken line, represent the average growth rate of the experimental population.• •2.01.5 ••a•••• • ••• ••••••-0•• • • • •o0^—2.0 ^• •t71^0^• •-1^0^ ° 001^ 0—2 — 05010000 20000 30000 40000 50000 60000 70000 80000 90000 100000--Prey concentration (mL1Fig. 3.5. Panel a, The numerical response of Strombidinopsis acuminatum strain SPJSC from a combination of 3experiments (sections 1.1-1.3). All data points represent growth rates. The solid line represents the modifiedMichaelis-Men ten response (Eq. 1 in text) between 0-1x105 prey mL4 using data represented by both open andsolid circles. The broken line represents the modified Michaelis-Menten response between 0-9x104 prey mL-1using only data represented by the solid circles (see section 6.0 for details on how outliers were removed). SeeTable 3.1 for the parameters associated with the curves. Panel b, The distribution of the residuals for theMichaelis-Menten fit to all the data in Fig. 3.5a (open and solid circles, solid line). Panel c, The distribution ofthe residuals for the Michaelis-Menten fit to a subset (solid circles, broken line) of the data in Fig. 3.5a (seesection 6.0).---- 25al 20rn 15a)105130i 254.320'CI 150:14)pa1050^10^20^30^40^50^60Time (min)80000^100000--1Bead concentration (# mL )Fig. 3.6. Panel a, The change in number of fluorescent beads ingested by Strombidinopsis acuminatum strainSPJSC over time at two bead concentrations (open circles and solid line, 4x104 beads mr...-1; solid circles andbroken line, lx 1O4 beads mL4). Both lines are linear regressions forced through the origin. Note: these twoexamples were measured every 15 min over an hour while most estimates were made by measurements madeevery 5 min over 30 min (see section 7.0). Panel b, The functional response of stain SPJSC ingesting 5 Amfluorescent beads (section 7.0 for details). All data points jepresent grazing rates. The solid line is the Michaelis-Menten fit (Eq. 2 in text) to all the data between 0-1.1x10" beads mrl, and the broken line is the same fit butdoes not include the open circles (see section 7.2 for details). See Table 3.1 for the parameters associated withthese curves.0^20000^40000^6000052CHAPTER 4GROWTH AND GRAZING RATES OF Strobilidium spiralis STRAIN IA AS AFUNCTION OF FOOD CONCENTRATIONIntroductionStrobilidium spiralis was the last ciliate I maintained in culture. Therefore,experiments conducted on it were refined, compared to those described in the previouschapter. Unlike Strombidinopsis acuminatum (Chapter 3) which was maintained for almost ayear, this ciliate was only maintained for 3 months before the culture died. This was typicalfor many of the ciliates I isolated, and I was often not able to perform a complete suite ofexperiments before the clones were lost. This will become apparent in later chapters wheredata are either missing or unsatisfactory. I was however, able to obtain estimates of grazingand growth rates for S. spiralis. Further, I documented the change in culture viability duringthe period the ciliate was maintained. This provided criteria to evaluate the growth andgrazing response data.PART I GROWTH RATES: Estimating the Numerical Response1.0 General MethodsExperiments were conducted to determine the numerical response of Strobilidiumspiralis strain IA grazing on the prey flagellate Chroomonas sauna, at food concentrationsranging from 0 to 9.5x104 cells mL-1. The ciliates were maintained in semi-continuouscultures by transfers to new prey and medium on a daily basis, as outlined in Chapter 3.Modifications of this method follow.After S. spiralis was isolated (see Chapter 2), it was fed 3 prey flagellates(Chroomonas sauna, Isochlysis galbana, and Rhodomonas lens). The ciliate grew on all 3prey individually and in combinations, but it grew faster on some prey (Table 4.1).Strobilidium spiralis grew well on C. sauna alone, often as well as it grew on combinations of2 or 3 prey. Since a major goal of this study was to determine the impact of a mono-clonal53ciliate bloom on a single prey species bloom (see later chapters), experiments were conductedusing C. sauna alone to determine the ciliate's numerical response.Chroomonas sauna was maintained in log phase, in 300 mL flasks, by serial dilution,on a 14:10 light:dark cycle with irradiance of 50-100 /Imo' photons m-2 s-1 at 16-17°C. Priorto the experiments, S. spiralis was maintained in 20 mL tissue culture wells containingenriched natural seawater with C. sauna on a 14:10 light:dark cycle with lighting of 25-50Amol photons m-2 s-1 at 16-17°C. For experiments, the ciliates were collected during loggrowth phase; prey concentrations, from which the ciliates were collected, were notcontrolled.During the experiment, ciliates were grown in the dark in 20 mL tissue culture platescontaining 10 mL of prey and medium. Prey concentrations were prepared and enumerated asdescribed for Thalassiosira pseudonana in Chapter 3 (section 1.0). Typically, eachconcentration was replicated 3 times. The semi-continuous culture method was similar to thatdescribed in Chapter 3 (section 1.0), for Strombidinopsis acuminatum grown on T.pseudonana, except that daily transfers were continued for 7 days rather than 5 and theaverage growth rate was determined for days 3-7 (days 1 and 2 were an acclimation periodafter which growth rate was relatively constant, data not shown). The numerical response datawere fitted to a modified Michaelis-Menten model using an iterative fit (Eq. 1, Chapter 3).A total of 32 treatments (prey concentrations) were examined in two consecutiveexperiments; treatments ranged from 0 to 9.5x104 prey mL-1. Prey concentrations weremeasured before and after each transfer of the ciliates to examine the stability of the preyconcentrations. Typically prey concentrations varied in a manner similar to that seen for T.pseudonana in Chapter 3: prey number increased slightly over the 24 h period and lowerconcentrations varied, on a relative scale, more than higher ones. Growth rates were plottedagainst the average prey concentration during the incubation period (5 days).2.0 ResultsGrowth rate followed a rectangular hyperbolic response between 0-9.5x104 prey mL4(Fig. 4.1). At concentrations <6x103 prey mL-1 mortality occurred within the first 48 h (the54acclimation period). After the acclimation period, at concentrations <1.0x104 prey mL-1,mortality occurred, presumably due to starvation. Equation 1 (Chapter 3) was fitted to pointsfrom 0-9.5x104 prey mL-1 (Fig. 4.1). The parameters and estimates of error of this equationare presented in Table 4.2.3.0 Change in Culture Growth Rate Over TimeStrobilidium spiralis was collected in early March. By early April it was beingmaintained as a monoclonal culture grown on a single prey: C. sauna. At 3 times, betweenearly April and mid May, the culture growth rate was measured. Ciliates were acclimated to5x104 C. sauna mL-1. Then, 70-100 ciliates were removed from the cultures and placed inseparate wells in a 24 well tissue culture plate (well size = 3.5 mL). Each well contained 2mL of culture medium plus 5x104 C. sauna ml.,4 (a concentration yielding maximum growthrate, see Fig. 4.1). The ciliates in tissue culture plates were placed in conditions similar tothose in which the ciliates had been previously maintained (see section 1.1). After 24 h thenumber of ciliates well-1 was determined.In the first of these isolations, more than a month after the ciliates were collected, themodal number of progeny from a single cell was 4 and some cells produced up to 8 progeny.Nineteen days later, in the second isolation, the modal number of progeny was 2 and in nocases were 5 cells produced. Eight days later, in the final isolation, many cells had died,few were dividing and the maximum number of progeny was 2 (Fig. 4.2). Two weeks laterthe culture was dead. These results indicate how the culture vitality decreased over 2 months.The reason for this decrease in growth rate was not determined. This was a trend fornot only this ciliate but for several others I isolated (data not presented). In fact, it is a featuretypical of most cultured oligotrichs (Gifford 1985). In the last chapter I argued that geneticdeterioration was the primary cause of culture decline, but other reasons may exist for thispersistent feature of oligotrich culturing.As mentioned in Chapter 3, growth rate in most ciliate cultures decreases over time dueto genetic changes (Bell 1988). Such reductions in growth rate are typical over several55months, but S. spiralis stopped growing after several weeks. Possibly one of the geneticinfirmities, that plague isolated cultures, already existed in the collected S. spiralis. Thiswould have escalated the culture's deterioration. Alternatively, it may have been inadequateculturing methods that caused the decline of the ciliates.Lack of essential nutrients in the food supply may cause a decline in ciliate culturevitality. For a nutrient to be virtually diluted out of S. spiralis by the number of observedfissions (20-40), its initial concentration would need to be <220-240 ( 105- 1012) cell-1.The likelihood of a nutrient being diluted out of the S. spiralis culture was high. For example,the carbon content of S. spiralis is only 6x1016 atoms cell-1 (assuming a cell -50 Am indiameter and 0.19 pg carbon Am- Putt and Stoecker 1989). Since most essential nutrientshave concentrations far below that of carbon, the depletion of a nutrient was plausible in mycultures. In fact, empirical data support this reasoning. Growth rate of the ciliate Didiniumnasutum decreased due to "poor" food quality over 20 days (<40 fissions), and the ciliate wasrevitalized by the introduction of "good" food (Beers, summarized in Bell 1988). It appearsthen that growth rate can be depressed over an extended period by poor food quality. Severalsubstances (e.g. amino acids, purines, pyrimidines, lipids and trace elements) are required forParamecium (Wichterman 1986), trace metals (either in the medium or prey) appear to berequired for oligotrichs (Gifford 1985), and chloroplasts are required in the diet ofStrombidium capitatum (Stoecker and Silver 1990). Possibly one of these was depleted in myculture of S. spiralis.An alternate scenario, still within the confines of culturing technique, would be toinvoke the role of some sub-lethal toxin. Such a substance could accumulate in the ciliatesover time and eventually reach lethal levels. Trace metals may also act in this manner(Gifford 1985, Stoecker et al. 1986).Regardless of the cause of growth rate decline, its impact has bearing on the results ofthis chapter. The growth rate experiments (section 1.0) were conducted from March 30 toApril 9 during which time growth rate was high (Fig. 4.2). Possibly though, growth rate waseven higher prior to these experiments. The grazing rate experiments (section 6.0) were56conducted on April 18-20 by which time growth rate had likely declined. Assuming thatgrowth and grazing rates are correlated, the grazing rates of the ciliates that were used todetermine growth rates may have been higher than those described in section 6.0.4.0 Examining the Growth Estimates and Developing a Numerical Response ModelThe problems associated with the methods used in this chapter parallel those discussedin Chapter 3 (section 4.0). For instance, mortality was likely over-estimated and handling ofciliates was probably not an important bias. As discussed above, genetic degradation orculturing methods may also have effected growth rate, but this was not explored.In general, the results of the numerical response of S. spiralis were consistent with themodel (Eq. 1, Chapter 3). Some deviation occurred between 5x103-1.5x104 prey mL-1 (Fig.4.1). A change from rapid growth to mortality occurred over a narrow prey concentration.Then, mortality was almost constant for a range of prey concentration that were inconsistentwith those predicted by the model. This may indicate switching behaviour on the part of theciliate that is not consistent with the simple Michaelis-Menten dynamics. Possibly reducedactivity (either physiological or physical) occurred below mortality-inducing concentrations.Weight specific respiration can decrease over starvation periods for protozoa (Fenchel 1982,Caron et al. 1990). Further, starvation reduces cell size in protozoa, and this may alter thegrowth rate (Fenchel 1982, 1990).The change in cell size with food concentration was measured for strain IA, todetermine if this might explain the anomalous mortality rates. In these experiments, theciliates cultured at concentrations at <1.8x104 C. sauna mL-1 died, and there was no trend inthe change of cell volumes between 1.8x104-1x105 prey mL-1. However, these experimentswere made in mid-May near the end of the culture's life, and thus the results are suspect.There may, in fact, have been a change in cell volume with food concentration in theexperiment in section 1.0. Alternatively, the mortality rate data may be misleading since, asmentioned above (Chapter 3, section 4.1), they are likely overestimated. Such anoverestimation could cause the jump in response seen at the concentration where mortality57occurred. With these potential deviations from the model in mind, I have still used Eq. 1 forestimating the response curve.5.0 Aspects of the Growth Rates and Their Ecological RelevanceIn general, the data for this species provided a good description of the numericalresponse. There are 4 useful pieces of information provided by the above experiments: 1) thethreshold concentration of C. sauna for S. spiralis was -104 mL-1; 2) the maximum increasein growth rate was between 104-2x104 prey mL4 but may actually be over a narrower rangenear 1.5x104 prey mL-1; 3) the maximum growth rate was -2.5 divisions d-1 but was as highas 3 d-1 (Fig. 4.1 and 4.2); and 4) the growth rate of this culture declined to zero over aperiod of three months (Fig. 4.2), suggesting either culture conditions were inadequate,genetic deterioration was rapid or a combination of both.During the late winter, when strain IA was collected in Indian Arm (see Chapter 2), theabundance of phytoplankton, of a similar size to C. sauna (3-10 gm), was typically <102mL4 (Martin and Montagnes 1993). These prey levels would not sustain the ciliate,suggesting that either S. spiralis grows much better on other prey, the ciliates were in areduced metabolic state (winter temperatures are 4-10°C), or the prey were distributed in smallconcentrated patches. However, other numerical response-curve estimates for S. spiralisindicate that it is able to survive at much lower food levels (see Chapter 8, Fig. 8.1, Table8.1); possibly my estimates are erroneous or this clone is distinctly different from previouslystudied (Atlantic) strains (see Chapter 2).PART H GRAZING RATES: Estimating the Functional Response6.0 General MethodsInitially, to determine if S. spiralis consumed fluorescently labeled prey (Gonzalez etal. 1990, Sherr and Sherr 1993) the ciliates were fed a 1:1 mixture of stained and live C.sauna at 105 particles mL-1. The ciliates died after <45 min in this treatment. However, theciliates lived and consumed fluorescent beads in experiments using a 1:1 mixture of beads58(used in Chapter 3) and live C. sauna at -105 particles mL-1. Hence, beads were used as asurrogate food to estimate grazing rates.Fluorescent beads and C. sauna were fed to strain IA over an 80 min period at 5x104prey mL-1. Bead ingestion was measured at 2 to 10 mm intervals and followed a rectangularhyperbolic response. Over the first 20 min, this response was virtually linear (Fig. 4.3).Thus, a 20 mm incubation period was used for subsequent experiments, assuming that grazingrate was constant from when beads were added to 20 mm later.An experiment was conducted to determine the effect of prey concentration on thegrazing rate of the ciliate S. spiralis. Fluorescent beads (see Chapter 3) were fed to the ciliatesin a 1:1 ratio with live C. sauna. The experiments were run at particle concentrations rangingfrom 103 to 9.5x104 prey (beads plus C. sauna) mL4. Live prey concentrations were madeby diluting a culture of C. sauna (maintained as described in section 1.0) with naturalpasteurized sea water. After diluting the C. sauna culture to the desired concentrations, thetreatments were counted using a Coulter Counter. Twenty-two concentrations (treatments)were made this way. Five mL of each prey treatment were placed in tissue culture plate wells.Each treatment was replicated 3 times, making a total of 66 wells.The ciliates were grown on C. sauna prior to the experiments and harvested when thecultures were in log growth phase. Ciliates (40-50) were transferred, using finely drawnpipets, to the wells containing the live-treatments and acclimated to these treatments for 4 h at16-17°C at 5 Amol photons m-2 s4. During the acclimation period, bead treatments thatequalled the prey treatments were prepared (following methods similar to those described inChapter 3) and counted on the Coulter Counter. After the acclimation period, 5 mL of bead-treatments were added to the appropriate live treatment, thus exposing the ciliates to 10 mL of1:1 beads:live prey at 22 concentrations. The mixtures were incubated at 16-17°C, 5 Amolphotons m-2 s-1 for 20 mm and agitated every 5 mm to reduce settling of beads and prey.After 20 mm, the 10 mL in each well were transferred to a vial, fixed with acidLugol's iodine (final concentration 2% volume/volume) and later bleached with concentratedsodium thiosulphite (since Lugol's masks fluorescence). Some ciliates were lost during this59manipulation; typically 20-40 ciliates were observed. The ciliates were examined using 10 mLsettling chambers and an inverted fluorescent microscope (see Chapter 3). The mean numberof beads ciliated was determined and ingestion rate estimated; ingestion rate wasapproximately twice the number of beads in the ciliates (depending on the actual ratio ofbeads:live prey at each treatment level). I assumed no preference for beads or prey and noegestion of prey due to fixation, both of which may underestimate grazing rates (Sieracki et al.1987, Stoecker 1988).7.0 Results, Discussion and Modeling the ResponseGrazing rates ranged from 3 to 60 beads ciliate-1 h-1 and increased with preyconcentration, but there was variation in the grazing rate at single concentrations (Fig. 4.4, alldata). The grazing rate data were fitted to a rectangular hyperbolic response using an iterativefit (Eq. 2, Chapter 3) between 0-9.5x104 prey mL4 (Fig. 4.4, solid line). At concentrations<3x103 prey mL-1 grazing rate dropped.The relation determined above did not give a good fit to the grazing rates above4.5x104 prey mL-1. Possibly the data were not accurately represented by Eq. 2 (Chapter 3).However, assuming that gross growth efficiency was constant (at concentrations elicitingmaximum growth rates), the notion that grazing rates >4.5x104 prey mL-1 were elevated, isnot supported by the growth data (Fig. 4.1) since a corresponding increase in growth rate didnot occur above 4.5x104 prey mL-1. Thus, I have used the curve based on Eq. 2 (Chapter 3)as an estimate of the functional response, to examine grazing rates in Chapter 8. Theparameters of the equation, from the line depicted in Fig. 4.4, and its estimates of error arepresented in Table 4.2.CONCLUDING REMARKSStrobilidium spiralis (= Lohmanniella spiralis, see Chapter 2) is found in many coastalmarine waters (Smetacek 1981, 1984, Rassoulzadegan 1982, Sheldon et al. 1986, Lynn andMontagnes 1988, Verity 1991a). This genus has been reported at high densities, suggesting60that it forms blooms (Smetacek 1981, 1984). The data from my studies also suggest that S.spiralis has the potential to form blooms over short periods, as it grows rapidly under optimalconditions. My estimates and those of Smetacek (1984) indicate that S. spiralis can undergoup to 3 divisions d4 and likely has a maximum sustainable growth rate of 2.5 divisions d-1(see Chapter 8, Table 8.1). The ubiquitous nature of this ciliate, its large size and its rapidgrowth and grazing rates undoubtedly make it one of the more important components of themicrozooplankton.6 1Table 4.1. The relative growth rate of Strobilidium spiralis strain IA over a 2 and a 4 day period. The ciliateswere grown on a 14:10 light:dark cycle at 16-17°C on all the combinations of the three prey: Isochtysis galbana,Chroomonas sauna, and Rhodomonas lens (the initial combined prey concentrations was -3x104 cells mI.-1).Growth rates were qualitative estimates of increase in numbers: cells died, -; cells grew slowly, +; cells grew,+ +; cells grew quickly +++.PREYCONCENTRATION REPLICATE #1day2^day4GROWTH RATEREPLICATE #2day2 day4REPLICATE #3day2^day4R. lens + - ++ - ++I. galbana + + + + + +C. sauna + +++ ++ +++ ++ +++1. galbana & R. lens + + + + + + + + +++ ++I. galbana & C. sauna +++ ++ +++ +++ +++ +++C. salina & R. lens ++ +++ +++ +++ +++ +++1. galbana & C. sauna +++ ++ +++ ++ +++ +++&R. lens62Table 4.2. Growth and grazing parameters for Strobilidium spiralis strain IA; parameters and estimates of error ofthe numerical and functional response curves presented in Figs. 4.1 and 4.4. See section 1.0 in Chapter 3 for theequations used. Values are in units of numbers m1:1 for concentrations (k ,x'), and c14 for growth rate (A x).valueMichaelis-Menten fit for Fig. 4.1standarddeviationcoefficientof variationAmax 1.845 0.2572 13.94k 19070 2866 15.03x' 10230 919.8 8.989Michaelis-Menten fit for Fig. 4.4, solid lineGmax 45.45 2.751 6.0524028 1359 33.74Michaelis-Menten fit for Fig. 4.4, broken lineGmax 50.21 3.869 7.7064473 1951 43.622,^I^,^I^ I^.^1 0^20000^40000^60000^80000^100000—Prey concentration (# mL 1 )Fig. 4.1. The numerical response of Strobilidium spiralis strain IA grown on the prey flagellate Chroomonassauna. The circles represent growth rates and the line represents the modified Michaelis-Menten fit to the data(Eq. 1, Chapter 3). See Table 4.2 for the parameters and estimates of error associated with this curve.63;April 27, 1992-ICIMay 5,, 1992II500to 40a)ci 20c.)10C.)^0cda)504030rn 20-8 1004-150C)40ti)cd 3020C) 10C) 0a.. 0^1^2^3^4^5^6^7^8^9Number of progeny fromone cell after 24 h64Fig. 4.2. The change in culture growth rate of Strobilidium spiralis strain IA over time. Each panel is thedistribution of progeny from 70 to 100 single cell isolates after 24 h. (see section 3.0 for details).40.-II^30cu-4—)ccs•,--4r--1• ■•■•1c) 20m--c5cticuCr.) 100650^20^40^60^80Time (min)Fig. 4.3. The number of beads ingested by Strobilidium spiralis strain IA over time. At two different times,ciliates were fed beads and prey as described in section 6.0, at 2 concentrations: 4.7x104 beads mL-1 (solidcircles and 4.4x104 beads mL-I (open circles). The dotted line was fit to the function (beads ciliate d ) = 3.2(1-e41.043*min) using a combination of both data sets. The solid line is the linear regression fitted to only the opencircles; it is bound by the 95% confidence intervals around the linear regression (dashed lines).40000^60000^80000-1Prey concentration (# mL )100000Fig. 4.4. The functional response of Strobilidium spiralis strain IA ingesting a 1:1 ratio of 5 i2m fluorescentbeads and Chroomonas sauna (see section 6.0 for details). The curve is the Michaelis-Menten fit (Eq. 2, Chapter3) to the data. See Table 4.2 for the parameters and estimates of error associated with these curves.6667CHAPTER 5GROWTH AND GRAZING RATES OF Strobilidium sp. STRAIN JERC AS AFUNCTION OF FOOD CONCENTRATIONIntroductionThis chapter is placed here to offer, by its proximity, a comparison to the last one (adetailed comparison of species is presented in Chapter 8). Both chapters examine species inthe genus Strobilidium, but the growth and grazing rates of the two species differ considerably.There are also differences in the methodologies used to study the two species. Thesedifferences stem from logistic problems and differences in the ciliate's behaviour. Still,comparisons can easily be made between the two species.PART I GROWTH RATES: Estimating the Numerical Response1.0 General MethodsExperiments were conducted to determine the numerical response of Strobilidium sp.strain JERC grazing on a combination of two flagellates, lsochrysis galbana and Chroomonassauna at concentrations ranging from 0 to 9.2x104 cells mL-1. The prey were maintained, inlog phase, by serial dilution, in 150 mL flasks at 50-100 Amol photons m-2 s-1 on a 14:10 hlight:dark cycle at 16-17°C. Prior to the experiments, Strobilidium sp. was maintained inenriched natural seawater with I. galbana and C. sauna in 6-well 20 mL plastic tissue cultureplates at 16-17°C and on a 14:10 h light:dark cycle at 25-50 Amol photons m-2 s-1.For these experiments, prey concentrations were prepared and quantified as describedin Chapter 3, but the prey concentrations were made by combining the two species, I. galbanaand C. sauna, in a 1:1 ratio. Although the initial food concentrations were in this 1:1 ratio, itwas not determined if they remained this way over the experimental period.For the growth experiments, ciliates were collected during log or early stationaryphase. Then, 10-25 ciliates were transferred to tissue-plate wells containing 10 mL of prey attreatment-concentrations and acclimated in the dark for 24 h. After 24 h, the ciliates weretransferred to new prey, at the same concentrations they were acclimated to, and allowed togrow for 96 h, for most treatments. This was too long a period for starving ciliates (usuallyall died after <96 h), and the concentrations below 4x103 prey mL-1 were left for only 24 h68after the initial 24 h acclimation period. Growth rate was estimated as the change in numbersover the incubation period; exponential growth was assumed over the 24-96 h (i.e. growth ratewas calculated as A = lit In [cellst+ i/cellst], where t = time in days).Typically, each treatment was replicated 3 times. The prey concentrations used todetermine the numerical response were estimated, for each well, prior to the addition of, andafter the removal of the ciliates; these two values were then averaged for each "replicate".But, prey were often depleted to different extents in each of the 3 "replicate" wells (whichwere initially the same concentration). Thus, the "replicates" did not have the same final, oraverage, concentration and were not considered true replicates. Typically, prey concentrationsdropped to 5-35% of the initial concentration over the 96 h, but at times the prey levelsincreased (-15%) while at one extreme there was an 80% decrease. There was a negativecorrelation between initial food concentration and percent change (i.e. low concentrationsdecreased more than higher ones). However, there did not appear to be any relation betweenchange in prey concentration and growth rate. Since the rate of change over the 96 h was notmonitored, these differences were ignored. This procedure could alter the precision ofestimates of prey concentration and therefore overestimate the threshold concentration by -15-20%. Further, if the two prey species were not grazed equally (a parameter which could notbe determined by Coulter Counter enumeration), this overestimation could be higher.Growth rate for each well was plotted against the average prey concentration of thatwell. The numerical response data were fitted to a modified Michaelis-Menten model (Eq. 1,Chapter 3) using an iterative fit.2.0 Results and Discussion2.1 GeneralAt first, semi-continuous growth rate experiments, like those described in Chapters 3and 4, were attempted for Strain JERC. However, this ciliate was smaller than the otherspecies and tended to "jump" when a finely drawn pipette tip was brought near it (see Chapter2). It was therefore not possible to collect and transfer the ciliates on a daily basis and the6 9method described above was adopted to estimate growth rate. This long incubation methodwas not as satisfactory as the semi-continuous method because it did not ensure that foodconcentration remained constant (on a daily basis) for the duration of the experiment and it didnot allow repeated measurements.Strain JERC was initially cultured on a combination of I. galbana, C. sauna and R.lens. Later it was grown on all combinations of these species; the ciliate grew on a number ofprey combinations (Table 5.1). Although a diet of R. lens alone induced maximum growth(Table 5.1), later experiments indicated that this growth rate was not sustainable (data notshown). Consequently, the numerical response was conducted using a 1:1 mixture of the preyI. galbana and C. sauna, which also gave a maximum sustained growth rate (Table 5.1).2.2 Numerical ResponseA total of 76 treatments with no replication (see section 1.0) were run in 3 consecutiveexperiments (as described in section 1.0). The growth rate followed a rectangular hyperbolicresponse between 1.7x103-5.7x104 prey mL-1 (Fig. 5.1). The threshold concentration of preywas -4x103 prey mL-1. At concentrations above 5.7x104 prey mL-1 growth rate wasinhibited, and these data were not used to estimate the numerical response. Equation 1(Chapter 3) was fitted to points between 1.7x103-5.7x104 prey mL4 (Fig. 5.1). Theparameters of the equation and their estimates of error are presented in Table 5.2.The calculation of mortality rate in these experiments differs from that described inprevious chapters. In Chapters 3 and 4 mortality, was the average rate over several days, aftera 2 day acclimation period. Here, the mortality was typically estimated over 24 h, after a 24 hacclimation period. This would give relatively lower mortality rates than those obtained by theprevious methods, assuming mortality rate due to starvation is not constant (see Chapter 3).Consequently, the mortality response, presented in Fig. 5.1, might have a different shape if itwas obtained by previous methods.7 03.0 Aspects of the Growth Rates and Their Ecological RelevanceThe data for this species provide a good prediction of the numerical growth response.There are 3 pieces of information provided by the above experiment: 1) the thresholdconcentration of prey for strain JERC was -4x103 mL-1; 2) the maximum increase in growthrate was between 4x103-1.5x104 prey mL-1, and 3) the maximum growth rate was onedivision d*During May when this ciliate was collected at Jericho Pier in Vancouver Harbour,potential prey (10 m in diameter) concentrations in British Columbian coastal watersranged from 102-x103 cells mL-1 (Appendix 3). Thus on average, natural levels of foodcould be sufficient to support strain JERC but would be unlikely to support it at its maximumgrowth rate.PART II GRAZING RATES: Estimates of the Functional Response4.0 General MethodsInitially to determine if strain JERC consumed fluorescently labeled beads, (seeChapters 3 and 4) the ciliates were fed a 1:1 mixture of beads and C. sauna at 3x104 particlesmL-1. Virtually no beads were ingested after an hour, indicating a "reluctance" of the ciliateto eat beads. Hence, a different surrogate food was tested: fluorescently stained C. sauna(Gonzalez et al. 1990, Sherr and Sherr 1993). The ciliate consumed fluorescently labeled C.sauna at concentrations of 2x104 mL-1 and prey uptake was linear over 1 h, sampled at 10min intervals (data not shown). Thus, a 1 h incubation period was used for subsequentexperiments.Two identical (replicate) experiments were conducted to determine the effect of preyconcentration on the grazing rate of strain JERC. Fluorescently stained C. sauna were used astracers of ingestion in a 1:1 ratio with live C. sauna. The experiments were run at foodconcentrations ranging from 1.0x102-4.8x104 prey mL-1 (open circles, Fig. 5.2) and2.5x102-4.1x104 prey mL-1 (solid circles, Fig. 5.2).71The stained prey were prepared as described by Sherr and Sherr (1993), andconcentrations of live and stained prey were made up as described for live cells and beads inChapter 3 and 4. Prior to estimating their concentration, the stained prey were forced througha 10 Am Nitex mesh to insure monodispersion. Stained prey were made the day of theexperiment and no preservatives were used (see Sherr and Sherr 1993).Live-prey concentrations were made by diluting a log phase culture of C. sauna, grownat 50 Amol photons m-2 s-1 in enriched culture medium, with pasteurized natural seawater.After diluting the stock culture to the desired treatment concentrations, they were countedusing a Coulter Counter and allocated to 20 mL wells (5 mL well-1) in tissue culture plates.For both experiments, 20 live C. sauna concentrations (treatments) were made this way andeach was replicated 3 times making a total of 60 wells.Prior to the experiments, the ciliates were grown on I. galbana and C. sauna. Bothciliates and prey were harvested when the cultures were actively growing (log phase). Toremove all prey, -3000 ciliates were transferred (using finely drawn pipettes) through 3-4rinses of pasteurized natural seawater. Subsequent to rinsing, 30-50 ciliates were transferredto the wells containing the live-prey treatments. The ciliates were acclimated in these live-prey treatments for 4-5 h at 16-17°C and 5 Amol photons m-2 s-1.During the 4-5 h acclimation period, stained-prey treatments were made up toconcentrations equaling those of the live-prey; these were also counted on the Coulter Counter.After the 4-5 h period, 5 mL of the appropriate stained-prey were added to the ciliate and live-prey mixtures. Thus, the ciliates were exposed to 1:1 stained:live prey at 20 concentrations.The mixtures were left at 16-17°C and 5 timol photons m-2 s-1 for 1 h, with agitation every10 min to reduce prey settling. After 1 h, the 10 mL were transferred to vials and processedby a method similar to that used for Strobilidium spiralis in Chapter 4; typically 20-40 ciliateswere observed.7 25.0 Results and DiscussionGrazing rate ranged from 0 to 24 prey ciliated hd and showed a trend to increase withprey concentration from 0 to 5-7x103 prey mLd, but there was a steady decline in grazingrate above 5-7.5x103 prey mLd (Fig. 5.2, both data sets). The grazing rates, for both datasets depicted in Fig. 5.2, were fitted to rectangular hyperbolic functions using an iterative fit(Eq. 2, Chapter 3) between 0 and the maximum grazing rates. At concentrations above themaximum grazing rate, an asymptotic response, not in accordance with the data, was assumedfor both curves.The rectangular hyperbolic response, typically invoked to explain a functional response(see Appendix 2), did not fit the data in Fig. 5.2. The reasons for this are unclear; possiblyfood concentrations above a certain level inhibited feeding. However, a concomitant decreasein growth rate would then be expected (assuming growth and grazing are positivelycorrelated); this was not seen (Fig. 5.1). Alternatively, there may have been something in thestained-prey preparation that inhibited feeding and as the concentration of this "substance"increased, its effect increased. In the absence of an inhibitor, feeding rates would havepresumably continued to increase and/or become asymptotic as food concentration increased.Consequently, I have used Eq. 2 (Chapter 3) for the first part of the curves depicted in Fig.5.2 and have predicted an asymptote for the remainder of the curves; these asymptotes wouldthen be minimum responses.A second anomalous trend, exhibited by the data in Fig. 5.2, is the difference inresponses between experiments (solid vs. hollow circles). These two data sets were fromexperiments performed 10 days apart but -45 days after the growth rate experiments (section1.0) and several months after the ciliate was isolated. The experiments used different"batches" of ciliates but were otherwise identical. The difference between the two curves mayhave been due to a genetic difference between batches (as discussed in Chapter 3) oruncontrolled differences in technique between the experiments (i.e. more of the "inhibitorysubstance", proposed above, may have reduced the grazing rate). In either of these cases, theupper response curve (solid circles, Fig. 5.2) would be the more accurate representation of7 3natural conditions. Hence, I have used this upper curve when comparing grazing rates inChapters 8. The parameters and estimates of error of both curves are presented in Table 5.2.CONCLUDING REMARKSSpecies of Strobilidium in the size range of strain JERC are found in many coastalmarine waters (Smetacek 1984, Andersen and Sorensen 1986, Lynn and Montagnes 1988,Martin and Montagnes 1993). Such species have been reported at high densities and have highgrowth rates, suggesting that they forms blooms (Smetacek 1984, Andersen and Sorensen1986). The data from my studies also suggest that strain JERC has the potential to formblooms. This is further supported by the model presented in Chapter 9. Often species ofStrobilidium (= Lohmanniella) are lumped together in field studies. This likely misrepresentsthe ciliate assemblage since the differences between two species of Strobilidium, stains IA(Chapter 4) and JERC, were pronounced (see Chapter 8).74Table 5.1. The relative growth rate of Strobilidium sp. strain JERC over 5 days. The ciliates were grown on a14:10 light:dark cycle at 16-17°C on all the combinations of the three prey: Isochtysis galbana, Chroomonassauna and Rhodomonas lens (the initial combined prey concentrations was 3x104 cells mL-1). Growth rates werequalitative estimates of increase in numbers: cells grew slowly, +; cells grew, + +; cells grew quickly, + + +.PREYCONCENTRATIONGROWTH RATEREPLICATE #1 REPLICATE #2 REPLICATE #3R. lens +++ ++ + +++ I. galbana + + +C. sauna + + +I. galbana & R. lens +++ + + ++I. galbana & C. sauna +++ + + + + + +C. sauna & R. lens ++ + + ++ + + +I. galbana & C. sauna & + + ++ + +R. lens'75Table 5.2. Growth and grazing parameters for Strobilidium sp. strain JERC; parameters and estimates of error ofthe numerical and functional response curves presented in Figs. 5.1 and 5.2. See section 1.0 in Chapter 3 for theequation used. Values are in units of numbers mL-1 for concentrations (k ,x.), and d-1 for growth rate (Amax).value^standard^coefficientdeviation of variationMichaelis-Menten fit for the line depicted in Fig. 5.1Amax 0.7212 0.1097 15.22k 10580 3473 32.83x 4075 545.8 13.39Michaelis-Menten fit for the solicl line depicted in Fig. 5.2between 100-4000 particles nil,' after which Gmax = 12Gmax^12.38^0.921^7.436711.687 159.7 21.96Michaelis-Menten fit for the brolicen line depicted in Fig. 5.2between 200-5000 particles mU after which Gm ax = 23Gmax 41.21^5.918^14.393865 903.0 23.3620000^40000^60000^80000^100000Prey concentration (# mL760.80.60.40.2cd 0.02-0.4—0.6Fig. 5.1. The numerical response of Strobilidium sp. strain JERC grown on a 1:1 ratio of the prey flagellatesChroomonas sauna and lsochrysis galbana. The circles represent growth rates and the line represents themodified Michaelis-Menten fit to the data (Eq. 1, Chapter 3). See Table 5.1 for the parameters and estimates oferror associated with this curve.0^80^e@ 008o 8......%v-II77251^20C.)as• ,-.4^.---1^15C)›-.Q.)^10;..1Ca,N.--^C)^5-6-1cc;;•40to0NMS —50C...4•^•• • •• ••10000^20000^30000^40000--Prey concentration (# mL1 )oo^•^•••^• 8• • •^•_-_0.o0 _50000Fig. 5.2. The functional response of Strobilidium sp. strain JERC ingesting a 1:1 ratio of stained and liveChroomonas sauna. The solid and open circles represent two separate experiments (see section 4.0). The twolines are fitted to these data using the modified Michaelis-Menten model (Eq. 2, Chapter 3) and assuming thatgrazing rate becomes asymptotic after maximum grazing rates were reached (see section 5.0). See Table 5.2 forthe parameters and estimates of error associated with these curves.7 8CHAPTER 6GROWTH AND GRAZING RATES OF Strombidium acuminatum Strain BJSCH AS AFUNCTION OF FOOD CONCENTRATIONIntroductionAlthough Strombidium acuminatum strain BJSCH was collected from the plankton, itsbehaviour indicated that it was semi-benthic (see Chapter 2). As well as swimming in thewater column, it "crawled" along the bottom of the culture dishes. This made conductingsemi-continuous growth experiments impossible and the technique for growth ratemeasurement described in Chapter 5 was used.As indicated earlier, I have not presented the experiments chronologically. StrainBJSCH was one of the first species that I examined. Hence, some methods in this chapter mayseem poorly designed compared to those previously described. This species is presented nearthe end since it is semi-benthic and therefore less comparable to the other planktonic forms.However, both the grazing and growth responses were obtained for strain BJLSC and anestimate of growth using two different prey was made. Further, the semi-benthic nature ofthis species suggested that it may be linked to suspended detritus (marine snow), and thereforeit might be used to illustrate grazers in regions of dense suspended material.PART I GROWTH RATES: Estimating the Numerical Response1.0 General MethodsThe numerical response experiments conducted on strain BJSCH were similar to thosein Chapter 5 but, in this case, two diets were examined: 1) the diatom Thalassiosirapseudonana at food concentrations ranging from 102 to 2.5x104 cells mL4 and 2) a 1:1:1combination of 3 flagellates, Isochtysis galbana, Chroomonas sauna, and Rhodomonas lens, atcombined concentrations ranging from 2.5x102 to 4.3x104 cells mL-1.Prior to the 3-flagellate numerical-response experiment, strain BJSCH was cultured ona combination of I. galbana, C. sauna, and R. lens. Later it was grown on all combinationsof these species; the ciliate grew on a number of prey combinations, but all 3 flagellates7 9combined gave the maximum growth rate (Table 6.1). Consequently, the numerical responsewas conducted using a 1:1:1 mixture of these 3 flagellates.Thalassiosira pseudonana was cultured as described in Chapter 3. The three flagellateswere cultured separately in 150 mL flasks at 16-17°C on a 14:10 h light:dark cycle at 50-100Amol photons m-2 s-1. Prior to the experiments, strain BJSCH was maintained in enrichednatural seawater in 6-well 20 mL plastic tissue culture plates at 16-17°C and collected duringlog or stationary phase.As in previous experiments, the ciliates were grown in the dark in tissue culture plates(with 10 ciliates and 10 mL of prey-concentration well-1); treatments were performed intriplicate. The ciliates were acclimated, at the various prey concentrations, in the wells for 24h. Then, the ciliates were transferred to new prey concentrations and allowed to grow for 124h (for the diatom experiment) and 96 h (for the 3-flagellate experiment). All preyconcentrations were measured using a Coulter Counter.As described in Chapter 5, there were changes in prey concentration over theincubation period, but the changes were not monitored in the same manner. The initial preyconcentration was the same for all 3 replicates of a single treatment. However, preyconcentrations were not measured in each well after the incubation period. Instead, theaverage of the 3 wells was measured (by mixing the three). Prey concentrations, used todetermine the numerical response, were averages of the initial and final (mixed)concentrations.Growth rate was estimated as the increase in numbers over the incubation period;exponential growth was assumed (i.e. growth rate was calculated as p. = lit In[cellst+ licellst], where t = time in days). The numerical response data were fitted to amodified Michaelis-Menten model (Eq. 1, Chapter 3) using an iterative fit.8 02.0 Results and Discussion2.1 GeneralAt first, semi-continuous growth rate experiments, like those described in Chapter 3,were attempted for strain BJSCH but this ciliate was small and tended to "hug" the bottom (seeChapter 2). Consequently, it was not possible to collect and transfer the ciliates and themethod described above was adopted to estimate growth rate. As mentioned in Chapter 5, thismethod has disadvantages, but given that the semi-continuous method was not feasible, itprovided an estimate of the numerical response.2.2 Numerical ResponseThe T. pseudonana experiment: The growth rate followed a rectangular hyperbolicresponse between 1x102-2.5x iO4 prey mL-1 (Fig. 6.1, solid circles). At concentrations below2.7x103 prey mL-1 mortality occurred, presumably due to starvation. Equation 1 (Chapter 3)was fitted to points between 9.3x10-2.6x104 prey mL-1 (Fig. 6.1, solid line). The parametersof the equation and their estimates of error are presented in Table 6.2.The 3-flagellate experiment: The growth rate followed a rectangular hyperbolicresponse between 2.5x102-4.3x104 prey mL-1 (Fig. 6.1, open circles). At concentrationsbelow 7.5x102 prey mL-1 mortality occurred, presumably due to starvation. Equation 1(Chapter 3) was fitted to points between 2.5x102-4.3x104 prey mL-1 (Fig. 6.1 broken line).The parameters of the equation and their estimates of error are presented in Table 6.2.3.0 Aspects of the Growth Rates and Their Ecological RelevanceIn general, strain BJSCH divided less than once a day (-0.88 d-1) at saturating preyconcentrations, had a threshold concentration of 7.5x102-2.7x103 prey mL-1, and themaximum numerical response was between 1x103-5x103 prey mL-1. However, theseparameters varied with prey type; the 3 flagellates gave a lower mortality rate, more rapidnumerical response and lower maximum growth rate than the diatom.Food type and quality affect ciliate growth (Blackbourn 1974, Repak 1983, Gifford1985, Verity and Villareal 1986, Skogstad et al. 1987, this study Tables 4.1, 5.1, 6.1, and817.1). The differences in growth rate response (Fig. 6.1) could be due to physical (Gifford1985) or algal biochemical composition (Thompson and Harrison 1992, Verity 1991a)differences of the prey or they may be due to sub-clonal differences of the ciliate (as seen inChapter 3). However, the relation of the two curves changes when viewed in terms of carbonrather than cells mL-1. The two curves are similar, but ciliates fed T. pseudonana reached ahigher growth rate than those fed the 3 flagellates (see Fig. 6.1, inset), but given the errorestimates of the parameters associated with these curves (Table 6.1.), there may be nodifference in these responses.PART II GRAZING RATES: Estimating the Functional Response4.0 General MethodsAn experiment was conducted to determine the effect of prey concentration on thegrazing the rate of strain BJSCH. Fluorescently labeled 5 ,am beads were used as a tracer ofingestion rate. Ciliates were fed beads and T. pseudonana in a 1:1 ratio at combinedconcentrations ranging from 4.5x102 to 5x104 mL4 .For two days prior to the feeding experiment, strain BJSCH was grown in tissue-platewells containing 18-20 mL of pasteurized natural seawater and T. pseudonana at 12concentrations. After two days the ciliates were removed from the wells. Then, 3-6 mL ofthe growth medium was used to estimate the well's abundance of prey, with a CoulterCounter. The remaining 15 mL of the medium were allocated to 3 wells in a tissue plate (5mL well-1). The ciliates were replaced in the medium and allowed to sit for 30 min. Thebeads, were prepared as described in Chapter 3 and the 1:1 ratio of beads to T. pseudonanawas established as described in Chapter 4. Beads were added to the medium and the ciliatesincubated for 30 min. The ciliates were kept at 16-17°C for all stages of the experiment. Thesamples were processed as described in Chapter 4. Typically, 20-40 ciliates were observed.8 25.0 Results and DiscussionGrazing rate ranged from 0 to 7 particles ciliate-1 h-1 and showed a trend to increasewith prey concentration between 4.5x102-3x104 prey mL-1 (Fig. 6.2). Although there wasconsiderable scatter in the data, they were fitted to the Michaelis-Menten model (Eq. 2,Chapter 3) using an iterative fit, and this curve was used to compare grazing rates in Chapter8.6.0 Swimming BehaviourStrain BJSCH grazed along the bottom of the tissue plates as well as swimming in thewater column. Some benthic ciliates tend to swim into the water column when foodconcentration is high (unpublished data) while a semi-benthic species of Strombidium movesoff the bottom when food becomes depleted (Fenchel and Jonsson 1988). I thought that strainBJLSC might alter its swimming behaviour at varied food concentrations, affecting its feedingrate and/or acting as a mechanism for dispersal. Therefore, I conducted an experiment todetermine the effect of food concentration on the swimming behaviour of strain BJLSC.Eight concentrations of T. pseudonana, ranging from 103 to 5x104 mL-1, wereprepared with natural pasteurized seawater, as described in previous sections. These 8treatments were allocated to 20 mL tissue plate wells (10 mL we11-1), with 3 replicatestreatment-1. Ciliates (50 well-1) were placed in the treatments and acclimated for 2 h at 16-17°C and 50 timol photons m-2 s-1. Then, the number of ciliates swimming in the watercolumn was microscopically (10-20x) observed (after a 3 min acclimation period on adissection microscope stage); ciliates not seen in the water column were assumed to be on asurface.Food concentration had no effect on the ciliate's position in the water column (Fig.6.3). Typically 15% of the ciliates were in the water column regardless of food concentration.Apparently, change in food concentration does not change the swimming behaviour of strainBJSCH and can not be invoked as a dispersal mechanism.83CONCLUDING REMARKSThis work is the first study of the grazing and growth responses of an elongate, semi-benthic species of Strombidium. There have been other studies on what are likely semi-benthicspecies of Strombidium (e.g. Rivier et al. 1985, Fenchel and Jonsson 1988, Bernard andRassoulzadegan 1990, Ohman and Snyder 1991), but these were all done on truncated forms(often referred to as Strombidium sulcatum but see Montagnes et al. 1990). The shapedifference of strain BJSCH may be related to its feeding and swimming behaviour, as thisspecies differed from others (see section 6.0 and Chapter 8). Further, this elongate morphseems to be a common but poorly examined type (see Chapter 2). Thus, this study adds newinsight into the adaptive diversity of the genus Strombidium. Further, since semi-benthicciliates may be important links in the formation and mineralization of ocean sinking detritus(Silver et al. 1984, Lochte 1991), this species and its response curves could be used toinvestigate the impact of ciliates on marine snow.8 4Table 6.1. The relative growth rate of Strombidium acuminatum strain BJLSC over a 4 day period. The ciliateswere grown on a 14:10 light:dark cycle at 16-17°C on all the combinations of the three prey: lsochrysis galbana,Chroomonas sauna, and Rhodomonas lens (the initial combined prey concentrations was 3x104 cells mL,-1).Growth rates were qualitative estimates of increase in numbers: cells died, -; cells grew slowly, +; cells grew,+ +; cells grew quickly, + + +.PREY^ GROWTH RATECONCENTRATION^REPLICATE #1 REPLICATE #2 REPLICATE #3R. lens^ -I. galbana +^+C. saunaI. galbana & R. lens^++^ -I. galbana & C. sauna + + +C. sauna & R. lensI. galbana & C. salina &^++^+++ ^+ +R. lens85Table 6.2. Growth and grazing parameters for Strombidium acuminatum strain BJLSC; parameters and estimatesof error of the numerical and functional response curves presented in Figs. 6.1 and 6.2. See section 1.0 inChapter 3 for the equation used. Values are in units of numbers nal.:1 for concentrations (k ,x'), and d-1 forgrowth rate (Amax).value^standarddeviation of variationMichaelis-Menten fit for the solid line in Fig. 6.1coefficientPmax 0.6118 0.07112 11.62k 5812 1224 21.05x' 2709 335.9 12.40Michaelis-Menten fit for the broken line in Fig. 6.1Amax 0.4169 0.02959 7.098k 1247 327.1 26.22x' 753.4 112.5 14.94Michaelis-Menten fit for the line in Fig. 6.2Gmax 4.214 0.7463 17.713401 2758 81.080.6,—,1rti 0.4—0.40^10000^20000^30000^40000--1Prey concentration (# mL )Fig. 6.1. The numerical response of Strombidium acuminatum strain BJLSC grown on 2 diets: 1) the diatom7halassiosira pseudonana (solid circles) and 2) a 1:1:1 combination of the three flagellates lsochlysis galbana,Chroomonas sauna and Rhodomonas lens (open circles). The lines represent the Michaelis-Menten fit (Eq. 1,Chapter 3) to the two data sets: the diatom (solid line) and the three flagellates (broken line). See Table 6.1 forthe parameters and estimates of error associated with these curves. Inset, same as above but the Rmyconcentrations have been converted to carbon using: carbon (pg) = 0.109 * live cell volume(Am')991(Montagnes et al. submitted).86,. 8.....10000 20000 30000 40000 50000Prey concentration (# mC1)Fig. 6.2. The functional response of Strombidium acuminatum strain BJLSC grazing on a 1:1 mixture of thediatom Thalassiosira pseudonana and 5 Am fluorescent beads. The line is the Michaelis-Menten fit to the data(Eq. 2, Chapter 3). See Table 6.1 for the parameters and estimates of error associated with this curve.87•-^ -••_ --1 - - ---^_•• •- --• - - - '0 ------------------ - _ _ _ _ -- - - -^ ---- .....-. - - '0 - - •^•^•^•^-•••1^I^10^10000 20000 30000 40000 50000--1Prey concentration (# mL )0E350 300;..Ia)+.7 25ctSa) 204-i—)15ul••••••4r.4 1 0Q.)C.)0+70 0Q.)C.);-,a)0.i88Fig. 6.3. The change in position (in the water column vs. on a surface) of Strombidium acuminatum strainBJLSC at different prey (Thalassiosira pseudonana) concentrations. The solid line is the regression through thedata and the broken lines are the 95% confidence intervals around the regression.8 9CHAPTER 7GROWTH RATES OF Strombidium capitatum STRAIN APAG AS A FUNCTION OFFOOD CONCENTRATIONIntroductionThis species is presented last because only growth rate data were obtained for it.Although S. capitatum was isolated on two separate occasions, neither culture ingestedsurrogate particles. However, a reasonable numerical response was obtained for it.PART I GROWTH RATES: Estimating the Numerical Response1.0 General MethodsStrombidium capitatum strain APAG was cultured on all combinations of I. galbana,C. sauna and R. lens. The ciliate grew on a number of prey combinations but did not surviveon single species (Table 7.1). An experiment was conducted to determine the numericalresponse of S. capitatum grazing on a 1:1 ratio of the two flagellate species Isochlysis galbanaand Chroomonas sauna (which gave maximum growth rates, Table 7.1), at foodconcentrations ranging from 0 to 105 cells mL-1. The ciliates were maintained by the semi-continuous culture method described in Chapters 3 and 4.Isochlysis galbana and C. sauna were maintained in log phase in 150 mL flasks byserial dilution at an irradiance of 50-100 Amol photons m-2 5-1 at 16-17°C. Prior to theexperiments, S. capitatum was maintained in enriched natural seawater with !. galbana and C.salina at 16-17°C on a 14:10 light:dark cycle at 25-50 gmol photons m-2 s-1. The ciliateswere collected during log or stationary phase; prey concentrations, which the ciliates werecollected from, were not maintained at a constant concentration.During the experiment, ciliates were grown in 6-well 20 mL tissue culture platescontaining 10 mL of prey medium. Prey concentrations were prepared and enumerated asdescribed in Chapter 3. Typically, each concentration was replicated three times. The semi-continuous culture method was similar to that described in Chapter 3 except for the followingmodifications.9 0After the ciliates were placed in the tissue plate they were kept in the dark for 48 h(rather than 24 h) at 16-17°C to allow the ciliates to acclimate. After the acclimation perioddaily transfers were continued for 5 days, and growth rate was calculated as the 5-day average.The numerical response data were fitted to a modified Michaelis-Menten model (Eq. 1,Chapter 3) using an iterative fit.A total of 22 treatments (prey concentrations) were examined in two consecutiveexperiments; treatments ranged from 0-105 prey mL-1.A test of methodology was conducted to determine the stability of the preyconcentrations. Prey concentration was measured after the initial dilution (prior to adding theciliates) and after the ciliates had been removed (24 h later). Prey concentrations varied little(-15%) over the 24 h period. Thus, the semi-continuous culture method described aboveappeared to maintain constant prey levels.2.0 Results and DiscussionThe growth rate followed a rectangular hyperbolic response between 1x104-6x104 preymL-1 (Fig. 7.1). At concentrations below 104 (0 and 5x102) and above 6x104 (8x104 and105) prey mL-1 mortality occurred within the first 48 h (the acclimation period). After theacclimation period at concentrations below 1.4x104 prey mL-1, mortality occurred,presumably due to starvation. Above concentrations of 4x104 growth was also inhibited.Similar inhibition of growth at high concentrations was seen in Chapters 3 and 5 and by Verity(1985). Equation 1 (Chapter 3) was fitted to points from 104-6x104 prey mL-1 (Fig. 7.1).The parameters of this equation and their estimates of error are presented in Table 7.2.3.0 Aspects of the Growth Rates and Their Ecological RelevanceIn general, S. capitatum divided more than once a day (1.5 d-1 if the curve in Fig. 7.1is extended beyond the data) at saturating prey concentrations; its threshold concentration was1.4x104 prey mL-1; and it had a maximum numerical response between 2x104-3x104 prey91mL-1. There was also a decrease in growth rate with food concentration above 4x104 preymL-1.This ciliate is typically mixotrophic (Montagnes et al. 1988b, Stoecker and Silver1990). However, since growth rate was positive in the dark, it is not an obligate mixotroph,like Laboea strobila (Stoecker et al. 1988, Putt 1990). This seems to contradicts the findingsof Stoecker and Silver (1990) who suggested that S. capitatum requires chloroplasts forcontinued growth. However, I constantly provided the ciliate with new autotrophic prey,which had been grown in the light. Thus, S. capitatum may not be an obligate mixotroph butmay require chloroplasts in its diet, as has been suggested by Stoecker and Silver (1990).Possibly, the ciliates would have survived at lower prey concentration if the they had beengrown in the light.PART II GRAZING RATES4.0 A Discussion of Grazing Experiments.Grazing experiments using 5 gm latex beads, similar to those described in Chapter 4,were conducted with strain IA. However, this ciliate did not ingest beads at any of 8concentrations ranging from 1.7x104 to 7x104 mL-1; these concentrations covered the range inwhich S. capitatum grew. Further, at a later date, a second clone of S. capitatum wasestablished and grown on I. galbana and C. sauna. It was fed fluorescently stained C. sauna(see Chapter 5), but it did not eat this food analog either. It is not known why this ciliate didnot ingest the beads or stained prey. Regardless, no grazing rate measurements were possible.92Table 7.1. The relative growth rate of Strobilidium capitatum strain APAG over a 7 day period. The ciliateswere grown on a 14:10 light:dark cycle at 16-17°C on all combinations of the three prey: Isochrysis galbana,Chroomonas sauna, and Rhodomonas lens (the initial combined prey concentration was 3x104 cells ml.).Growth rates were qualitative estimates of increase in numbers: cells died, -; cells grew slowly, +; cells grew,+ +; cells grew quickly, + + +.PREY^ GROWTH RATECONCENTRATIONI. galbanaR. lensC. saunaI. galbana & R. lensI. galbana & C. saunaC. sauna & R. lens1. galbana, C. sauna & R. lensDAY 7+++++9 3Table 7.2. Growth parameters for Strombidium capitatum strain APAG; parameters and estimates of error of thenumerical response curve presented in Fig. 7.1. See section 1.0 in Chapter 3 for the equation used. Values are inunits of numbers mL-1 for concentrations (k ,x'), and d -1 for growth rate (&m ).value^standard^coefficientdeviation of variationMichaelis-Menten fitAmax 1.074 0.983 92.53k 12370 6410 51.81x' 13860 898.7 6.4831 .00.8Pc!0.6cp 0.40.2O -0.2;•4— 0.4— 0.6940^10000 20000 30000 40000 50000 60000Prey concentration (# mLFig. 7.1. The numerical response of Strombidium capitatum strain APAG grown on a 1:1 ratio of the twoflagellates Isochtysis galbana and Chroomonas sauna. The circles represent growth rates and the line representsthe modified Michaelis-Menten fit to data (Eq. 1, Chapter 3). See Table 7.1 for the parameters and estimates oferror associated with this curve.9 5CHAPTER 8AN EXAMINATION OF THE FUNCTIONAL AND NUMERICAL RESPONSES OFOLIGOTRICHSIntroductionIn the past chapters I investigated the numerical and functional responses of 5 ciliatespecies. These responses were isolate specific (Figs. 8.1 and 8.2), ranging from the slowlygrowing and grazing Strombidium acuminatum strain BJLSC to the rapidly growing andgrazing Strobilidium spiralis strain IA. Now I will compare my results with similar work byothers (Figs. 8.1 and 8.2, Tables 8.1 and 8.2) and evaluate my findings. There was no"typical" ciliate response to varied food concentrations (Figs. 8.1 and 8.2). This implies thatciliates cannot be considered a single functional group for food web modeling. The maximumdifferences between parameters (see Tables 8.1 and 8.2) were often greater than thedifferences suggested by the coefficients of variation of these parameters (Table 8.3). Thus,while some differences were undoubtedly artifactual, others were probably real. In thefollowing sections I: 1) suggest reasons for variation in response between species and studies;2) examine the effect of scaling growth and grazing responses to determine if a common sizeand/or temperature dependent response might be invoked for all ciliates; and 3) present threegrowth and grazing responses which were used to model ciliate blooms (Chapter 9).1.0 REASONS FOR RESPONSE VARIATION1.1 Methodological Biases and the Use of Steady State Measurements1.1.1 Methodological BiasesThe variation in responses (Figs. 8.1, 8.2) can be partially attributed to methodologicalerrors in data acquisition and manipulation. Errors in estimating prey abundance (see Fig.3.2), cell volumes, and carbon quotas (Verity et al. 1993, Montagnes et al. submitted) wouldcause several fold errors on the independent axis. Error on the dependent axis might occur, asgrowth rate may not be constant over the incubation period (see Chapter 3).It may also be inappropriate to universally apply the modified Michaelis-Mentenkinetics to the data (Chapter 3 Eqs. 1 and 2). This model can be a good predictor of both9 6functional and numerical responses and is based on sound theoretical mechanisms (e.g. Holling1959, Fenchel 1986, Appendix 2) which have been supported by direct observation (Taniguchiand Takeda 1988). However, these mechanisms vary due to external factors. The modelassumes that predators are non-selective and have a constant encounter rate. Both theseassumptions can be violated. Oligotrichs can be selective and may alter their behaviour(Blackbourn 1974, Stoecker 1988, Taniguchi and Takeda 1988, Verity 1991a). The modelalso assumes that the gross growth efficiency is constant, which it may not be (Stoecker andEvans 1985, Verity 1985). Further, my results indicate that: 1) there is deviation from thehyperbolic response at high food concentrations, indicating inhibition (see Chapters 3, 5, and7, also see Verity 1985); 2) there may be switching in behaviour at lower food concentrations(see Chapter 4); and 3) Eq. 1 may not be applicable to mortality rate estimates (see Chapter3). Finally, piecewise regression (Blackman kinetics) may, at times, fit functional andnumerical response data better (Condrey 1982, Condrey and Fuller 1985). Thus, applyingEqs. 1 and 2 (Chapter 3) indiscriminately to all ciliates is innapropriate. However, given that:1) Michaelis-Menten-like kinetics are a first approximation of grazing and growth responsesfor ciliates (see Holling 1959, Spain 1982, Fenchel 1986, Taniguchi and Takeda 1988,Legovic 1989, Appendix 2); 2) my data follow a rectangular hyperbolic response (rather than apiecewise, Blackman, response); and 3) from the parameters I measured, there was no meansto determine the reasons for the deviant ciliate behaviour, I consider Eqs. 1 and 2 (Chapter 3)to adequately represent the responses for use in my food web model (see below).1.1.2 The Use of Steady State MeasurementsThe time over which experiments are run and the extent of the acclimation period canaffect the results of both grazing and growth experiments (Verity 1985). Ciliates exhibitbehavioural changes to variation in food concentration over minutes (e.g. Verity 1991b), butthey exhibit morphological and physiological changes over hours to days (Lynn 1975, Lynn etal. 1987, Fenchel and Jonsson 1988, Fenchel 1990). Failure to acclimate ciliates, over aperiod of days, to steady state conditions may result in variable an inaccurate growth andgrazing estimates. Because of the difficulty of maintaining such conditions, few studies have97established steady state estimates. Other than chemostat methodologies (e.g. Scott 1985),small containers and regular transferring of ciliates seems to be the only means to maintainciliates at a constant prey concentration for extended periods. By virtue of their large surfacearea and shallow depth, small containers could increase microbial growth, facilitate the settlingand accumulation of prey (on the bottom), and provide surfaces for cells to collide into (andpossibly lyes). The 10 mL containers I used may therefore misrepresent natural conditions,and these potential errors should be considered when evaluating my results. However themethodology I used was the only practical means to acclimate and maintain these ciliates overextended periods. Therefore, the precision of such estimates is at present constrained byculturing logistics.My estimates of growth responses may be better than previous ones on marineplanktonic ciliates, as they were conducted for several days under steady-state conditions.Further, most studies of ciliate numerical responses have likely underestimated the thresholdfood concentration (the concentration where growth equals mortality), as ciliates may notimmediately die at sub-threshold levels. For instance, Verity (1985) acclimated two tintinnidsfor 12-18 h prior to growth experiments at several food concentrations, but there was no (orlittle) indication of a threshold concentration (Verity 1985, Fig. 3). In another case,Heinbokel's (1978) response curves on tintinnids, which have been used in several models(e.g. Frost 1987, Frost and Franzen 1992), were based on ciliates that were only acclimated"overnight". In some of my experiments the threshold concentration may have beenoverestimated due to imprecision of prey concentration measurements (see previous chapters).However, these imprecise measurements should have been overcome to some extent by themultiple measurements that were made near the threshold levels (see previous chapters). SinceI also made extensive measurements below the threshold concentrations (measurements whichother methods are unable to make) these studies should provide more realistic (higher)estimates of mortality rate (see Fig. 8.1a and b, insets) and would be more appropriate formodeling systems where the ciliates deplete their food source.9 8My grazing rate estimates (except Chapter 4) are like those obtained by others, whichwere conducted on ciliates acclimated for short periods (several hours, see previous chapters).I expect these estimates are biased. For instance, the grazing rates may be overestimated atlow food concentrations since these experiments often used ciliates maintained at higher foodconcentrations (Chapters 4 and 5). These ciliates would be larger and more active than thosemaintained on leaner diets, and therefore they might graze at higher rates. The problem ofobtaining grazing estimates on prey-acclimated ciliates lies in the difficulty of maintaininglarge ciliates numbers (needed for grazing experiments) at constant food concentrations forextended periods. Therefore logistics limit the resolution of accurate grazing rates.1.2 Use of Surrogate Particles to Determine Grazing Rates.Surrogate particles, both beads and stained prey, have been extensively used to measureprotozoan grazing rates in the laboratory and field (e.g. Jonsson 1986, Pace and Bailiff 1987,Stoecker 1988, Rublee and Galagos, 1989, Gonzalez eta!. 1990, McManus and Olcubo 1991,Putt 1991, Sherr and Sherr 1993), and there have been criticisms of this technique (e.g.ciliates may have prey preference, Stoecker 1988, Putt 1991; fixation may cause egestion ofprey, Sieracld et al. 1987; particles may not be monodispersed, Pace and Baliff 1987, but seePutt 1990 and McManus and Olcubo 1991). Based on a literature review and an assessment ofavailable techniques I chose surrogate uptake to estimate grazing (see previous chapters). Inretrospect, I now believe this methodology inadequately measured grazing rates, as there wasconsiderable variation in my data (Chapters 3-6). Analysis of my data also indicated that thevariability in grazing rates measured by surrogate uptake can significantly over- orunderestimate the predicted grazing rates.Assuming my growth rate estimates are accurate, the ingestion rate required to sustaingrowth can be calculated using the bioenergetic equation:Ingestion = (growth + respiration)/ assimilation efficiency^(3)Respiration for ciliates and flagellates is dependent on cell size and follows the relation: log ioR (nL 02 cell-1 111) = log Vol (.4m3) * 0.75 - 4.09 (Fenchel and Finlay 1983). Thisrespiration rate can be converted to carbon (ng) by converting nL 02 to moles of 02,9 9converting this to moles of CO2, assuming a respiratory quotient ranging from 1 (forcarbohydrate) to 0.7 (for fat) (Schmidt-Nielsen 1980), and then converting to mass of carbon(multiplying by 12 g C mole-1). Assimilation efficiencies, ranging from 0.65 to 0.9, havebeen established for protozoa (Laybourn-Parry 1984, Stoecker 1984, Verity 1985), althoughthe accuracy of these measurements may be suspect (Caron and Goldman 1993). Using mygrowth rates and cell volumes (cell volume was determined from 30-100 growing cells, seeTable 8.1) and a range of respiratory quotient and assimilation efficiencies and Eq. 3, I havecompared the predicted ingestion rates to my direct estimates, for the ciliates examined inChapters 3-6 (Fig. 8.3). This analysis indicates that the surrogate particle method: 1)underestimated the calculated (bioenergetically determined) grazing rate by up to an order ofmagnitude for Strombidinopsis acuminatum and Strombidium acuminatum; 2) overestimatedcalculated grazing by up to four-fold for Strobilidium sp. strain JERC; and 3) comparedreasonably to the calculated grazing rate of Strobilidium spiralis.The gross growth efficiency (carbon incorporated/carbon consumed, GGE) of ciliates,which is typically 0.3-0.6 (see Caron and Goldman 1990 for a review) can also be used todetermine the accuracy of grazing estimates (although the GGE can vary with foodconcentration, Verity 1985, 1991). The observed growth rates of the ciliates described inChapters 3-6 were divided by their observed and predicted (by Eq. 3) ingestion rates (Fig.8.4). In all 4 cases the predicted GGE plateaued near 0.4-0.5 while in 3 of the 4 cases theobserved GGE was far from this range. Thus, predicting grazing rates from growth ratesapears to be a more accurate than using surrogate particles to estimate grazing.Grazing rates can be affected by food quality (type of prey analog). Oligotrichs can beselective in their feeding (Blackbourn 1974, Stoecker 1988), and this will affect grazingestimates. Strobilidium spiralis consumed latex microspheres but not stained algae (Chapter4), Strobilidium sp. strain JERC consumed stained algae but not microspheres (Chapter 5), andStrombidium capitatum consumed neither (Chapter 7). These data support the evidence ofother researchers that prey analog choice influences grazing estimates, and it indicates thatthere is not a single "best" prey analog to use (e.g. Putt 1991). Thus, my grazing estimates100from surrogate uptake are likely inaccurate, and I have used Eq. 3 to establish grazing rateresponses in the model presented in Chapter 9.1.3 Physiological-Behavioural VariablesTo simplify modeling, it would be convenient if all ciliates behaved in a similarfashion, but they do not (see Chapters 3-7 and Figs. 8.1 and 8.2). Still, it is possible thatsome factors might be used to scale growth responses and allow us to simplify the "ciliate"component of models. For instance, differences in ciliate cell size, prey type, and ambienttemperature might affect the ciliates and alter the growth responses (Muller and Geller 1993).Increases in temperature by 10°C may raise ciliate responses by 2-2.5 fold (i.e. Q10 =2-2.5, Fenchel and Finlay 1983, Fenchel 1987, Caron et al. 1990). This relation can be statedas:log io (ratei/rate2) = log10Q10 * {(tempi-temp2)/10}^(4)However, variation exists in Qio estimates (Muller and Geller 1993, Verity 1985). Therelationship between physiological rates and temperature may not be log-linear and maydeviate significantly at high temperatures, where it is often strongly negative (Aelion andChisholm 1985, Caron et a/. 1990). Thus, differences in ambient temperature may explainsome of the response variations seen in Figs. 8.1 and 8.2, but caution should be applied whenusing Q 10 scaling.Cell volume is another parameter used to scale growth rate. Dogma suggests that largecells grow slower than small ones, but there is variation, as ciliates can have "r" or "K" likestrategies (Taylor and Shutter 1978, Turley et a/. 1986). Generally, protozoa follow theallometric response A = aVb, where it (d-1) is growth rate, V (tad) is volume and "a" (d-1V-1), and "b" are constants (Taylor and Shutter 1981, Fenchel and Finlay 1983). At 20°C thevalue of "a" ranges from 20 to 40 (1-1 V-1 (from Fig. 1, Taylor 1981 and Fig. 3, Fenchel andFinlay 1983) while "b" is approximately -0.25 (Taylor 1981, Fenchel and Finlay 1983,Fenchel 1987, Muller and Geller 1993).Using the above relationships, differences in ambient temperature and cell volume maybe accounted for. I have scaled the growth responses presented in Table 8.1 and Fig. 8.1b to10120°C (using Q10 = 2) and standardizing growth rate by solving for "a" in the allometricequation = aVb (using b = -0.25) (Fig. 8.5). Note that applying these transformations tothe entire curve (rather than only to the maximum rate) may be inappropriate, as theserelationships are typically calculated from maximum growth rates.After scaling for temperature and cell volume, some conformity between the ciliatesexisted, but differences between the responses remained (Fig. 8.5). The transformedmaximum growth rate generally fell within the previously observed range of 20-40 d ^,with most of the ciliates exhibiting a volume specific growth of near 20 (Fig. 8.5, inset). Notethat two of the three studies on Strobilidium spiralis (curves 2 and 8 vs. 9) gave a highertransformed growth rate (-40 d^m-3, Fig. 8.5, inset). The response of three medium sizedciliates were also similar (Fig. 8.5, curves 6, 9, and 10). Thus, although differences inambient temperature and cell size may account for some of the variation between responses, itis likely that other factors alter the response.I have also scaled the grazing rates (Fig. 8.2a, Table 8.2) by dividing them by thesquare of the sum of the prey and predator radii ([r ciliate+rprey]2) and using a Qio = 2(scaled to 20°C). The former manipulation was based on the assumption that ingestion isdirectly proportional to prey-predator encounters, and such encounters will vary directly withthe cross-sectional areas of both prey and predator. These manipulations changed the relativeposition of the curves (cf. Fig. 8.2b vs. Fig. 8.6) but indicated that size and temperature didnot control grazing rate.Thus, differences between ciliate responses are not solely due to variation in ambienttemperature and cell size. As indicated above (section 1.1), methodological biases likelycontribute to these differences. However, species specific factors such as behaviouralresponses or possibly physiological responses to food processing may also cause thedifferences. For instance, swimming speed (Buskey and Stoecker 1988, 1989), prey rejection(Blackbourn 1978, Taniguchi and Talceda 1988), filtration efficiency (Bernard andRassoulzadegan 1990), vacuole formation (summarized in Capriulo 1990), and prey detection(Verity 1988) could affect the ingestion rate.102Variation in food quality within and between experiments may have affected the growthrates presented in Figs. 8.1 and 8.2. Nutritional quality differs between and within preyspecies. This was discussed in Chapter 6, but see the data in Chapters 4, 5, and 7. The extentto which different prey species affect growth was qualitatively examined in Chapters 4-7 andhas been shown by others (Capriulo 1990, Caron and Goldman 1990, Caron et al. 1990).Initially, I examined the suitability of several algal species for maintaining ciliates:Thalassiosira pseudonana, Pavlova lutheri, Pyramimonas orientalis, Tetraselmis suecia,Tetraselmis sp., Croomonas sauna, Isochlysis galbana, Heterocapsa triquetra, Scrippsiellatrochoidea, Micromonas pusilla, and Rhodomonas lens. These all proved to be adequate foodfor some ciliates, but not for all (data not shown). Ultimately, I chose to use the four speciesThalassiosira pseudonana, Isochrysis galbana, Croomonas sauna and Rhodomonas lens, asthese algae stimulated growth for most ciliates that I isolated, and have been shown by othersto be good food items (e.g. Repak 1983, Verity and Villareal 1986).Variation of oligotrich growth rates may be in part due to food preferences (Verity1991a and b). Not only does prey type affect ciliate growth but there are species specificgrowth responses to prey type (cf. Tables 4.1, 5.1, 6.1, and 7.1). My response curves werebased on prey which illicited a maximum growth response, but I only examined 3-4 preyspecies. Possibly, other prey would produce higher rates.From these studies, I suggest that food type is the primary factor affecting ciliatespecies composition and distribution in the plankton. Concomitantly, prey abundance is theprimary factor affecting ciliate growth rate and therefore abundance in the plankton. Theformer suggestion may appear as an obvious platitude, but it is a concept often overlooked inplankton models. Recognition of such species specific responses should allow us to betterassess the role of ciliates in the plankton. It also suggests that monospecific ciliate bloomscould be caused by monospecific prey patches, if the optimum ciliate and prey combinationcoincide. Such an event is explored in Chapter 9.1031.4 Species DifferencesVariations in growth and grazing rates are likely species specific, as implied by Figs.8.1 and 8.2. Such differences in ciliate numerical responses may be due to a number offactors: 1) clonal decline (see Chapter 3 and 4); 2) clonal variation (see Chapter 3); 3)inclusion of both planktonic and semi-benthic species in analyses (both Strombidiumacuminatum strain BJSCH and Strombidium sulcatum from Fenchel and Jonsson, 1988, Table8.1 are likely semi-benthic); 4) inclusion of mixotrophic and non-mixotrophic species(oligotrichs range from non-mixotrophic to obligate mixotrophic species, e.g. Stoecker et al.1988, Putt 1990, Stoecker and Silver 1990, Stoecker and Michaels 1991); and 5) tendenciestoward r and K life strategies (Taylor 1978).It would be presumptuous to pursue a discussion of species variations based on myresearch and the limited studies on planktonic oligotrichs. However, two comparisons fromthe data presented in Fig. 8.1 are exemplary of the response differences from similar ciliates.Curves 2, 8 and 9 in Fig. 8.1 are all from different studies on ciliates that appear to beStrobilidium spiralis (see Chapter 2), but all three responses are different. Further, curves 2and 3 are from my studies on two species of Strobilidium of different size but similarmorphology (see Chapter 2), and they are different. Thus, lumping ciliates into a singlefunctional group, as has been done for past models, ignores the diversity of responses thatciliates may exhibit. This could inaccurately represent their role as microzooplankton.2.0 Ciliate Growth and Grazing Responses for Food Web ModelingVariation in grazing and growth responses exists between planktonic ciliates. Ratherthan devise a single numerical response to represent all ciliates, I have proposed three growthresponses which cover a range (Fig. 8.7, Table 8.4): species A was based on (parameters wereadjusted slightly, see previous chapters and Table 8.4) the growth rate of Strombidinopsisacuminatum strain SPJSC; species B was based on the growth rate of Strobilidium spiralisstrain IA; and species C was based on the growth rate of Strobilidium sp. stain JERC. Thegrazing rates of these 3 ciliates were calculated as a function of growth rate, based on Eq. 3, a104respiratory quotient of 0.80, and an assimilation efficiency of 0.75. The latter two parametersgave a response near the middle of the expected range, see Figs. 8.3 and 8.4 (although arespiratory quotient of 0.9 has been suggested for tintinnids, Verity 1985). These 3 ciliateresponses cover a large portion of the ranges in growth responses presented above and are usedto examine the impact of ciliates on nanoplanlcton blooms (Chapter 9).105Table 8.1. A comparison of numerical response constants for 10 oligotrichs. The parameters are the same asthose described in Eq. 1, Chapter 3. Cell volumes (pm 3) and densities (# mL-1 ) are presented in bold type. Cellcarbon quotas (pg C cell-1) and carbon concentrations (ng C mL-1) are presented in normal type. Prey carbonwas estimated using the conversion: carbon (pg) = 0.109 * live cell volume (Am 3)0.991 (from Montagnes et al.submitted), except for marked cases. 1 A value obtained from a figure rather than the text. 2Recalculated fromraw data graphically presented in Verity (1991a) but provided by the author. 3Calculated using 0.1 pg C Am -3(Borsheim and Bratbak 1987). 4Ciliate carbon estimated using 0.148 pg C Arn-3 from Putt and Stoecker (1989)except in marked cases (*) where it was presented in the literature.SPECIES CILIATEvolume &carbon(jim3)(pg C celrl)It max(c1-1) (# inL-1)(ng C mL-1)x'm11)(ng C mL-1)SOURCE °C PREY SPECIES PREYCARBON(pg CPREY RANGECONC.niL-1)(ng C mL4)Strombidinopsis 117193 0.985 4091 1567 This study 16 Thalassiosira 5.9 400 to 90000acuminatumstrain SPJSC17368 24.30 9.31 pseudonana 2.38 to 534Strobilidium spiralis 110000 1.85 19070 10230 This study 16 Chroomonas sauna 32.4 0 to 95000strain IA 16302 617.87 331.45 0 to 3080Strobilidium sp. 19635 0.721 10580 4075 This study 16 Chroomonas sauna 19.7 1700 to 57000strain JERC 2910 208.27 80.22 Isoclnysis galbana 33.5 to 1120Strombidium sp. 28575 0.612 5812 2709 This study 16 7halassiosira 6.0 100 to 50000strain BJLSC 4235 34.52 16.09 pseudonana 0.59 to 297Strombidium 64140 1.07 12370 13860 This study 16 Chroomonas sauna 19.7 14000 to 60000capitatumstrain APAG9506 243.50 272.90 Isochrysis galbana 275 to 1181Strombidium sp. 6500 21 6500 0 Fenchel & 20 Pteridomonas 8.03 0 to 20000963 524 0 Jonsson 1988 danich 0 to 160Eutintinnus pectinis 18000 1.54 24530 1390 Heinbokel 18 Isochrysis galbana 10.0 0 to 400002668 24.530 13.90 1978 Monochrysis lutheri 0 to 400Dunaliella tertiolectaStrobilidium (=-- 150000 1.18 5400 730 Jonsson 1986 12 Pyramimonas sp. 10.2 0 to 50000Lhomanniella )spiralis22230 54.81 74.10 0 to 507Strobilidium spiralis 114942 1.79 11915 5025 Verity 1991 20 Isochrysis galbana 3.5 7000 to 12000003200* 41.7 17.56 25 to 420Strombidium 40000 0.84 6300 800 Jonsson 1986 12 Pyramimonas sp. 10.2 0 to 100000reticulatum 5928 63.95 8.12 0 to 1015106Table 8.2. A comparison of functional response constants for 8 oligotrichs. The parameters are the same as thosedescribed in Eq. 2 (Chapter 3) or as modified in Eq. 1 (Chapter 3), which allows for a non-zero x-intercept. Cellvolumes (=3) and densities (# mL-1 ) are presented in bold type. Cell carbon quotas (pg C cell -1 ) and carbonconcentrations (ng C mL-1) are presented in normal type. Prey carbon was estimated using the conversion:carbon (pg) = 0.109 * live cell volume (Am3 )0.991 (from Montagnes et al. submitted), except for marked cases.1 A value obtained from a figure rather than the text. 2Recalculated from raw data graphically presented in Verity9(1991a) but provided by the author. •No temperature was presented but 20°C was assumed.Strombidinopsis^101.00acuminatum^0.60strain SPJSCStrobilidium spiralis 45.45strain IA^1.47Strobilidium sp.strain JERCStrombidium sp.strain BJLSC41.21 butmax=2441.34 butmax = 0.784.210.025Eutintinnus pectinis 160.16SPECIES^g ma,x(#( pg C 11-1)(# mL-1)(pg C m1:1)x'(# mL-1)(pg CSOURCE °C PREY SPECIES PREY(pg C ce11-1)PREY RANGECONC.(pg C426300 0 This study 16 Thalassiosira pseudonana 5.9 1800 to 700002532 0 and 5 gm beads 10.7 to 4164028 0 This study 16 Chroomonas sauna and 5 32.4 1000 to 95000130.51 0 gm beads 32 to 30803865 0 This study 16 Chroomonas sauna and 32.4 100 to 48000125.2 0 5 gm beads 3.24 to 15553401 0 This study 16 Thalassiosira pseudonana 5.9 450 to 3000020.203 0 and 5 gm beads 2.67 to 1781628 1000 Heinbokel 18 Isochrysis galbana 10.0 1000 to 4000016.28 10 1978 Monochrysis lutheri 10 to 400Dunaliella tertiolecta4521 0 Stoecker 1988 15 Hetrocapsa triquetra 110 0 to 250049.76 0 0 to 275198492 42292 Verity 1991 20 Isochrysis galbana 3.5 7000 to 120000069.47 14.80 25 to 4204681 336 Bernard & 20? Dunaliella minuta 3.3 0 to 120000015.21 1.09 Rassoulzadegan 0 to 175201990Favella sp.^13.5511.49Strobilidium spiralis 169.0020.593Strombidium^34.40sulcatum 0.11107Table 8.3. A comparison of error estimates of numerical and functional responses (coefficients of variation) fromdata of the five species presented in Chapters 3-7. The symbols (Amax, k, x', Gm, k) are the parameters usedto fit data to Eqs. 1 and 2 (Chapter 3). No data (--).Coefficients of variation ofgrowth parameters^grazing parametersSpecies^gmax k x' Gma)c kStrombidinopsisacuminatum4.7 15.0 10.0 182.9 2169Strobilidiumspiralis13.9 15.0 9.0 6.1 33.7Strobilidium sp.strain JERC15.2 32.2 13.4 7.4 22.0Strombidiumacuminatum strain BJLSC11.6 21.1 12.4 17.7 81.1Strombidiumcap itatum27.8 51.8 6.5108Table 8.4. Parameters used to establish 3 different ciliate numerical response curves (for ciliates A, B, C) usingEq. 1, (Chapter 3). These are use in the model of ciliate-prey population dynamics in Chapter 9. The symbols(Amax, k, x') are the parameters used to fit data to Eq. 1 (Chapter 3). Values are in units of ng carbon mL I forconcentrations (kM. x'A ), and d-1 for growth rate (A^). Values in bold are the number (mL -1) of 8 Am./2(diameter) particles that would be made from the respective carbon value (carbon particle- 1 was calculated from:carbon (pg) = 0.109 * live cell volume (Am 3)0.991 , from Montagnes et a/. submitted). These particleconcentrations were used in the model discussed in (Chapter 9).species Amax x'A 0.99 24 10921 380B 1.85 618 33124436 13014C 0.72 208 808143 310540010950000^ 100000Prey concentration (# mL )930.1—1.5—2.00.0—0.1—0.2200 30000^500^1000^1500^2000—1Carbon (ng mL )Fig. 8.1. The growth rates (numerical responses) of 10 oligotrichs in response to varied prey concentration asprey numbers (panel a) and prey carbon (panel b). The lines are the modified Michaelis-Menten fits (Eq. 1,Chapter 3) to the data; see Table 8.1 for the parameters associated with these curves and particulars associatedwith each study. The solid curves were obtained from this study and represent: Strombidinopsis actuninatumstrain SPJSC (1); Strobilidium spiralis strain IA (2); Strobilidium sp. strain JERC (3); Strombidium acuminatumstrain BJLSC (4); and Strombidium capitation strain APAG (5). The broken curves were obtained from theliterature and represent: Strornbidium sulcatum (Fenchel and Jonsson 1988) (6); Eutintinnus pectinis (Heinbokel1978) (7); Strobilidium^Lohmanniella) spiralis (Jonsson 1986) (8); Strobilidium spiralis (Verity 1991a) (9);and Strombidiurn reticulatum .(Jonsson 1986) (10). The insets are an enlargement of the region where the curvesintercept the zero growth (threshold) concentrations; see Table 8.1 for the value (x') of these intercepts. Note,curve 9 is not in the insets as it does not extend to A=0.2500^3000a.-*..............................................6040c.)2011020000^40000 60000^80000^100000 120000Prey concentration (# mL )••••tt1.1••■••1.51.0i .. ........... .50^500^1000^1500^2000^2500^3000^3500Prey concentration (ng C mL - 1)Fig. 8.2. The grazing rates (functional responses) of eight oligotrichs in response to varied prey concentration asnumbers (panel a) and prey carbon (panel b). The lines are the Michaelis-Menten fits (Eq. 2, Chapter 3) to thedata; see Table 8.2 for the parameters associated with these curves and particulars associated with each study.The solid curves were obtained from this study and represent: Strombidinopsis acuminatum strain SPJSC (1);Strobilidium spiralis strain IA (2); Strobilidiutn sp. strain JERC (3); and Strombidium acuminatum strain BILSC(4). The broken curves were obtained from the literature and represent: Eutintinnus pectinis (Heinbokel 1978)(5); Favella sp. (Stoecker 1988) (6); Strobilidium spiralis (Verity 1991a) (7); and Strombidium sulcatum(Bernard and Rassoulzadegan 1990) (8).T 1.5C.)C.)0.50.0 o1.51.0 -0.5^,10.0 ^''00-1 ''' 2000^00.3 ^111•=0500 1000 1500 2000rci00ci0.80.6-i 0.20.4 -0.10.2^' ' '^' '0.0  ^0.00^500 1000 1500 2000^0Prey concentration• II1W; 500 1000 1500 2000--(ng C mL1Fig. 8.3. A comparison of observed functional response rates (dotted lines) and those predicted from equation 3(see section 1.2), using growth rates predicted by values presented in Chapters 3-6 and Eq. 1, Chapter 3. The 3predicted rates used the following parameters: a respiratory quotient of 1 and an assimilation efficiency of 0.8(short dashes); a respiratory quotient of 0.8 and an assimilation efficiency of 0.75 (medium dashes); a respiratoryquotient of 0.75 and an assimilation efficiency of 0.7 (solid lines). Data were obtained from: Strombidinopsisacuminatum strain SPJSC, Chapter 3 (panel a); Strobilidium spiralis strain IA, Chapter 4 (panel b); Strobiliditunsp. strain JERC, Chapter 5 (panel c); and Strombidium acuminatum strain BJLSC, Chapter 6 (panel d).112C)10 0.8rn8 - 0.76 - 0.60C.)4......^- 0.52 - 0.40.30.8 0.20anCti0.4 0.1C) 0.0 0.0\.,^0^500 1000 1500 2000^0^500 1000 1500 20000.6 1 I^i404 ^.4..... .... __ ---------1^606610.2 r ..... 0.4(31.1^ri............................. - 0.20.0^I^".^' I 1 0.00^500 1000 1500 2000^0^500 1000 1500 2000Prey concentration (ng C mLFig. 8.4. A comparison of gross growth efficiencies (carbon incorporated/carbon consumed) based on observedgrowth and grazing data (dotted line lines) and those based on observed growth and predicted grazing rates(predicted grazing rate was determined using Eq. 3 (see section 1.2) and growth rates predicted by valuespresented in Chapters 3-6 and Eq. 1, Chapter 3). The three predicted grazing rates used the following parameters:a respiratory quotient of 1 and an assimilation efficiency of 0.8 (short dashes); a respiratory quotient of 0.8 and anassimilation efficiency of 0.75 (medium dashes); a respiratory quotient of 0.75 and an assimilation efficiency of0.7 (solid lines). Data were obtained from: Strombidinopsiis acuminatum strain SPJSC. Chapter 3 (panel a);Strobilidium spiralis strain IA, Chapter 4 (panel b); Strobilidium sp. strain JERC, Chapter 5 (panel c); andStrombidium acumincuum strain BJLSC, Chapter 6 (panel d).113cv)40"O 20a)4-2CC,—20 —ocu —4004-•oti —600E-4• • 81......, .^9501:1g. 407305 -:32,01 20574 '2.4 10oF1- •0 0500^1000 1500^2000^2500^300010Prey concentration (ng C mL )Fig. 8.5. The growth rate of 10 oligotrichs in response to varied prey concentration, corrected for differences incell volume and ambient temperature (to 20°C). The lines are the transformed modified Michaelis-Menten fits(see Fig. 8.1) to the data; see Table 8.1 for the particulars associated with each study and the text (Chapter 8section 1.3) for the specific transformations. The solid curves were obtained from this study and represent:Strombidinopsis acuminatum strain SPJSC (1); Strobilidium spiralis strain IA (2); Strobilidium sp. strain JERC(3); Strombidium acuminatum strain BJLSC (4); and Strombidium capitatum strain APAG (5). The broken curveswere obtained from the literature and represent: Strombidium sulcatum (Fenchel and Jonsson 1988) (6);Eutintinnus pectinis (Heinbokel 1978) (7); Strobilidium (= Lohmanniella) spiralis (Jonsson 1986) (8);Strobilidium spiralis (Verity 1991a) (9); and Strombidium reticulatum (Jonsson 1986) (10). The inset is acomparison of the transformed maximum growth rates predicted by Eq. 1, Chapter 3 (the untransformedmaximum growth rates are presented in Table 8.1).114............... 7.......... .... ........ ............•••••••••............................................................................................3********************************* 52A.61^ 1^ 10^20000 40000 60000 80000 100000 120000—Prey concentration (# inL 1 )Fig. 8.6. The grazing rate of 8 oligotrichs in response to varied prey concentration, corrected for differences incell size and ambient temperature (to 20°C). The lines are the transformed Michaelis-Menten fits (see Fig. 8.2)to the data; see Table 8.2 for the particulars associated with each study and the text (Chapter 8 section 1.3) for thespecific transformations. The solid curves were obtained from this study and represent: Srrombidinopsisacuminatum strain SPISC (1); Strobilidium spiralis strain IA (2); Strobilidium sp. strain JERC (3); andStrombidium acumination strain BJLSC (4). The broken curves were obtained from the literature and represent:Eurintinnus pecrinis (Heinbokel 1978) (5); Favella sp. (Stoecker 1988) (6); Strobilidium spiralis (Verity 1991a)(7); and Srrombidium sulcarum (Bernard and Rassoulzadegan 1990) (8).0.350.30r. 0.254.) 0.20s.ea0 .15• 0.10 500.05E-*0.00.................. 80^500^1000^1500^2000-Prey concentration (ng C mLI )Fig. 8.7. Three growth responses used to represent different ciliate types (A-C) (see section 2.0 for details).These responses are used to model ciliate-nanoplankton prey interactions in Chapter 9. The parameters for thesecurves are presented in Table 8.4.115116CHAPTER 9A MODEL OF CILIATE AND PHYTOPLANKTON POPULATION DYNAMICSIntroductionIn this chapter I investigate short-term and small-scale ciliate-phytoplanlcton bloomdynamics using the BASIC model presented in Appendix 4. The goal of this exercise was touse the data, collected in the previous chapters, to answer the following general questionsabout ciliate blooms. 1) Can ciliates control blooms of small (<10 Am) phytoplankton bygrazing them down over short periods (several days)? 2) Are ciliates a link or a sink forphytoplankton biomass (i.e. are they important in the transfer of biomass to higher trophiclevels)? 3) Can ciliates and copepods compete for the same food source? 4) Can differentciliate species cause different types of algal and ciliate blooms (i.e. are species differencesimportant)? I then examined a phytoplankton-ciliate-copepod food web, to determine howciliate blooms could influence the flow of carbon to copepods. These questions were designedto determine if ciliate blooms could be an important component of marine food webs.The phytoplankton and/or ciliate blooms I have considered are mono-specific, rapidincreases in numbers or biomass, visible as transient peaks (Legendre 1990, Lynn andMontagnes 1991). In coastal waters, short term phytoplankton blooms could be stimulated bya number of factors: tidal or wind mixing events, changes in irradiance, or by allochthonousnutrient inputs from terrestrial run off (Macias et al. 1985, Harris 1986, Legendre 1990).Such blooms probably do not exist for more than 2-3 weeks, as mixing processes (e.g. windand tides) dissipate them, and mesozooplankton populations (e.g. copepods) can respond tothem if they do persist. For the model, I have assumed that over periods of <20 days,phytoplankton blooms are not influenced by increases in mesozooplanIcton populations(Raymont 1983) and that advection of organisms and nutrients in or out of the system wouldbe negligible. These are the conditions where ciliate blooms might occur and where ciliatescould have a selective advantage over mesozooplankton due to ciliates' rapid growth rates.Blooms with these parameters are probably meters to kilometers in size and fall into the "fine"117scale patches described by Haury et al. (1978). Macicas et al. (1985) suggested that patches ofthis magnitude persist for 1-10 days and are dominated by "episodic predation stress".1.0 The Model (see Appendix 4)1.1 General DescriptionThe model simulated the population dynamics of ciliates and phytoplankton in a non-steady state situation, where the ciliates and copepods encountered a patch of water with adefined initial phytoplankton (algal) concentration. There were 5 types of organisms in thesystem (Fig. 9.1): algae, 3 ciliate species (A, B, C, see Table 8.4, Fig. 8.7), and a copepodspecies (data obtained from Frost, 1972 for Calanus pacificus, see section 1.3.2). The algaewere small (8 p.m in diameter) and during the simulation grew at a constant rate of /.4 = 0.71 , or = 0 d".1 when algal abundance reached its maximum level (see below). Algalmortality was by predation from the ciliates (see section 2.0, Chapter 8) and the copepods.Ciliate growth rates were species specific functions of algal concentration (as illustrated inprevious chapters). For all 3 ciliates, mortality was due to starvation (a function of algalconcentration, Table 8.4, Fig. 8.7) and to predation by the copepod (a function of ciliateconcentration and size). There was no growth or mortality of the copepods over the simulationperiod, as 20 days is less than the gestation period for many mesozooplankton (Raymont1983).The model was designed to examine the dynamics of nanoplankton-ciliate blooms thatwere graazed by a copepod. This was a test of the potential impact and temporal extent ofciliate blooms. In no way should the following analysis be extended to assess the role ofciliates under more complex conditions. In the future, more elaborate, models may do this;these would likely include: advection terms, dispersal of ciliates and prey, migration terms,more predators, estimates of nutrient regeneration, and cohort analysis of the zooplankton.1181.2 ConstraintsAlgal concentration was always 55x104 cells mL-1. This maximum was based on theassumption that algae can be nutrient-limited, and nitrogen, which may be the limiting nutrient(e.g. Harrison et al. 1983, Harris 1986), can have pre-bloom concentrations of 10-20 AM(140-280 Ag L-1, e.g. Talcahashi era!. 1977, Harrison et al. 1983, Haigh et al. 1992). Anaverage 8 Am (diameter) algal cell has 5.24 pg N in it (based on the relation: nitrogen (pg) =0.0172 * (volume in /.4m3)1.023, from Montagnes et al. submitted). Thus, a 20 AM nitrogensource would allow the maximum production of -5x104 cells mL-1. Algal production waszero when the maximum concentration of 5x104 cells mL4 was reached, but as soon as thismaximum concentration was depressed, by grazing, there was algal growth.1.3 Parameters and Rate Equations1.3.1 CiliatesThe ciliate growth parameters (Table 9.1) were established in Chapter 8 (Table 8.4)and were used in a Eq. 1 (Chapter 3) to determine ciliate growth rate for a given algalconcentration.Ciliate grazing rates were calculated using Eq. 3 (Chapter 8), assuming growth rateestimates were accurate, a respiratory quotient of 0.8, and an assimilation efficiency of 0.75(see Chapter 8). Growth and grazing rates were converted to similar units (carbon time4)using the equations: ciliate carbon = 0.148 pg C Am-3 (Putt and Stoecker 1989) and algal.4.carbon (J)g)=0m3)0991g)= 0.109 * live cell volume^(Montagnes et al. submitted).1.3.2 CopepodThe parameters for the copepod were obtained from work conducted on Calanuspacificus by Frost (1972) and modified by Mullin et al. (1975). Frost (1972) establishedpiecewise numerical responses for C. pacificus feeding on diatoms that were 11 gm(Thalassiosirafluviatilis), 35 Am (Coscinodiscus angstii), and 75 gm (Coscinodiscuseccentricus) in mean diameter. Mullin et al. (1975) reanalyzed Frost's data for T. fluviatilisusing a modified Michaelis-Menten equation (with a non-zero intercept, similar to Eq. 1,119(Chapter 3). In the model, I have used the parameters established by Mullin et al. (1975) todetermine the copepod grazing rate on the 8 gm algae (i.e. assuming that 11 = 8 Am algae).From Frost (1972, Fig. 4), I estimated (by eye) the half-saturation concentration (k) andmaximum grazing rate (Gmax) for Cal. pacificus capturing Cos. angstii (35 Am) and Cos.eccentricus (75 gm) (Table 9.1). In the model I used these two sized organisms to establishMichaelis-Menten like responses (with zero intercepts, Eq. 2, Chapter 3) for ciliate C (whichis 25 gm = 35/1m) and ciliates A and B (which are -75 p.m).I assumed that ciliates would be grazed at the same rate as diatoms of a similar size.There are studies which support and refute this assumption (Stoecker and Egloff 1987, Giffordand Dagg 1988, 1991, Tiselius 1989, Wiadnyana and Rassoulzadegan 1989, Jonsson andTiselius 1990). Copepods may sense ciliates and preferentially capture them, but some ciliates"jump" and may avoid capture. Ciliate A, in the model, was based on a species ofStrombidinopsis (which does not jump) while ciliates B and C were based on species ofStrobilidium (which do jump). These variable behaviours could reduce the validity of themodel. However, as there are conflicting data in the literature, and I have no data on thegrazing rates of copepods on these ciliates, I have assumed that ciliates and diatoms would begrazed in a similar fashion by C. pacificus.Copepod grazing of ciliates A, B, and C and algae followed Eq. 5. This equation isused for a single predator consuming n prey types, when the predation rates have beendetermined from separate experiments on single prey types using Michaelis-Menten-likekinetics (see Appendix 2, Legovic 1989).Gi = (Gmaxi*Pi/ki)/ (1 +E Pi/ki )^ (5)where,Gi = grazing rate on prey i (prey d-1)Gmaxi = the maximum grazing rate on prey i (prey d4)Pi = the concentration of prey i (mL-1)= the half-saturation concentration for prey i (mL-1)120E pi/ki = the sum of all prey concentrations divided by their respective half-saturationconstants1.4 Variables (Table 1)I have varied the initial abundance of algae (.05-25x103 mL-1), the invariant (i.e.unchanging during the simulation) abundance of copepods (0.01-5 L4), and the initialabundance of the 3 ciliates (0-1 mL-1), to examine the effect of these variables on populationdynamics and carbon flow.1.5 A Summary of AssumptionsThe following assumptions were made: 1) the time for a potential ciliate bloom tooccur was limited to 20 days, and blooms occurring after 20 days were ignored; 2) advectionwas considered negligible during the bloom period, and there were no terms associated with it;3) only small phytoplankton, ciliates, and copepods were considered in the system (e.g. nolarger phytoplankton, metazoan microzooplankton, or fish were included); 4) copepods grazedciliates at the same rate they would graze diatoms of similar size (i.e. predation was strictlysize dependent); 5) nutrient regeneration was considered only in terms of phytoplanktonmaximum abundance (i.e. phytoplankton growth was limited by nutrient availability only whenthe maximum phytoplankton density was reached, and when phytoplankton density droppedbelow this maximum, growth resumed); and 6) the algal species elicited maximum ciliategrowth rates.1.6 EquationsThe following equations were numerically integrated over a 20 d period (time step =0.05 days) using an iterated fourth order Runge-Kutta procedure (see Appendix 4). Cellabundances were converted to carbon using the relationships of Putt and Stoecker (1989) andMontagnes et al. (submitted), stated above.Rate of change in abundance of ciliate species i (Ci) at time t was modeled as:121dCi/dt = (Aci*C0 - (ZGo*Z)Rate of change in abundance of algae (P) at time t was modeled as:dP/dt = (Ap*P) - (ZGp*Z) - E(CGip*Ci)where P 5x104 cells m1: otherwise P = 5x104 cells m1:1The gross production of ciliate i (CPO over the 20 day period wasf 20CPi= o j Aci(t)*Ci*dtwhere Aci(t) = Aci if Aci .?.. 0, otherwise go(t) = 0The gross production of algae (PP) over the 20 day period was1 20PP = o j Ap(t)*P*dtwhere Ap(t) = /A p if P .. 5x104, otherwise Aci(t) = 0The total number of algae grazed by the copepod (PZG) over the 20 day period was120PZG = o j ZGp(t)*Z*dtThe total number of ciliate i grazed by the copepod (CiZG) over the 20 day period was120CiZG = 0 j ZGo(t)*Z*dtThe total number of algae grazed by the ciliate i (PCiG) over the 20 day period was1 20PCiG = 0 j CGip(t)*Ci*dtwhere,t = 0.05 (d)Aci = the growth rate of ciliate i (d-1), determined by Eq. 1 (Chapter 3), parameters stated inTable 9.1 for each ciliate, and the algal concentration at time t.Ap = the growth rate of the algae (d-1) = 0.7 d-1Z = the copepod concentration (mL-1), set as a constant at the start of each simulationZGci = the grazing rate of the copepods on ciliate i (cells m1:1 d-1), determined by Eq. 4(Chapter 9), the parameters in Table 9.1, and the concentration of ciliate i at time tZGp = the grazing rate of the copepods on the algae (cells m1:1 d-1), determined by Eq. 4(Chapter 9), the parameters in Table 9.1, and the concentration of ciliate i at time t122CGip = the grazing rate of ciliate i on the algae (cells mL-1 d-1), determined by Eq. 1(Chapter 3), Eq. 3 (Chapter 8), the parameters stated in Table 9.1 for each ciliate, andthe algal concentration at time t.2.0 Results of the Model and Discussion of Their Implications2.1 Ciliate Blooms Control Blooms of Small PhytoplanktonThe model (Appendix 4) predicted that ciliates can bloom under conditions wherecopepods are rare and phytoplankton are initially abundant (Figs. 9.2). Under theseconditions, the copepod population was unable to control primary production, phytoplanktonbloomed, and blooms of ciliates grazed down the phytoplankton; Fig. 9.3c-e illustrates someof these bloom conditions.Copepod abundance may vary in coastal and oceanic waters from 0-10 L-1 (e.g.Harrison et al. 1983, Raymont 1983, Landry and Lehner-Fournier 1988, Nielsen andRichardson 1989, Nielsen and Kiorboe 1991). The model indicated that at an invariantconcentration of 0.5-1 copepods L-1 and initial algal concentrations of near 103 mL-1, ciliateblooms do not occur, although there were minor oscillations in ciliate A abundance (Fig. 9.3aand b). These low algal levels were because the copepod population was large enough to grazethe algae down to the copepod's threshold concentration of -800 algae mL-1 (i.e. this is theconcentration where copepods stop eating algae, Table 9.1).Copepod and nano-phytoplankton levels of 0.5-1 copepods L-1 and 103 algal mL-1 arenear average conditions for many coastal waters (e.g. Harrison et al. 1983, Raymont 1983,Middlebrook and Roff 1986). Thus, ciliate blooms probably do not occur under "typical"coastal conditions, and copepods are the dominant grazers of small phytoplankton. Thisconclusion supports Banse's (1982) argument that on average ciliates consume little foodrelative to copepods because the concentrations of suitable food particles tend to be too low.One of the ciliates (A) survived under "typical" conditions but the other ciliates (B andC) did not (Fig. 9.3a and b). The mortality rates of ciliates B and C resulted from their highgrowth-threshold levels (i.e. they need high concentrations of algae to survive, Table 8.4).123How then can ciliates B and C persist? Likely, they persist by existing in a patchyenvironment.Plankton distributions are patchy in both time and space (Haury et al. 1978, Harrisonet al. 1983, Macicas et al. 1985, Harris 1986, Owen 1989, Sime-Ngando 1992). Therefore,conditions should exist where ciliates will bloom and A and B will persist. Ciliates bloomedwhen algae were initially abundant and/or copepods are rare (Fig. 9.3). Such conditions mightexist in the summer (post-spring bloom) or in the winter when mesozooplankton abundancecan be low. They might also occur at any time if mesozooplankton distribution is patchy andphytoplankton are abundant (e.g. Fig 9.3c-e). Such blooms would fall into the "fine" scalepatches (tens to thousands of meters, Haury et al. 1978) which can exist for several days underunperturbed conditions.Ciliate and nanophytoplankton blooms occur in British Columbian waters (Blackbourn1974, Takahashi et al. 1977, Harrison et al. 1983, Appendix 3) and in other waters (Guillardand Ryther 1962, Verity 1986, Dale and Dahl 1987, Lynn and Montagnes 1991, Gallegos1992). In Sechelt Inlet, British Columbia (Appendix 3), ciliate and <10 Am phytoplanlctonabundances occasionally exhibited relatively high levels that could be considered blooms, butrarely achieved the high levels predicted by the model. In contrast, blooms of ciliates occurthat are higher than those predicted by the model (Dale and Dahl 1987, Lynn and Montagnes1991). Thus, there are factors other than those employed in the model which affect bloomdevelopment.British Columbian waters, unlike the model, typically contain more than 3 ciliatesspecies (Martin and Montagnes 1993). A greater diversity of ciliate responses might alter theformation of blooms, as other responses likely exist. Also, to examine the maximum potentialfor ciliate blooms, I established an upper (but atypical) limit for the phytoplankton bloom(5x104 cells mL-1). Field data on phytoplankton abundances (Appendix 3) and observednitrogen pulses of -10 AM (Takahashi et al. 1977) suggest that maximum algal levels of 1-2.5x104 mL-1 are more representative for small scale blooms. Limiting the phytoplankton124biomass does decrease the size of ciliate blooms in the model. Thus, the simulations I havepresented can be viewed as a maximum for ciliate blooms.High abundances of ciliates may result from concentration by physical factors (Mackaset al. 1985). An aggregation of up to 2x103 ciliates mL-1 observed by Dale and Dahl (1987)in a Norwegian bay was probably due in part to concentration from wind mixing, and thesubsequent dispersal of the ciliates was probably due to flushing of the bay. Dale and Dahl(1987) argue that the bloom could only have occurred without mixing if the ciliates divided -3times d-1 (at 9-10°C). This is unlikely, but not impossible, as I found a maximum of 3divisions (1-1 for one ciliate at 16°C (Chapter 4).These differences between the model output and field data do not diminish thelikelihood that ciliate blooms exist or that they graze down phytoplankton blooms. Thedifferences do suggest though, that blooms are more complex than the model simulates. Ofsome interest, and of possible consequence to aquaculture, is the observation that blooms ofsmall phytoplankton and ciliates often occur in enclosures (Grice et al. 1980, Smetacek 1984,Sheldon 1986, Riemann et al. 1990, Turner and Graneli 1992, Graneli et al. 1993).Enclosures, by removing some physical and biological components (e.g. loss of ciliatesthrough advection and mesozooplankton grazing), are simplifications of larger planktonicsystems (Steele and Gamble 1982). Therefore, it is not surprising that ciliate blooms are seenin them. Since the aquaculture of marine organisms often employs small enclosures, suchblooms might be expected there.The primary goal of this dissertation was to substantiate the claim that planktonicciliates bloom and graze down algal blooms over short periods. This section has indicated thatciliates can graze down prey populations. Below I show that these blooms occur over shortperiods of several days.2.2 Ciliates Can Bloom Over 10 to 20 Day PeriodsThe model predicted that ciliate booms typically occurred over a 5-10 day period andwere always less than 20 days, when the 3 ciliates (species A, B, C) were present at initialconcentrations of 0.3 mL-1 (e.g. Fig. 9.3c-e). However, when only one species was125introduced (at 1 mL-1, see section 2.4), some blooms developed after-20 days. Typically,blooms after 20 days occurred when there were low initial algal and low copepod conditions(under these conditions net algal growth was not limited by copepod grazing, but the initialconcentrations were insufficient to immediately stimulate rapid ciliate growth). When bloomsoccurred after 20 days, I did not consider them to be plausible, as conditions, not accountedfor in the model, would likely occur (e.g. mixing events, Mackas et al. 1985 and copepodgrowth and migration, Raymont 1983).Blooms of ciliates often do occur over 10 to 20 days: natural populations of smallautotrophic flagellates and the ciliate Strobilidium (= Lohmanniella) were placed in 1 m3containers, and over 6 days the ciliates bloomed, grazed down the flagellates and then died(Smetacek 1984); a 16-32 day cycle of ciliates and flagellates existed in the Mediterranean Sea(Ibanez and Rassoulzadegan 1977); and ciliate peaks followed heterotrophic flagellate peakswithin 4-6 days in Danish coastal waters (Andersen and Sorensen 1986). These studies wereonly able to establish such relationships by sampling every 1-2 days. Unfortunately, logisticand financial constraints limit most field studies to less regular sampling. For instance,although peaks of ciliates and small phytoplankton occurred in Sechelt Inlet (Appendix 3), themonthly to weekly sampling did not provide information on the formation of these blooms.The model and field data indicate that ciliates can bloom over <20 day periods. As will beshown below, ciliate blooms may be important in the transfer of carbon to upper trophiclevels. Therefore, sampling regimes should be designed to capture these short blooms. Thiswould likely require daily sampling.2.3 Ciliates Are Both a Link and Sink, and Ciliates Can Compete With CopepodsFor the last two decades it has been argued that microzooplankton (20-200 gm) can bean important link between nanoplankton (2-20 Am) and mesozooplankton ( >200 11m) (e.g.Pomeroy 1974, Porter et al. 1979, Gifford and Dagg 1988, 1991). However, this paradigmdoes not consider that ciliates can bloom in the fashion described above. This was possibly anoversight, as considerable production occurs during ciliate-nanoplanlcton blooms (see below),126and ciliates are probably not important in food webs at "typical" concentrations of food andcopepods (see above).Ciliates may not be a link if they graze down their prey and die (due to starvation),before they are captured by mesozooplankton. As was shown in previous chapters, planktonicciliates do not survive when food concentrations are low. The half life of ciliate cultures in theabsence of food is generally much less than 1 d (Chapters 3-7, Wickham et al. 1993). Usingthe model, I have examined the effect of different copepod concentrations and initial algalconcentrations on carbon flow in the food web depicted in Fig. 9.1. The simulations for thisanalysis were conducted with an initial ciliate concentration of 0.3 mL-1 of each of the threeciliates (A, B, and C) and the parameters described in section 1.0. Since the outcome of themodel was affected by both the proportion of the different ciliates initially introduced (seesection 9.2.4) and variation in algal growth rate (data not shown) the following analysis wouldnot apply to all situations. However, it does illustrate the general response of ciliates toaverage conditions.As was shown above (Fig. 9.2), ciliate blooms occurred when copepods were rareand/or initial algal concentrations were high. Under these conditions, both ciliate and primaryproduction were also high (Figs. 9.3e, 9.4), and the majority of algae were consumed byciliates (Figs. 9.3e, 9.5a and b). In contrast, at high copepod concentrations and/or low initialalgal concentrations the majority of algae were consumed by copepods (Fig. 9.5a and b).Between these two extremes there was an abrupt transition (Fig. 9.5a and b): the point wherecopepod grazing became insufficient to deplete phytoplanlcton to the copepod thresholdconcentration of -800 cells mL1 (Table 9.1). Thus, under some conditions ciliate populationshad a greater impact than copepod populations on small phytoplankton (Fig. 9.3 cf. a and bvs. c and e). This is also illustrated in Fig 9.6, where at low copepod and/or high algalconditions, ciliates consumed more of the algae than copepods, over the 20 day period.However, the ciliates may then have been eaten by the copepods and are therefore not simplycompetitors.127At high copepod concentrations and/or low initial algal concentrations, ciliateproduction was low (Fig. 9.4a), but the majority of ciliates produced were consumed bycopepods (Fig. 9.5c). In contrast, at low copepod concentrations and/or high initial algalconcentrations (bloom conditions where ciliate production was high, Figs. 9.2, 9.4a) themajority of ciliates died (due to starvation, e.g. Fig. 9.3e) without being consumed bycopepods (Fig. 9.5c). This is also illustrated in Fig. 9.6a, where the number of ciliatesingested by the copepod population decreased at low copepod concentrations, but ciliateproduction remained high (Fig. 9.4a). Not surprisingly, the proportion of ciliate carbontransferred to upper trophic levels decreased as the number of copepods decreased.However, during bloom conditions, ciliates were a link in the transfer of carbon tocopepods: at low copepod concentrations and high initial algal concentrations, 40 to 50% ofthe carbon ingested by copepods was from ciliates (Fig. 9.5d). Gifford and Dagg (1988) alsofound that when ciliates were abundant, they constituted a significant fraction of the diet of thecopepod Acartia tonsa, even though phytoplankton were also abundant. In contrast, at highcopepod concentrations and low initial algal concentrations, virtually all the carbon ingested bycopepods was from phytoplankton (Fig. 9.5d).In summary, under ciliate bloom conditions in the model ciliates were a link, providing40-50% of the carbon available to copepods (Fig. 9.5d). It was only when copepods wereabundant and initial algal levels were low that ciliates were not an important carbon source.Field data support these conclusions: after a storm, in coastal waters off Denmark, wheremixing caused increased primary production of the <11 Am fraction, ciliate abundanceremained between 0.05-0.3 mL4 (Nielsen and Kiorboe 1991). This low ciliate abundancewas attributed to the high mesozooplankton abundance (5-10 L-1). However, in other Danishcoastal waters, where copepod abundance is typically low (often <1 L-1, Blanner 1982), therewere ciliates blooms of up to 160 mL4 (Andersen and Sorensen 1986). In another case, thecopepod Neocalanus plumchrus was an efficient grazer of 2-30 Am phytoplankton and, atdensities of 1 animal L-1 it was able to control small (4-10 Am) phytoplankton over 6 dayincubations (Landry and Lehner-Fournier 1988). But, in the open Pacific ocean, where this128study was conducted, N. plumchrus is likely not to be abundant enough to controlphytoplankton, as it was typically found at concentrations of 0.2 L-1 (Landry and Lehner-Fournier 1988). Thus, as suggested by others (e.g. Frost 1987, Strom and Welschmeyer1991), this is a region where microzooplankton may be important grazers. A study inmesocosms also indicated that when copepods were less abundant (due to introduction of aplanktivorous fish), flagellates and then ciliates began to bloom over 10 days (Graneli et al.1993).Under "typical" coastal conditions (103 algae mL-1 and 1 copepod L-1), the modelindicated that ciliates act as a link (Figs. 9.3a, 9.5d), but under these "typical" conditions therewas little primary production, as a low phytoplankton biomass was maintained by copepodgrazing (see Fig. 9.3a, 9.4b). This supports the view that on average, ciliates are not animportant component of coastal planktonic systems (Banse 1982, Montagnes et al. 1988a).Although ciliates were a link under bloom conditions, a large proportion of the ciliatesdied of starvation in post-bloom conditions and were therefore not available tomesozooplankton (Fig. 9.3a, 9.5c). Incidental observations from my experiments (Chapters 3-7) suggest that starved and dead ciliates do not remain intact (data not shown). If this is so,ciliates would not sink from the water column and would fuel the microbial loop,remineralizing nutrients to the primary producers (Azam et al. 1983).2.4 Species Differences Are ImportantIn the model, when the initial concentration of the 3 ciliates was kept constant (1:1:1,at 0.3 ciliates mL-1), variations of both initial algal and invariant copepod concentrationscaused changes in the species composition of the blooms (Fig. 9.2). For instance, whencopepods were rare and algae abundant, ciliates A and B both bloomed and had similar peakabundances, but at slightly lower algal concentrations only ciliate A bloomed (Fig. 9.2a andb). When only one species was introduced into the model, the species differences were morepronounced (Fig. 9.7); examples of these are illustrated in Fig. 9.8. Ciliate C was anexceptional example of these differences: it reaches peak abundances of only 6 mL-1 when theother two species were present, but it bloomed to 500 mL-1 when introduced alone (Fig. 9.2,1299.7b). These differences in model results were caused by differences in thresholdconcentrations, half-saturation constants, and maximum rates of the 3 ciliates (Table 9.1) andillustrate the uniqueness of species specific responses. As mentioned above, the values ofciliate abundance may be unreasonably high, but the relative differences between the results isinstructive: the abundance of species in a bloom is affected by the type and relative proportionof ciliates encountering the phytoplankton bloom.Bloom dynamics and the flow of carbon through the food web also depended on theciliate species present. Examples of these differences are depicted in Fig. 9.8. For instance,when only ciliate A was present or it was present in equal concentrations with the other twospecies (Fig. 9.8a and d), primary production was low, and copepods consumed a relativelylarge portion of the ciliates. However, when only ciliates B or C were present, primaryproduction was high, relatively few of the ciliates were eaten by copepods, and many of theciliates died due to starvation when algae were depleted (Fig 9.8b and c).In summary, the differences between numerical responses, presented in Chapter 8, canbe of consequence in terms of ciliate bloom dynamics and carbon flow. Such differences arelikely species dependent (see Chapters 2 and 8). Thus, both species-level analysis of fieldsamples and laboratory examination of species differences will be warranted if we wish tounderstand the range of influence that ciliates could have on planktonic food webs. Anobvious corollary to this is that we must also continue taxonomic studies on oligotrichs, whichare still a poorly studied group of organisms (Martin and Montagnes 1993).2.5 How Ciliate Blooms Influence Carbon Flow to CopepodsCiliate blooms occur when there were initially many algae and copepods were rare (seeabove). The model provided an indication of the flow of carbon through the food web and canoffer some information on the relative importance of ciliate blooms.Under bloom conditions, both ciliate and algal production, over a 20 day period,increased from non-bloom conditions by an order of magnitude (Figs. 9.3, 9.4). At the regionwhere both algae and ciliate production begin to increase (Fig. 9.4a and b), the carbon flow(total, ciliate, or algae) to the copepod population was maximized (Figs. 9.6a, 9.9a and c).130Thus, this transition region appears to be important in terms of maximum mesozooplanktonproduction. However, the maximum benefit in terms of carbon ingestion by individualcopepods lay in the entire region where ciliates bloomed (Figs. 9.9b and d). Therefore,during ciliate blooms the flow of carbon to individual copepods was high, even when copepodabundance was low. This suggests that while ciliate blooms do not enhance the production ofmesozooplankton populations over short periods, they may be a food source for individuals.Thus, ciliate blooms which are <20 days may aid in the long term production ofmesozooplankton.The predictions of the model agree with field data. During a phytoplankton bloom,when copepods were abundant, copepods controlled the ciliate population, and ciliatescontributed -10% to the copepod diet (Nielsen and Kiorboe 1991). Under these conditions,ciliates were not a link between the microbial and classical (phytoplankton-copepod) foodwebs. A similar situation existed in coastal waters with the copepod Acartia clausi grazing onciliates (Tiselius 1989): at low ciliate concentrations (<1 mL-1), ciliates were a minorcomponent of the copepod's diet. However, in other coastal waters, where ciliates wereabundant (as a result of grazing on nanoplanlcton) they were a significant portion of the diet ofAcartia tonsa (Gifford and Dagg 1988). Similar results were also found in the NorthBering/Chukchi Seas, where the link from nanoplankton to copepods, through ciliates,increased at regions of higher nanoplankton biomass (Andersen 1988). Thus, ciliates are notan important component of copepod production when nano-phytoplankton are rare, but theymay be when these autotrophs bloom.CONCLUDING REMARKSAssuming that the ciliates used in the model (Table 8.4, Fig. 8.7) are representative ofcoastal ciliates, the above analysis suggests that, over a 5-20 day period, ciliates can act as"bloom and bust" organisms in a "feast and famine" situation. During blooms the ciliates areable to graze down populations of small phytoplankton, and subsequently the ciliate populationcrashes due to predation by copepods and starvation. Ciliate blooms are a link in the flow of131carbon from phytoplankton to individual copepods but channel much of the carbon away fromupper trophic levels. It also appears that the species composition of the ciliate assemblage canaffect the bloom dynamics.We must therefore rethink our understanding of ciliates in planktonic systems. Theymay not be a constant link between small phytoplankton and mesozooplankton, as previousanalyses implied (e.g. Pomeroy 1974, Conover 1982, Laval-Peuto et al. 1986). Rather, theycan be, under transient conditions, important as both links and sinks of carbon, but under"typical" coastal conditions (103 algae mL-1 and 1 copepod L-1) ciliates may not be importantcomponents of food webs. One of the most important findings of my studies was that thethreshold food concentration (where ciliates began to die due to starvation) was often higherthan "typical" food concentrations (Chapter 8 and Appendix 3). This means that ciliates cannot survive at "typical" concentrations and rely on patches and/or blooms of phytoplankton.The question then arises: how do ciliates survive between patches?There are few data explaining how planktonic ciliates survive periods of starvation.My studies indicated that oligotrichs die within 2-3 days of starvation (Chapters 3-7), and at notime did these ciliates form resting cysts. However, tintinnids can form cysts (Reid and John1978, 1983, Takahashi and Aizawa 1990) and naked oligotrichs may also (Reid 1987). It issurprising that encystment has never been reported for cultured planktonic oligotrichs. This isone area that requires more work if we are to understand the germination of ciliate blooms.Mixotrophy may also be a mechanism that allows some ciliates to survive periods ofstarvation, as they can derive nutrition from enslaved chloroplasts and may maintain them forhours to days (Stoecker and Silver 1990, Stoecker and Michaels 1991). Finally, micro-scalepatchiness (Owen 1989) is another mechanism that may allow ciliates to survive betweenblooms. If phytoplankton are concentrated in small patches (centimeters to meters), thenciliates may be moving from one patch to another. Such associations have been shown fortintinnids and dinoflagellates (Stoecker et al. 1984) and ciliates may migrate (Jonsson 1989).Again, this is an area that requires study.132There is also a lack of information on where ciliates go when they die. If after a bloomciliates are not ingested by upper trophic levels (e.g. mesozooplanlcton), what is the fate oftheir biomass? There is good evidence that ciliates decrease in cell size as they starve (e.g.Lynn et al. 1987, Fenchel 1990). Thus, a substantial portion of ciliate biomass may berespired in post bloom periods, and ciliates would act as a carbon sink (i.e. a loss of carbonfrom the system). It is also possible that starved cells eventually lyse and their contentsbecome available to bacterial degradation. Thus, ciliates blooms may contribute to themicrobial loop (Azam et al. 1983). The fate of such blooms may be important in ourunderstanding of carbon cycling, especially in regions like the subarctic Pacific wheremicrozooplankton may be the dominant grazers.Finally, now that I have supported the notion that ciliates can bloom, it would beconstructive to investigate these blooms in the field. The problems with doing this areobvious: if ciliate blooms are transient and brief, then observing an entire bloom will requireexhaustive sampling. My studies suggest ways to simplify such a search by narrowing therange of conditions where blooms should occur. Ciliates should bloom when:mesozooplankton are not abundant; small phytoplanlcton are abundant; conditions allow smallphytoplankton to bloom; and lack of mixing processes provide stable conditions over 1-3weeks. Where are these places? Ciliate blooms should occur in small embayments and inletswhere there is little mixing but nutrient enrichment may occur, either through periodic mixingor allochthonous processes. Blooms may also occur in enclosures such as aquaculture tanks orexperimental mesocosms. Finally, blooms may occur in the open water column if conditionsallow for patches in the 100 m to 10 km range, which persist for days to weeks. Such patchescould, for instance, be caused by fronts or vertical shear (Haury et al. 1978, Mackas et al.1985). Daily examination of ciliate distributions over several weeks may reveal that ciliateblooms are in fact a regular but ephemeral occurrence in many planktonic systems.1 3 3Table 9.1. Parameters and variables used in the model of ciliate-phytoplankton (algal) interactions.Parameter^Value^Units^Descriptionin modeldt^0.05^days^time stepCiliates-1P1^0.99^d Ciliate A, maximum growth rate (Amax)P2 380 algae mL4^Ciliate A, growth threshold concentration (x)'P3^921^algae mL4 Ciliate A, growth half-saturation constant (k)P4 1.85 (II^Ciliate B, maximum growth rate (Amax)P5^13014^algae mL 1^Ciliate B, growth threshold concentration (x)'P6 24436 algae mL4^Ciliate B, growth half-saturation constant (k)P7^0.72^d-1^Ciliate C, maximum growth rate (it x)P8 3105 algae mL4^Ciliate C, growth threshold concentration (x)'P9^8143^algae mL4^Ciliate C, growth half-saturation constant (k)P10 0.8 CO2/02 respiratory quotient for ciliatesP11^0.75none^assimilation efficiency for ciliates-1P12 0.7^d algal growth rate (A)Copepod3312000^algae (11^the maximum grazing rate on the algaecopepod-1day-1767^algae mL4^the grazing threshold concentration for the algae (x)'785 algae mL-1^the grazing half-saturation constant for the algae (k)30000^ciliates d4^the maximum grazing rate on ciliate Ccopepod-1day-1100^algae mt-1^the grazing half-saturation constant for ciliate C (k)14400^ciliates d-il^the maximum grazing rate for ciliates A and Bcopepod-1day-1^.45^algae mL1^the grazing threshold concentration for ciliates A and B (x)'921 algae mL4^the grazing half-saturation constant for the algae (k)Variables^Range^Units^Descriptionin modelxl^50-25x103^# mL4^initial algal concentrationx2 0-1^# mL4-1^initial concentration of ciliate Ax3^0-1 # mL initial concentration of ciliate Bx4 0-1^# mL 1^initial concentration of ciliate CL-1zoo^0.01-5^# invariant copepod concentrationciliates^copepods1 3 4Fig. 9.1. A schematic diagram of the model described in Appendix 4. There are five types of organisms in themodel: small (8 Am) phytoplankton (algae), ciliate species A, B, C, (see Table 8.4, Fig. 8.7) and a copepodspecies (data obtained from Frost, 1972 for Calanus pacificus). During the 20 day simulation, the algae had aconstant growth rate of AL = 0.7 d4 (but see text). Algal mortality was by predation from the ciliates and thecopepod. Ciliate growth rates were functions of algal concentration. For all three ciliates, mortality was due tostarvation and to predation by the copepod. There was no growth or mortality of the copepods over thesimulation period. Four calculations of carbon transfer, over the 20 day simulation period, are presented insubsequent examples of this diagram (Figs 9.3 and 9.8). These values were determined from the integratedproduction (carbon time-1) over the 20 day simulation. They are represented here as: PP, the amount ofautotrophic carbon produced; ZP, the percentage of autotrophic carbon produced that was consumed by thecopepods; CP, the percentage of autotrophic carbon produced that was consumed by the ciliates; ZC, thepercentage of ciliate carbon produced that was consumed by the copepods; and CZ, the percentage of the copepoddiet contributed by ciliates (i.e. ciliate carbon ingested / [ciliate carbon ingested + algae carbon ingested]).Species A100755025o0 4oqbe^24 0^1Q'^0.5,Species C3 250002000015000100005000 Species B60453015o4A.,D,kiv11.2500020000000015000^1^....1. NI5000LIC1-149.It..xe2500020000^.....1. N15000 )50001000023.te.e135Fig. 9.2. Output from the model of ciliate-phytoplankton dynamics (Appendix 4): the effect of variations ininitial algal concentration and constant copepod abundance (i.e. it did not change during the simulation) on theformation and peak magnitude of ciliate blooms, a, b, c, peak numbers of ciliates A, B, and C, respectively(Table 8.4), respectively. All the simulations for these panels were initiated with a total of 0.9 ciliates mL4 (0.3of each ciliate). The initial algal concentration was varied from 25 to 2.5x104 mL-1 and the constant copepodabundance was varied from 0.025 to 5 L-1.^initial prey = 103 mL-1^copepods .. 1 L-1^ a1000 ,,. ^1 1 .5^/-----------0---7505002500_-__-,--^__^-- -- -7..- - - "._^••..^r 1^I1.0^'3"3 48 i■lelaelaitt=0.5 0.0-iinitial prey = 103 =. L-1^copepods a, 0.5 L b1000.1^250■ o,^•^•--1.5^--Aill.6--eAlba4z0-____L.ALLF78 1-1W0.50.0■Ittill.\'^.0,„„,,,/.131W50°, -1• initial prey = 10 3 mLcopepods = 0.15-1L --------------0-- CI'^22501 5 00750• 01^I -.^-%. -- ,'^. .' *******43210 I^I h I I g4111^1.11110400. ciBatesd I 4 IA4011111=I I I III I I I I I I I• - - - - . . _ . _ _ _ _ _ 3. . . .100initial prey = 5x103 mL-1 copepods a. 1 L-1di^i^1^- 56000-4800360024001200-'-'•\a' -,^.- -t- :^\ --43 6 2 =Mates21^ 180 ,: ****Initial prey = 5x103 mL- 1 copepods a, 0.1 L --1600004.8000-36000024000120001^,I 1 - 80- --^i, .,-^J. ,• ...,_^!,,,,,_-------0--60 (1)-^tw.),-:3=,0o^510^15^20p-o136DaysFig. 9.3. Output from the model of ciliate-phytoplankton dynamics (Appendix 4): five examples of bloomdevelopment over 20 days and the flow of carbon during those 20 days: a, initial algal concentration = 103 ml..-1and constant copepod concentration = 1 L I; b, initial algal concentration = 10"tilL4 and constant copepodconcentration = 0.5 L; c, initial algal concentration = 103 m1,4 and constant copepod concentration = 0.1L 1 ; d, initial algal concentration = 5x103 mL4 and constant copepod concentration = 1 L; e, initial algalconcentration = 5x103 mL4 and constant copepod concentration = 0.1 L-1. Phytoplankton, solid line. CiliateA, large dashed line. Ciliate B, small dashed line. Ciliate C, dotted line. All the simulations for these panelswere initiated with a total of 0.9 ciliates ad:1 (0.3 of each ciliate). The calculation of values in circles, ovals,and diamonds are described in Fig 9.1.200015001 0005000Ciliateproduction(ng C over20 days)2500015000200001000050004000300020001 00004C' 3oto 2ei)^1oct 0N't5000111-t?le'\25000200001500010000Primaryproduction(ng C over20 days)LI"c1)5.1g,e,„ealgeLe5000Fig. 9.4. Output from the model of ciliate-phytoplankton dynamics (Appendix 4): the effect of variations ofinitial algal concentration and constant copepod abundance on 2 bloom parameters: a, gross ciliate (ng C)production over the 20 day simulation and b, gross primary (algae) production (ng C) over the 20 day simulation.The simulations for these panels were initiated with a total of 0.9 ciliates mL4 (0.3 of each ciliate). The initialalgal concentration was varied from 100 to 2.5x10 4 mL4 and the constant copepod abundance was varied from10.01 to 5 L.13713880 a^. 4 0 ri■- -.111■ -., .10040^Algae eaten by600ciliates/primaryOa 4 3.i•- ■-■041F1.n-46o 2o ■%4Ir 150002000025000production201^loonre , 05000-I..)Algae eaten bycopepods/primaryproductionAWr11111  ilitill175002000205000\gfrel0000^Q0')•Ciliates eatenby copepods/ciliate production25000200001500010000500050400501140Ailli0 3oxo4^. 4.44 OP. .4 0 .■j ,I,I.b ., .- -111 -11. 14Alltp-410""2 ANI■ tr^200005°°° ....120.eac,461^5000100005n°°Q,^0 1 0 z.e Q51:js rz. ,i,IT:Ile'Fig. 9.5. Output from the model of ciliate-phytoplankton dynamics (Appendix 4): the effect of variations ininitial algal concentration and constant copepod abundance on 4 bloom parameters measured over the 20 daysimulation: a, (algal carbon consumed by copepods)/(algal carbon produced); b, (algal carbon consumed byciliates)/(algal carbon produced); c (ciliate carbon consumed by copepods)/(ciliate carbon produced); d. (ciliatecarbon consumed by copepods)/(ciliate carbon consumed by copepods + algal carbon consumed by copepods).All the simulations for these panels were initiated with a total of 0.9 ciliates mL-1 (0.3 of each ciliate). Allproduction and consumption measurements are the integrated value over the 20 day simulation. The initial algalconcentration was varied from 100 to 2.5x104 mL -1 and the constant copepod abundance was varied from 0.01 to5.0 L-1.1008060402001 008060402004.46e,2 210c" ^0_)Ciliates eaten bycopepods/(ciliateseaten by copepods+ algae eaten bycopepods)4321Algal carboneaten bycopepods(ng C 20 d ' )250002000015000500010000^Tt1.13g e20001 5001 0005000 40 3.o.to^2!eo„ 1`Cs, ;-(L300020001 0000 4 3 2 1oo:s,^0rzAlgal carboneaten by ciliates(ng C 20 c1-1)e (Tflij5000Blg250002000015000^t10000Fig. 9.6. Output from the model of ciliate-phytoplankton dynamics (Appendix 4): the effect of variations ininitial algal concentration and constant copepod abundance on 2 bloom parameters: a, algal carbon (ng) eaten bycopepods over the 20 day simulation and b, algal carbon (ng) eaten by ciliates over the 20 day simulation. Thesimulations for these panels were initiated with a total of 0.9 ciliates mL4 (0.3 of each ciliate). The initial algalconcentration was varied from 100 to 2.5x104 mL4 , and the constant copepod abundance was varied from 0.01to 5 L-1.1394cr)Species A0I.-4^go6030ti--t^002634^1Or ^0recviSpecies C500400,300rn^200C.)^1000cvic0•r4r--4• r-4250002000015000100005000100BO604020043220000250001150000000 1. 231,8.e50001V‘2'"343Species B25000200001500010000^-)50001-1A6 seB. bi4cda);a.140Fig. 9.7. Output from the model of ciliate-phytoplankton dynamics (Appendix 4): the effect of variations ininitial algal concentration and constant copepod abundance on the formation and peak magnitude of ciliate bloomswhen only one species (A, B, or C, Table 8.4) was included in the model: a, only ciliate A present; b, only ciliateB present; c, only ciliate C present. All three simulations were initiated with an algal concentration of 25x10'mL4 and one ciliate mL -1 . A constant copepod concentration of 0.5 L-1 was maintained throughout thesimulations.3.02.52.01.51.00.50.03000250020001 5001000500Ciliate A present100806040206000050000400003000020000100000Ciliate B present6000050000400003000020000100000Ciliate C present^ 600— 500' — 400—300— 200— 100s^* 0Ciliate A. B. and C present3000 3.52500 — 3.02000 — — 2.51500 — — 2.01000 -..—1.51.0500,, ..—0.50 0.00^5 10 15 2044.4..••••••■E 1,4•••44...^1.4•••■•••••■••••riCS^"4JCLI^• r■I0 r.1• f■4›.%141DaysFig. 9.8. Output from the model of ciliate-phytoplankton dynamics (Appendix 4): the effect of 3 ciliate species(A, B, and C, Table 8.4) on bloom development over 20 days and the flow of carbon during those 20 days: a,only ciliate A present; b, only ciliate B present; c, only ciliate C present; d, ciliates A, B, and C present. Forpanels a-c dashed lines are ciliates. For all panels solid lines are algae. For panel d: ciliate A, large dashed line;ciliate B, small dashed line; and ciliate C, dotted line. All four simulations were initiated with an algalconcentration of 25x103 mL-I and one ciliate mL4 (0.333 mL4 of each ciliate in panel d). A constant copepodconcentration of 0.5 L-1 was maintained throughout the simulations. The calculation of values in circles, ovals,and diamonds are described in Fig 9.1.14224001 600800 - 1*#0,, 4^101110. 25000°2't"jc,,, 3 ■ 10001750020"n1^•5000•et-'Total carbon (ciliates + algae)eaten by copepods(ng C 20 d-1)..10900600Total carbon (ciliates +^300algae) eaten by an individual 0copepod (pg C 20 d-1) 250002000015000^..- \100005000600450300150oLb 4.ebe^3460 2C/4,Ciliate carbon eaten byan individuil copepod(pg C 20 d )41r 1111bit.^Ciliate carbonVI411F--4■1Ibb.^...■1111111111^eaten by corpodstrig I I 0111 25000 (ng C 20 d- )150020000010000500060045030015004o2, 0algae25000200001500000 10000Fig. 9.9. Output from the model of ciliate-phytoplankton dynamics (Appendix 4): the effect of initial algalconcentration and constant copepod abundance on 4 bloom parameters: a, total (algae + ciliates) carbon (ng)eaten by copepods over the 20 day simulation; b, total (algae + ciliates) carbon (pg) eaten by an individualcopepod over the 20 day simulation; c, ciliate carbon (ng) eaten by copepods over the 20 day simulation; d, ciliatecarbon (pg) eaten by an individual copepod over the 20 day simulation. The simulations for these panels wereinitiated with a total of 0.9 ciliates inL-1 (0.3 of each ciliate). The initial algal concentration was varied from 100-1to 2.5x103 mL-1 , and the constant copepod abundance was varied from 0.01 to 5 L1.143LITERATURE CITEDAdl, S. M., Berger, J. D. (1991). Timing of oral morphogenesis and its relation tocommitment to division in Paramecium tetraurelia. Exp. Cell Res. 192: 497-504Aelion, C. M., Chisholm, S. W. (1985). Effect of temperature on growth and ingestion ratesof Favella sp. J. Plank. Res. 7: 821-830Andersen, P. (1988). The quantitative importance of the "microbial loop" in the marinepelagic: a case study from the North Bering/Chukchi seas. Arch. Hydrobiol. 31: 243-251Andersen, P., Sorensen, H. M. (1986). Population dynamics and trophic coupling in pelagicmicroorganisms in eutrophic coastal waters. Mar. Ecol. Prog. Ser. 33: 99-109Azam, F., Fenchel, T., Field, J. G., Gray, J. S., Meyer-Reil, L. A., Thingstad, F. (1983).The ecological role of water column microbes in the sea. Mar. Ecol. Prog. Ser. 10: 257-263Banse, K. (1982). Cell volume, maximal growth rates of unicellular algae and ciliates, and therole of ciliates in the marine pelagial. Limnol. Oceanogr. 27: 1059-1071Bell, G. (1988). Sex and death in Protozoa. Cambridge Univ. Press, CambridgeBernard, C., Rassoulzadegan, F. (1990). Bacteria or microflagellates as a major food sourcefor marine ciliates: possible implications for the microzooplankton. Mar. Ecol. Prog. Ser.64: 147-155Berger, J. D. (1976). Gene expression and phenotypic change in Paramecium tetraureliaexconjugants. Genet. Res. Camb. 27: 123-134Blackbourn, D. J., Taylor, F. J. R., Blackbourn, J. (1973). Foreign orgenelle retention byciliates. J. Protozool. 20: 286-288Blackbourn, D. J. (1974). The feeding biology of tintinnid protozoa and other inshoremicrozooplankton. Ph.D. thesis. Univ. British ColumbiaBlanner, P. (1982). Composition and seasonal variation of the zooplankton in the Limfjord(Denmark) during 1973-1974. Ophelia 21: 1-40Borsheim, K. Y., Bratbak, G. (1987). Cell volume to cell carbon conversion factors for abacterivorous Monas sp. enriched from seawater. Mar. Ecol. Prog. Ser. 36: 171-175Buskey, E. J., Stoecker, D. K. (1988). Locomotory patterns of the planktonic ciliate Favellasp.: adaptations for remaining within food patches. Bull. Mar. Sci. 43: 783-796Buskey, E. J., Stoecker, D. K. (1989). Behavioral responses of the marine tintinnid Favellasp. to phytoplankton: influence of chemical, mechanical and photic stimuli. J. Exp. Biol.Ecol. 132: 1-16Capriulo, G. M. (1990). Feeding-related ecology of marine protozoa. In: Capriulo, G. M.(ed.) Ecology of marine protozoa. Oxford Univ. Press, Oxford, p. 186-259Caron, D. A., Goldman, J. C. (1990). Protozoan nutrition regeneration. In: Capriulo, G. M.(ed.) Ecology of marine protozoa. Oxford Univ. Press, Oxford, p. 283-306144Caron, D. A., Goldman, J. C., Fenchel, T. (1990) Protozoan respiration and metabolism. In:Capriulo, G. M. (ed.) Ecology of marine protozoa. Oxford Univ. Press, Oxford, p. 307-322Caron, D. A., Goldman, J. C. (1993). Predicting excretion rates of protozoa: reply tocomment by Landry. Limnol. Oceanogr. 38: 472-474Condrey, R. E. (1982). Ingestion-limited growth of aquatic animals: the case for Blackmankinetics. Can. J. Fish. Aquat. Sci. 39: 1585-1595Condrey, R. E., Fuller, D. A. (1985). Testing equations of ingestion-limited growth. Arch.Hydrobiol. 21: 257-267Conover, R. J. (1982). Interrelations between microzooplankton and other organisms. Ann.Inst. oceanogr. 58(S): 31-46Corliss, J. 0. (1979). The ciliated protozoa: characterization, classification and guide to theliterature, Second edition. Pergamon Press, LondonCorliss, J. 0., Esser, S. C. (1974). Comments on the role of the cyst in the life cycle andsurvival of free living protozoa. Trans. Am. Microsc. Soc. 93: 578-593Dale, T., Dahl, E. (1987). Mass occurrence of planktonic oligotrichous ciliates in a bay insouthern Norway. J. Plank. Res. 9: 871-879Faure-Fremiet, E. (1924). Contributions a la connaissance des infusoires planktoniques.Bulletin Biologique de la France et de la Belgique, Supplement no. 6, p 1-171Faure-Fremiet, E. (1969). Remarques sur la systêmatique des cilia oligotrichida.Protistologica 5: 345-352Fenchel, T. (1982). Ecology of heterotrophic microflagellates. III. Adaptations toheterogeneous environments. Mar. Ecol. Prog. Ser. 9: 25-33Fenchel, T., Finlay, B. J. (1983). Respiration rates in heterotrophic, free-living protozoa.Microb. Ecol. 9: 99-122Fenchel, T. (1986). Protozoan filter feeding. In: J. 0. Corliss, Patterson, D. J. (ed.)Progress in Protistology Vol. 1. Biopress Ltd., Bristol, p. 66-113Fenchel, T. (1987). Ecology of Protozoa, the biology of free-living phagotrophic protists.Springer-Verlag, New YorkFenchel, T., (1990). Adaptive significance of polymorphic life cycles in Protozoa: responsesto starvation and refeeding in two species of marine ciliates. J. Exp. Mar. Biol. Ecol. 136:159-177Fenchel, T. Finlay, B. J. (1983). Respiration rates in heterotrophic, free-living protozoa.Microb. Ecol. 9: 99-122Fenchel, T., Jonsson, P. R. (1988). The functional biology of Strombidium sulcatum, amarine oligotrich ciliate (Ciliophora, Oligotrichina). Mar. Ecol. Prog. Ser. 48: 1-15Foissner, W., Skogstad, A., Pratt, J. R. (1988). Morphology and infraciliature ofTrochiliopsis australis n. sp., Pelagohalteria viridis (Fromentel, 1876) n. g., n. comb.,and Strobilidium lacustris n. sp. (Protozoa, Ciliophora). J. Protozool. 35: 489-497145Frost, B. W. (1972). Effects of size and concentration of food particles on the feedingbehavior of the marine planktonic copepod Calanus pacificus. Limnol. Oceanogr. 17:805-815Frost, B. W. (1987). Grazing control of phytoplankton stock in the open subarctic PacificOcean: a model assessing the role of mesozooplankton, particularly the large calanoidcopepods Neocalanus spp. Mar Ecol. Prog. Ser. 39: 49-68Frost, B. W., Franzen, N. C. (1992). Grazing and iron limitation in the control ofphytoplankton stock and nutrient concentration: a chemostat analogue of the Pacificequatorial upwelling zone. Mar. Ecol. Prog. Ser. 83: 291-303Gallegos, C. L. (1992). Phytoplankton photosynthesis, productivity, and species compositionin a eutrophic estuary: comparison of bloom and non-bloom assemblages. Mar. Ecol.Prog. Ser. 81: 257-267Gifford, D. J. (1985). Laboratory culture of marine planktonic oligotrichs (Ciliophora,Oligotrichida). Mar. Ecol. Prog. Ser. 23: 257-267Gifford, D. J. (1988). Impact of grazing by microzooplankton in the Northwest arm of HalifaxHarbour, Nova Scotia. Mar. Ecol. Prog. Ser. 47: 249-258Gifford, D. J., Dagg, M. J. (1988). Feeding of the estuarine copepod Acartia tonsa Dana:carnivory vs. herbivory in natural microplanlcton assemblages. Bull. Mar. Sci. 43: 458-468Gifford, D. J., Dagg, M. J. (1991). The microzooplankton- mesozooplankton link:consumption of planktonic protozoa by the calanoid copepods Acartia tonsa Dana andNeocalanus plumchrus Murukawa. Mar. Microb. Food Webs 5: 161-177Graneli, E., Olsson, P., Carlsson, P., Graneli, W., Nylander, C. (1993). Weak "top-down"control of dinoflagellate growth in the coastal Skagerrak. J. Plank. Res. 15: 213-237Grice, G. D., Harris, R. P., Reeve, M. R., Heinbokel, J. F., Davis, C. 0. (1980). Large-scale enclosed water column ecosystems. An overview of foodweb I, the final CEPEXexperiment. J. mar. biol. Ass. U.K. 60: 401-414Grim, J. N. (1987). The lcinetid structures of the Choreotrichous ciliate Strobilidium velox andan assessment of its evolutionary lineage. J. Protozool. 34: 117-123Gold, K. (1970). Cultivation of marine ciliates (Tintinnida) and heterotrophic flagellates.Helgoldnder wiss. Meeresunters 20: 264-271Gonzalez, J. M., Sherr, E. B., Sherr, B. F. (1990). Size-selective grazing on bacteria bynatural assemblages of estuarine flagellates and ciliates. App. Environ. Microbiol. 56:583-589Guillard, R. L., Ryther, J. H. (1962). Studies of marine planktonic diatoms, I. Cyclotellananna, and Detonula confervacea (Cleve) Gran. Can. J. Microbiol. 8: 229-239Haigh, R., Taylor, F. J. R., Sutherland, T. E. (1992). Phytoplankton ecology of Sechelt Inlet,a fjord system on the British Columbia coast. I. General features of the nano- andmicroplankton. Mar. Ecol. Prog. Ser. 89: 117-134Harris, G. P. (1980). Temporal and spatial scales in phytoplankton ecology. Mechanisms,methods, models, and management. Can. J. Fish. Aquat. Sci. 37: 877-900146Harris, G. P. (1986). Phytoplankton ecology. Chapman and Hall, New YorkHarrison, P. J., Waters, R. E., Taylor, F. J. R. (1980). A broad spectrum artificial mediumfor coastal and open ocean phytoplankton. J. Phycol. 16: 28-35Harrison, P. J., Fulton, J. D., Taylor, F. J. R., Parsons, T. R. (1983). Review of thebiological oceanography of the Strait of Georgia: pelagic environment. Can. J. Fish.Aquat. Sci. 40: 1064-1094Harrison, P. J., Thompson, P. A., Calderwood, G. S. (1990). Effects of nutrient and lightlimitation on the biochemical composition of phytoplankton. J. Aplp. Phycol. 2: 45-56Haury, L. R., McGowan, J. A., Wiebe, P. H. (1978). Patterns and processes in the time-space scales of plankton distributions. In: Steele, J. H. (ed.) Spatial patterns in planktoncommunities. Plenum Press, New York, p. 277-327Heinbokel, J. F. (1978). Studies on the functional role of tintinnids in the southern CaliforniaBight. I. Grazing and growth rates in laboratory cultures. Mar. Biol. 4: 177-189Holling, C. S. (1959). Some characteristics of simple types of predation and parasitism. Can.Entomol. 41: 385-399Ibanez, F., Rassoulzadegan, F. (1977). A study of the relationships between pelagic ciliates(Oligotrichina) and planktonic nanoflagellates of the neritic ecosystem of the Bay ofVillefranche-Sur-Mer. Analysis of chronological series. Ann. Inst. oceanogr. 53: 17-30Jackson, K. M., Berger, J. (1984). Survival of ciliated protozoa under starvation conditionsand at low bacterial levels. Microb. Ecol. 10: 47-59Jackson, K. M., Berger, J. (1985). Survivorship curves of ciliate protozoa under starvationconditions and at low bacterial levels. Protistologica 21: 17-24Jerome, C. A. Montagnes, D. J. S., Taylor, F. J. R. (1993). The effect of the quantitativeprotargol stain and Lugol's and Bouin's fixatives on cell size: a more accurate estimate ofciliate species biomass. J. Euk. Microbiol. 40: 254-258Johannes, R. E. (1965). Influence of marine protozoa on nutrient regeneration. Limnol.Oceanogr. 10: 434-443Jonsson, P. R. (1986). Particle size selection, feeding rates and growth dynamics of marineplanktonic oligotrichous ciliates (Ciliophora: Oligotrichina). Mar. Ecol. Prog. Ser. 33:265-277Jonsson, P. R. (1989). Vertical distribution of planktonic ciliates- an experimental analysis ofswimming behaviour. Mar. Ecol. Prog. Ser. 52: 39-53Jonsson, P. R., Tiselius, P. (1990). Feeding behaviour, prey detection and capture efficiencyof the copepod Acartia tonsa feeding on planktonic ciliates. Mar. Ecol. Prog. Ser. 60: 35-44Kofoid, C. A., Campbell, A. S. (1939). Reports on the scientific results of the expedition tothe eastern tropical Pacific. The Ciliates. The Tintinnoinea. Bull. Mus. Comp. Zool.,Harvard 84: 1-473147Krainer, K. H. (1991). Contributions to the morphology, infraciliature and ecology of theplanktonic ciliates Strombidium pelagicum n. sp., Pelagostrombidium mirabile (Penard,1916) n. g., n. comb., and Pelagostrombidium fallax (Zacharias, 1896) n. g., n. comb.(Ciliophora). Euro. J. Protistol. 27: 60-70Landry, M. R., Lehner-Fournier, J. M. (1988) Grazing rates and behaviors of Neocalanusplumchrus: implications for phytoplankton control in the subarctic Pacific. Hydrobiologia167/168: 9-19Laval-Peuto, M., Brownlee, D. C. (1986). Identification and systematics of the Tintinnina(Ciliophora): evaluation and suggestions for improvement. Ann. Inst. oceanogr. 62: 69-84Laval-Peuto, M., Heinbokel, J. F., Anderson, R. 0., Rassoulzadegan, F., Sherr, B. (1986).Role of micro- and nanozooplankton in marine food webs. Insect Sci. Applic. 7: 387-395Laybourn-Parry, J. (1984). A functional biology of free-living protozoa. Univ. CaliforniaPress, BerkeleyLee, J. J., Small, E. B., Lynn, D. H., Bovee, E. C. (1985). Some techniques for collecting,cultivating and observing protozoa. In: Lee, J. J., Hutner, S. H., Bovee, E. C. (ed.) Anillustrated guide to the Protozoa. Soc. Protozool. Special Publ. Allen Press, Kansas, p. 1-7L,eegaard, C. (1915). Untersuchungen iiber einige planktonciliaten des Meeres. Nyt Mag. f.Naturv. 53: 1-37Legendre, L. (1990). The significance of microalgal blooms for fisheries and for the export ofparticulate organic carbon in the oceans. J. Plank. Res. 12: 681-699Legovic, T. (1989). Predation in food webs. Ecol. Model. 48: 267-276Lochte, K. (1991). Protozoa as makers and breakers of marine aggregates. In: Reid, P. C.,Turley, C. M., Burlcill, P. H. (ed.) Protozoa and their role in marine processes, NATOASI publication, Springer Verlag, New York, p. 237-246Lynn, D. H. (1975). The life cycle of the histophagous ciliate Tetrahymena corlissiThompson, 1955. J. Protozool. 22: 188-195Lynn, D. H., Montagnes, D. J. S., Riggs, W. (1987). Divider size and the cell cycle afterprolonged starvation of Tetrahymena corlissi. Microb. Ecol. 13: 115-127Lynn, D. H., Montagnes, D. J. S. (1988). Taxonomic description of some conspicuousspecies of Strobilidiine ciliates (Ciliophora: Choreotrichida) from the Isles of Shoals, Gulfof Maine. J. mar. biol. Ass. U.K. 68: 639-658Lynn, D. H., Montagnes, D. J. S., Small, E. B. (1988). Taxonomic descriptions of someconspicuous species in the family Strombidiidae (Ciliophora: Oligotrichida) from the Islesof Shoals, Gulf of Maine. J. mar. biol. Ass. U.K., 68: 259-276Lynn, D. H., Montagnes, D. J. S, Dale, T., Gilron, G. L., Strom, S. L. (1991). Areassessment of the genus Strombidinopsis Kent 1881 (Ciliophora, Choreotrichida) withdescriptions of four new planktonic species and remarks on its taxonomic phylogeny. J.mar. biol. Ass. U.K. 71: 597-612Lynn, D. H., Corliss, J. 0. (1991). Ciliophora: Chapter 5. In: Harrison, F. W. (ed.)Microscopic anatomy of invertebrates Vol. 1, 7. Wiley-Liss Inc., New York, p. 333-46148Lynn, D. H., Montagnes, D. J. S. (1991). Global production of heterotrophic marineplanktonic ciliates. In: Reid, P. C., Turley, C. M., Burki11, P. H. (ed.) Protozoa andtheir role in marine processes. NATO ASI publication, Springer Verlag, New York, p.281-307Mackas, D. L., Denman, K. L., Abbott, M. R. (1985). Plankton patchiness: biology in thephysical vernacular. Bull. Mar. Sci. 37: 652-674Maeda, M., Carey, P. G. (1985). An illustrated guide to the species of the familyStrombidiidae (Oligotrichida, Ciliophora), free swimming protozoa common in the aquaticenvironment. Bull. Ocean Res. Inst., Univ. Tokyo 19: 1-68Maeda, M. (1986). An illustrated guide to the species of the Families Halteriidae andStrobiliidae (Oligotrichida, Ciliophora), free swimming protozoa common in the aquaticenvironment. Bull. Ocean Res. Inst., Univ. Tokyo 21: 1-67Martin, A. J., Montagnes, D. J. S. (1993). Winter ciliates and their putative prey in a BritishColumbian fjord: with descriptions of six species. J. Euk. Microbiol. in pressMcManus, G. B., Okubo, A. (1991). On the use of surrogate food particles to measureprotistan ingestion. Limnol. Oceanogr. 36: 613-617Middlebrook, K., Roff, J. C. (1986). Comparison of methods for estimating annualproductivity of the copepods Acartia hudsonica and Eurytemora herdmani inPassamaquoddy Bay, New Brunswick. Can J. Fish. Aquat. Sci. 43: 656-664Montagnes, D. J. S., Lynn, D. H. (1987). A quantitative protargol stain (QPS) for ciliates:method description and test of its quantitative nature. Mar. Microb. Food Webs 2: 83-93Montagnes, D. J. S., Lynn, D. H., Roff, J. C., Taylor, W. D. (1988a). The annual cycle ofheterotrophic planktonic ciliates in the waters surrounding the Isles of Shoals, Gulf ofMaine: an assessment of their trophic role. Mar. Biol. 99: 21-30Montagnes D. J. S., Lynn, D. H., Stoecker, D. K., Small, E. B. (1988b). Taxonomicdescriptions of one new species and redescription of four species in the familyStrombidiidae (Ciliophora, Oligotrichida). J. Protozool. 35: 189-197Montagnes, D. J. S., Taylor, F. J. R., Lynn, D. H. (1990). Strombidium inclinatum n. sp.and a reassessment of Strombidium sulcatum Claparede and Lachmann (Ciliophora). J.Protozool. 37: 318-323Montagnes, D. J. S., Lynn, D. H. (1991). The taxonomy of the Choreotrichs, the majormarine planktonic ciliates with emphasis on the aloricate forms. Mar. Microb. Food Webs5: 59-74Montagnes, D. J. S., Lynn, D. H. (1993). A quantitative protargol stain (QPS) for ciliates andother protists. In: Kemp, P., Cole, J., Sherr, B., Sherr, E. (ed.) Current methods inaquatic microbial ecology. Lewis Publishers, Boca Raton, p. 229-240Montagnes, D. J. S., Berges, J. A., Harrison, P. J., Taylor, F. J. R. (1993). Estimatingcarbon, nitrogen, protein and chlorophyll a from cell volume in marine phytoplankton: acomparison of microscopic and electronic particle counting techniques and the effect offixation with acid Lugol's iodine. Limnol. Oceanogr. (submitted)Mailer, H., Geller, W. (1993). Maximum growth rates of aquatic ciliated protozoa: thedependence on body size and temperature reconsidered. Arch. Hydrobiol. 126: 315-327149Mullin, M. M., Stewart, E. F., Fuglister, F. J. (1975). Ingestion by planktonic grazers as afunction of concentration of food. Limnol. Oceanogr. 20: 259-261Nanny, D. L., (1980). Experimental ciliatology, and introduction to genetic anddevelopmental analysis on ciliates. John Wiley and Sons, TorontoNielsen, T. G., Richardson, K. (1989). Food chain structure of the North Sea planktoncommunities: seasonal variations of the role of the microbial loop. Mar. Ecol. Prog. Ser.56: 75-87Nielsen, T. G., Kiorboe, T. (1991). Effects of a storm event on the structure of the pelagicfood web with special emphasis on planktonic ciliates. J. Plank. Res. 13: 35-51Ohman, M. D., Snyder, R. A. (1991). Growth kinetics of the omnivorous oligotrich ciliateStrombidium sp. Limnol. Oceanogr. 36: 922-935Owen, R. W. (1989). Microscale and finescale variations of small plankton in coastal andpelagic environments. J. Mar. Res. 47: 197-240Pace, M. L., Bailiff, M. D. (1987). Evaluation of a fluorescent microsphere technique formeasuring grazing rates of phagotrophic microorganisms. Mar. Ecol. Prog. Ser. 40: 185-193Paranjape, M. A. (1980). Occurrence and significance of resting cysts in a hyline tintinnid,Helicostomella subulata (Ehre.) Jorgesen. J. exp. mar. Biol. Ecol. 48: 23-33Petz, W., Foissner, W. (1992). Morphology and morphogenesis of Strobilidium caudatum(Fromentel), Meseres corlissi n. sp., Halteria grandinella (Muller), and Strombidiumrehwaldi n. sp. (Protozoa, Ciliophora). J. Protozool. 39: 159-176Pomeroy, L. (1974). The ocean's food web: a changing paradigm. Biosci. 24: 499-504Porter, K. G., Pace, M. L., Nauwerk, A. (1979). Ciliate protozoa as links in planktonic foodchains. Nature 279: 563-565Putt, M., Stoecker, D. K. (1989). An experimentally determined carbon:volume ratio formarine "oligotrichous" ciliates from estuarine and coastal waters. Limnol. Oceanogr. 34:333-355Putt, M. (1990). Metabolism of photosynthate in the chloroplast-retaining ciliate Laboeastrobila. Mar. Ecol. Prog. Ser. 60: 271-282Putt, M. (1991). Development and evaluation of tracer particles for use in microzooplanktonherbivory studies. Mar. Ecol. Prog. Ser. 77: 27-37Rassoulzadegan, F. (1982). Dependence of grazing rate, gross growth efficiency and food sizerange on temperature in a pelagic oligotrichous ciliate Lohmanniella spiralis Leeg., fed onnaturally occurring particulate matter. Ann. Inst. oceanogr. 58(S): 177-184Raymont, J. E. G. (1983) Plankton and productivity in the oceans 2nd ed., Vol. 2Zooplankton. Pergamon Press, OxfordReid, P. C., John, A. W. (1978). Tintinnid cysts. J. mar. biol. Ass. U.K. 58: 551-557Reid, P. C., John, A. W. G. (1983). Resting cysts in the ciliate class Polyhymenophorea:phylogenetic implications. J. Protozool. 30: 710-713150Reid, P. C. (1987). Mass encystment of a planktonic oligotrich ciliate. Mar. Biol. 95: 221-230Repak, A. J. (1983). Suitability of selected marine algae for growing the marine heterotrichciliate Fabrea sauna. J. Protozool. 30: 52-54Riemann, B., Sorensen, H. M., Bjornsen, P. K., Horsted, J. S., Jensen, L. M., Nielsen, T.G., Sondergaard, M. (1990). Carbon budgets of the microbial food web in estuarineenclosures. Mar. Ecol. Prog. Ser. 65: 159-170Rivier, A., Brownlee, D. C., Sheldon, R. W., Rassoulzadegan, F. (1985). Growth ofmicrozooplankton: a comparative study of bactivorous zooflagellates and ciliates. Mar.Microb. Food Webs 1: 51-60Rublee, P. A., Gallagose, G. L. (1989). Use of fluorescently labeled algae (FLA) to estimatemicrozooplankton grazing. Mar. Ecol. Prog. Ser. 51: 221-227Schmidt-Nielsen, K. (1980). Animal physiology: adaptation and environment. CambridgeUniv. Press, New YorkScott, J. M. (1985). The feeding rates and efficiencies of a marine ciliate, Strombidium sp.,grown under chemostat steady-state conditions. J. Exp. Mar. Biol. Ecol. 90: 81-95Sheldon, R. W., Nival, P., Rassoulzadegan, F. (1986). An experimental investigation of aflagellate-ciliate-copepod food chain with some observations relevant to the linear biomasshypothesis. Limnol. Oceanogr. 31: 184-188Sherr, E. B., Sherr, B. F., McDaniel, J. (1991). Clearance rates of <6 Am fluorescentlylabeled alga (FLA) by estuarine protozoa: potential grazing impact of flagellates andciliates. Mar. Ecol. Prog. Ser. 69: 81-92Sherr, E. B., Sherr, B. F. (1993). Protistan grazing rates via uptake of fluorescently labeledprey. In: Kemp, P., Cole, J., Sherr, B., Sherr, E. (ed.) Handbook of methods in aquaticmicrobial ecology. Lewis Publishers, Boca RatonSieracici, M. E., Hass, L. W., Caron, D. A., Lessard, E. J. (1987). Effects of fixation onparticle retention by microflagellates: underestimation of grazing rates. Mar. Ecol. Prog.Ser. 38: 251-258Silver, M. W., Gowing, M. M., Brownlee, D. C., Corliss, J. 0. (1984). Ciliated protozoaassociated with oceanic sinking detritus. Nature 309: 246-248Sime-Ngando, T., Juiper, K., Vezina, A. (1992). Ciliated protozoan communities over CobbSeamount: increase in biomass and spatial patchiness. Mar. Ecol. Prog. Ser. 89: 37-51Skogstad, A., Granskog, L., Klaveness, D. (1987). Growth of freshwater ciliates offeredplanktonic algae as food. J. Plank. Res. 9: 503-512Small, E. B., Lynn, D. H. (1985). Phylum Ciliophora. In: Lee, J. J., Hutner, S. H., Bovee,E. C. (ed.) An illustrated guide to the Protozoa. Soc. Protozool. Special Publ. AllenPress, Lawerence, Kansas, p. 395-575Smetacek, V. (1981). The annual cycle of protozooplankton in Kiel Bight. Mar. Biol. 63: 1-11Smetacek, V. (1984). Growth dynamics of a common Baltic protozooplankter: the ciliategenus Lohmanniella. Limnologica 15: 371-376Smith-Sonneborn, J. (1981). Genetics and aging in protozoa. Int. Rev. Cytology 73: 319-354151Snyder, R. A., Ohman, M. 0. (1991). Description of a new species of Strombidinopsidae(Ciliophora: Choreotrichida) from coastal waters of Southern California, U. S. A. Trans.Am. Microsc. Soc. 110: 237-243Spain, J. D. (1982). BASIC microcomputer models in biology. Addison-Wesley PublishingCo., LondonSteele, J. H., Gamble, J. C. (1982). Predation control in enclosures. In: Grice, G. D, Reeve,M. R., (ed.) Marine mesocosms, biological and chemical research in experimentalecosystems. Springer-Verlag, New YorkStoecker, D. K. (1984). Particle production by planktonic ciliates. Limnol. Oceanogr. 29:930-940Stoecker, D. K. (1988). Are marine planktonic ciliates suspension-feeders? J. Protozool. 35:252-255Stoecker, D. K., Davis, H. L., Anderson, D. M. (1984). Fine scale spatial correlationsbetween planktonic ciliates and dinoflagellates. J. Plank. Res. 6: 829-842Stoecker, D. K., Evans, G. T. (1985). Effects of protozoan herbivory and carnivory in amicroplanlcton food web. Mar. Ecol. Prog. Ser. 25: 159-167Stoecker, D. K., Sunda, W. G., Davis, L. H. (1986). Effects of copper and zinc on twociliates. Mar. Biol. 92: 21-29Stoecker, D. K., Egloff, D. A. (1987). Predation by Acartia tonsa Dana on planktonic ciliatesand rotifers. J. Exp. Mar. Biol. Ecol. 110: 53-68Stoecker, D. K., Silver, M. W., Michaels, A. E., Davis, L. H. (1988). Obligate mixotrophyin Laboea strobila, a ciliate which retains chloroplasts. Mar. Biol. 99: 415-423Stoecker, D. K., Silver, M. W. (1990). Replacement and aging of chloroplasts in Strombidiumcapitatum (Ciliophora: Oligotrichida). Mar. Biol. 107: 491-502Stoecker, D. K., Michaels, A. E. (1991). Respiration, photosynthesis and carbon metabolismin planktonic ciliates. Mar. Biol. 108: 441-447Stout, J. D. (1980). The role of protozoa in nutrient cycling and energy flow. In: Alexander(ed.), Advances in microbial ecology Vol. 4, New York, Plenum Press, p. 1-50Strom, S. L., Welschmeyer, N. A. (1991). Pigment-specific rates of phytoplankton growthand microzooplankton grazing in the open subarctic Pacific Ocean. Limnol. Oceanogr. 36:50-63Takahashi, K., Aizawa, Y. (1990). Excystment of tintinnid ciliates from marine sediment.Bull. Plankk. Soc. Japan 36: 137-139Takahashi, M., Seibert, D. L., Thomas, W. H. (1977). Occasional blooms of phytoplanktonduring summer in Saanich Inlet, B. C., Canada. Deep Sea Res. 24: 775-780Taniguchi, A., Takeada, Y. (1988). Feeding rate and behavior of the tintinnid ciliate Fayellataraikaensis observed with a high speed VTR system. Mar. Microb. Food Webs 3: 21-34Taylor, G. T. (1982). The role of pelagic heterotrophic protozoa in nutrient cycling: a review.Ann. Inst. oceanogr. 58(5): 227-241152Taylor, W. D. (1978). Maximum growth rate, size and commonness in a community ofbactivorous ciliates. Oecologia 36: 263-272Taylor, W. D., Shuter, B. J. (1981). Body size, genome size, and intrinsic rate of increase inciliated protozoa. Amer. Nat. 118: 160-172Thompson, P. A., Harrison, P. J. (1992). Effects of monospecific algal diets of varyingbiochemical composition on the growth and survival of Pacific oyster (Crassostrea gigas)larvae. Mar. Biol. 113: 645-654Tiselius, P. (1989). Contribution of aloricate ciliates to the diet of Acartia clausi andCentropages hamatus in coastal waters. Mar. Ecol. Prog. Ser. 56: 49-56Turley, C. M., Newell, R. C., Robins, D. B. (1986). Survival stratigies of two small marineciliates and their role in regulating bacterial community structure under experimentalconditions. Mar. Ecol. Prog. Ser. 33: 59-70Turner, J. T., Graneli, E. (1992). Zooplankton feeding ecology: grazing during enclosurestudies of phytoplankton blooms from the west coast of Sweden. J. Exp. Mar. Biol. Ecol.157: 19-31Verity, P. G. (1985). Grazing, respiration, excretion, and growth rates of tintinnids. Limnol.Oceanogr. 30: 1268-1282Verity, P. G. (1986). Grazing of phototrophic nanoplankton by microzooplankton inNarragansett Bay. Mar. Ecol. Prog. Ser. 29: 105-115Verity, P. G. (1988). Aggregation patterns of ciliates from natural assemblages in response todifferent prey. Mar. Microb. Food Webs 5: 115-128Verity, P. G. (1991a). Measurement and simulation of prey uptake by marine planktonicciliates fed plastidic and aplastidic nanoplankton. Limnol. Oceanogr. 36: 729-750Verity, P. G. (1991b). Aggregation patterns of ciliates from natural assemblages in response todifferent prey. Mar. Microb. Food Webs 5:115-128Verity, P. G., Villareal, T. A. (1986). The relative food value of diatoms, dinoflagellates,flagellates and cyanobacteria for tintinnid ciliates. Arch. Protistenkd. 131: 71-84Verity, P. G. Robertson, C. Y., Tronzo, C. R. Andrews, M. G., Nelson, J. R., Seracki, M.E. (1993). Relationships between cell volume and the carbon and nitrogen content ofmarine photosynthetic nanoplankton. Limnol. Oceanogr. 37: 1434-1446Wailes, G. H. (1943). Canadian Pacific Fauna 1. Protozoa if Ciliata lg Suctoria. Fish. Res.Bd. Can. 1-46Wiadnyana, N. N., Rassoulzadegan, F. (1989). Selective feeding of Acartia clausi andCent ropages typicus on microzooplankton. Mar. Ecol. Prog. Ser. 53: 37-45Wichterman, R. (1986). The biology of Paramecium. Plenum Press, New YorkWickham, S. A., Gilbert, J. J., Berninger, U-. G. (1993). Effects of rotifers and ciliates onthe growth and survival of Daphnina. J. Plank. Res. 15: 317-334153APPENDIX 1THE GROWTH MEDIUM USED IN THIS STUDY TO GROW BOTH CILIATES ANDPHYTOPLANKTONBelow is the recipe for HESNW1 medium (modified) used by the Northeast PacificCulture Collection (NEPCC, Department of Oceanography, University of British Columbia,Vancouver, British Columbia) to grow their phytoplankton stocks. It was also the medium Iused for all my ciliates and phytoplanlcton cultures (unless otherwise stated). The sea waterused to make the medium was obtained from West Vancouver Laboratory, Biological ScienceBranch, Department of Fisheries and Oceans, 4160 Marine Dr., West Vancouver; the waterwas pumped from Burrard Inlet, 100 m from the shore at a depth of 15 m. The final salinityof the medium was between 27-29 0/09.Stock^, FinalENRICHMENT STOCKS^Conc. (g L-')^Conc.(micromoles L-1)1) NaNO3 46.67 549.092) Na2g1ycerophosphate 6.67 21.793)^Na25iO3.9H20 30.00 105.60*^4)^Na2EDTA.2H20 3.64 9.81Fe(NH4)2(504)2.6H20 2.34 5.97FeC13.6H20 0.16 5.92 x 10-15) Mn504.4H20 0.54 2.42ZnSO4.7H20 0.073 2.54 x 10-1Co504.7H20 0.016 5.69 x 10-2Na2Mo04.2H20 0.126 5.2 x 10-1**^Na2EDTA.2H20 1.89 5.056) H3B03 3.80 61.467)^Na2Se03 0.00173 1.0 x 10-2VITAMIN STOCKThiamine 0.1 2.97 x 104Vitamin B12 0.002 1.47 x 10-3Biotin 0.001 4.09 x 10-3154Vitamin Stock should be stored frozen. Enrichment stocks can be refrigerated.* Add before the trace metals. Heat solution 4) to dissolve the iron. Adjust pH to 6.Preparation of medium1) Filter natural seawater through a 0.45 micrometer filter.2) Autoclave medium flask with distilled water before first use.3) Add 1 mL of each enrichment stock and 1 mL of vitamin stock L-1 of seawater.4) To prevent precipitation during autoclaving add: 0.12 g NaHCO3 and 1.44 mL 1N HC1 L-1 of seawater.5) Autoclave 20 min (for 1 L medium).6) Let stand for at least 2 d before use (pH should be 8.2).7) Medium may be filtered aseptically through a glass fiber filter to remove orange-brownprecipitate.1Harrison, P. J., Waters, R. E., Taylor, F. J. R. (1980). A broad spectrum artificial mediumfor coastal and open ocean phytoplankton. J. Phycol. 16: 28-35155APPENDIX 2CONSUMPTION AND FILTRATION: THEIR DERIVATION AND APPLICATION.Below is a description of the derivation of Michaelis-Menten-like kinetics and theirapplication to ciliate feeding. I have only described ciliate feeding. However, assuming thatgrowth rate is a constant function of grazing (i.e. a constant gross growth efficiency, then thisderivation should also apply to growth rate. The only modification to this would be that unlikefeeding rate, growth rates may be negative (net mortality). Therefore, growth rates are likelyto have non-zero intercepts.Ciliate feeding has been modeled using Michaelis-Menten type equations to describe the"basic functional response curve" (Fenchel 1986, but see Holling 1959). Consumption rate(E) is the number of prey (P) consumed by a ciliate (C) per unit time (pc-lr1). Ifconsumption is a combined process of non-selective prey encounter and ingestion then, it iscontrolled by two rate constants: encounter rate and ingestion rate. Assuming that: 1) particlecapture is directly proportional to particle concentration; 2) encounter area and particleretention are invariant with particle concentration; and 3) ingestion takes a finite time duringwhich further ingestion can not take place, then the consumption of particles may be depictedas[P] + [C] ^--- [PC] ^—> [C]ki^k2where [P] is prey concentration, [C] is ciliate concentration, k1 (VC4(1) and k2 (pc1t4)are rate constants (where t is time and V is volume).In a steady state at concentration [P] each ciliate has a constant probability of capturinga prey. Then the velocities of these reactions are V1 and V2V1 = ki[P][C] or [C] = V i/ki[P]V2 = k2[PC][Ptotal] = [P] + [PC][Ctotal] = [C] ± [PC]Assume [total] ' [P] thus, [PC] is a small portion of [P]+[PC]In a steady state V1 = V2 = V = consumption rate (E)156E = k2([Ct0tal]) - [C]E = k2aCt0tai] - E/k [P])E + k2E/k1[P]) = k2([Ctotai] or E(1 + k2/k1 [P]) = k2([Ctotal])E = k2aCtotai]/((1 + k2/k1[P]))multiplying by k1[P]E = ki[P]k2aCtotag/(k2 + ki[P])dividing by k1E = [P]k2([Ctotai]/(k2/ki + [P])dividing by total ciliates to get E as "per ciliate"E = [P]k2/(k2/k1 + [P])wher k2 = Em = maximum ingestion rate and k is k2/k1E[p] = Em [1:]/(k+[P])^ (A2.1)If there is a minimum threshold on food (i.e. at a certain low food concentration ([P'])where feeding stops, thenE[p] = Em([P]-[P'])/(k+([P]-[P']))^ (A2.2)Jonsson (1986) has conducted feeding experiments on the ciliate Strobilidium (=Lohmanniella) spiralis; his results can be explained using the above terminology. Typically,the velocity of the first reaction, determined by kb was much slower than the second reaction,determined by k2. However, when food concentration was high k2 determined the rate ofingestion. For S. spiralis, eating 14.4 gm prey, ingestion rate was one cell -15 s-1 (k2 =0.067 prey C-1s-1). Jonsson found that the maximum clearance rate varied with prey size.Therefore, the "filter" did not have the same efficiency for all prey.Equation A2.1 is based on the assumptions that k1 and k2 are invariant. For aparticular prey it is likely that k2, the time it takes a ciliate to ingest ("handle" andphagocytose) prey is constant; at least it is for Favella (Taniguchi and Takeda 1988).However, kb the rate at which filtration occurs may change.Of course, the whole notion of filtration rate may be erroneous since oligotrichs maynot filter feed. The presence of striae (Capriulo et al. 1986) and other extrusomes (Montagnes157et al. 1988) used for prey capture suggests raptorial feeding. Chemosensory abilities alsosuggest a selectivity, and thus non-filtering, mechanisms (Verity 1988). Therefore, theamount of food ingested could have little to do with the amount of medium "filtered". Non-filter feeding does not necessarily discount the model developed above (Eq. A2.1). It doeshowever, imply that k1 may vary.Studies have examined the clearance rates (k1?) for planktonic ciliates. They rangefrom 1 to 100 Al h-1 (see Jonsson 1986, Stoecker 1988). For the tintinnid Fayella the timebetween prey capture and complete engulfment (k2?) was -700 ms (Taniguchi and Takeda1988). Jonsson (1988) has recommended that more data are needed to estimate theseparameters; especially for the small, <20 Am, oligotrichs which seem to be ubiquitous. Ingeneral, more studies need to be conducted to determine if the Michaelis-Menten relation holdstrue and to determine if k1 and k2 vary.LITERATURE CITEDCapriulo, G. M., Traveras, J., Gold, K. (1986). Ciliate feeding: effect of food presence orabsence on occurrence of striae in tintinnids. Mar. Ecol. Prog. Ser. 30: 145-158Fenchel, T. (1986). Protozoan filter feeding. In: Corliss, J. 0., Patterson, D. J. (ed) Progressin Protistology Vol. 1, Biopress Ltd., Bristol, p. 66-113Holling, C. S. (1959). Some characteristics of simple types of predation and parasitism. Can.Entomol. 41: 385-399Jonsson, P. R. (1986). Particle size selection, feeding rates and growth dynamics of marineplanktonic oligotrichous ciliates (Ciliophora: Oligotrichina). Mar. Ecol. Prog. Ser. 33:265-277Montagnes D. J. S., Lynn, D. H., Stoecker, D. K., Small, E. B.. (1988). Taxonomicdescriptions of one new and redescription of four species in the family Strombidiidae(Ciliophora, Oligotrichida). J. Protozool. 35: 189-197Stoecker, D. K. (1988). Are marine planktonic ciliates suspension-feeders? J. Protozool. 35:252-255Taniguchi, A., Talceada, Y. (1988). Feeding rate and behavior of the tintinnid ciliate Fayellataraikaensis observed with a high speed VTR system. Mar. Microb. Food Webs 3: 21-34Verity, P. G. (1988). Aggregation patterns of ciliates from natural assemblages in response todifferent prey. Mar. Microb. Food Webs 5: 115-128158APPENDIX 3THE ANNUAL CYCLE OF OLIGOTRICHS AND <10 Am PHYTOPLANKTON INSECHELT INLET, A BRITISH COLUMBIAN FJORDAL SYSTEMWork conducted in conjunction with: R. Haigh 1, F. J. R. Taylorl, and T. F. Sutherland21Departni5nt of Oceanography, University of British Columbia, Vancouver, B.C., V6T 1Z4,Canada. Dept. Oceanography, Dalhousie University, Halifax, N.S., B3H 4J1, CanadaINTRODUCTIONThis appendix provides a description of the changes in abundance and biomass ofciliates and their putative prey (<10 Am phytoplankton) over 1.3 years, in Sechelt Inlet.Sechelt Inlet is typical of many fjordal inlets and should therefore provide representativeestimates of ciliate and prey abundance and biomass for much of British Columbia's waters.This analysis is strictly exploratory; its goals were to establish initial ("typical") ciliate andprey concentrations for the population model presented in Chapter 9 and to recognize thenaturally occurring range of these two trophic groups.Sechelt Inlet (Fig. A3.1) is a southern British Columbian fjord located 40 km northwestof Vancouver. It consists of a main inlet and two adjoining ones, Narrows and Salmon Inlets;these are collectively referred to as the Sechelt Inlet complex. The main inlet is 29 km long,on average 1.2 km wide, has a maximum depth of 300 m, and has a shallow (14 m) sill at itsentrance. Three distinct water layers exist in the inlet: a surface layer (-5 m) of low-salinitywater (due to precipitation and river runoff); an intermediate layer occupying a depth between-5 and 65 m; and a deep layer of resident water, characterized by uniform temperature andsalinity (Lazier 1963). Estuarine circulation in Sechelt Inlet is slight due to the relatively lowfreshwater drainage (110 m3 s-1, Pickard 1961), and depends largely on the intrusion of atidal jet that flushes the indigenous deep water zone (Lazier 1963).Temperature, salinity, chlorophyll a, nitrate, ammonium and phosphate were measuredover the 18 months of this study. The method of analysis and results of these measurementsare presented elsewhere (Haigh et al. 1992).159MATERIALS AND METHODSWater samples were collected from 6 stations in Sechelt Inlet (49° 40' N, 123° 45' W,Fig. A3.1) between April 1989 and September 1990. Samples were taken using the segmentedintegrated pipe sampler (SIPS, Sutherland et al. 1992). This system is composed of eightsegments of PVC tubing (inside diameter 3.75 cm). The segments are connected by snap-together joints and have a closing device at the bottom of each segment. The two surfacesegments are 1.5 m long and the remaining six tubes are 3.0 m (total length 21 m). Thesegments were connected together and lowered into the water to collect eight integratedsamples. Each segment's contents were released into a bucket for sub-sampling aboard theboat. A sample was also taken from 30 m with a 1 L Nislcin bottle. For details of thesampling procedure see Haigh et al. (1992) and Sutherland et al. (1992)Subsamples (125 mL) from the bucket were preserved with 1% acid Lugol's and ciliateand phytoplankton abundances were analyzed using the UtermOhl technique (Hasle 1978).Depending on the biomass, 2.5 to 10 mL were settled in a counting chamber for._12 h.Ciliate prey (<10 pm) were counted at 500x along one or two 10-20 mm transects. Ciliateswere counted at either 240x along six transects across the chamber diameter (25-26 mm) or at95x, examining the entire chamber bottom. Oligotrich ciliates (see Chapter 2) were dividedinto three groups: <25 pm, 25-50 pm, and >50 pm in diameter. The counts were thentransformed to cells mL4 and were integrated over depths, assuming that samples fromsegments of the SIPS were representative of the entire depths sampled and the sample bottlerepresented the bottom 9 m.The plankton data were converted to carbon using species specific cell volumes.Where possible, representative cells were measured for each species and the volume wascalculated using equations for simple geometric shapes. For a number of the less commonspecies, dimensions were extracted from the literature (dinoflagellates: Dodge 1982;nanoflagellates: Watanabe 1976, Throndsen 1988). Phytoplankton volumes were converted tocarbon using the relation: C (pg) = 0.109 * (live cell volume , pm3)°.991 and recognizing that160Lugol's fixed cells are 80% of live volume, from Montagnes et al. (submitted). Ciliatevolumes were converted to carbon using the relation: C (pg) = 0.19 * (volume, p,m3), fromPutt and Stoecker (1989).The data were presented in two ways: 1) as the change with time of average abundance(cells mL-1) or biomass (jig carbon mL-1) from the integrated water column (either 6 or 30 m,see below) and 2) as correlations between prey and ciliate abundance or biomass for everysample taken over the 18 month sampling period (n = 1056). As stated above, these datawere used as baseline estimates for the model presented in Chapter 9.RESULTS AND DISCUSSIONCiliate and prey abundance and biomass were fairly constant throughout the year,although they both tended to be higher in the summer (Figs. A3.2, A3.3). Both prey andciliate abundance and biomass occasionally increased by one or two orders of magnitude aboveseasonal averages (Figs. A3.2, A3.3, A3.4). Possibly these ciliate pulses were due topopulation growth (blooms).Since the top -5 m of the inlet was probably physically distinct from the deeper waters(see Materials and Methods), the abundance and biomass of organisms were exclusivelyexamined from the top 3 segments of the SIPS (top 6 m), to better elucidate populationdynamics. Average integrated ciliate and prey abundance and biomass in the top 6 m weretypically higher than those in the top 30 m but showed similar seasonal trends (Figs. A3.5,A3.6).The abundance and biomass of ciliates that were 25-50 Am in diameter (similar in sizeto those examined in Chapters 3-7) were examined. Over the top 30 m of the inlet, bothabundance and biomass of 25-50 Am ciliates varied by up to 3 orders of magnitude (Fig.3A.4). However, in the top 6 m the average integrated biomass and abundance of 25-50 Amciliates rarely varied by more than one order of magnitude (Figs. A3.7-8).On average the ratios of total ciliate:prey abundance and biomass were near 10:1000mL4 and 104:104 ng C, respectively. The ratios of 25-50 Am ciliate:prey abundance and161biomass were near 1:1000 mL-1 and 3x103:104, respectively (Fig. A3.4). These may beconsidered "typical" levels for modeling (see Chapter 9).Many of the parameters observed in this study (e.g. the ciliate:prey ratio, the ciliateand prey concentrations, and the higher ciliate numbers in upper waters) agree with thoseobserved in other cold water studies (Revelante and Gilmartin 1987, Andersen and Sorensen1986, Haigh and Taylor 1991, Lynn and Montagnes 1991 [and references within], Verity andVernet 1992, Martin and Montagnes 1993).Since ciliates have rapid numerical responses, they may potentially exhibit a positivecorrelation with their prey (Stoecker et al. 1984). Studies support this notion: significantcorrelations existed between abundances of naked ciliates and <20 Am prey in an estuary inMaine (Revelante and Gilmartin 1987), and on a global level ciliate and phytoplanktonbiomass were positively correlated (Lynn and Montagnes 1991). The Sechelt data also showsignificant (a = 0.05) positive correlations (but not good predictors r2 = 0.38-0.55) betweenciliate abundance and biomass and their prey (Fig. A3.4). Further, there were some instanceswhere ciliate and prey populations simultaneously increased, but there did not appear to beregular coupling in the abundance or biomass of ciliates and prey (Figs. A3.2-3, A3.5-8), asmight be expected (Andersen and Sorensen 1986). As there are undoubtedly periods whereblooms of ciliates have grazed down their prey, but have not crashed themselves, there shouldalso be times when ciliates and their prey are negatively correlated. Such situations can beseen in the output of the model presented in Chapter 9. Given that the Sechelt data werecollected weeks to months apart and ciliate-prey interactions likely occur on the order of days(Andersen and Sorensen 1986, Chapter 9), it is not surprising that such coupled trends, if theyexisted, were not obvious. The data do provide an indication of the range and "typical"concentrations of these two trophic groups in British Columbian coastal waters. In Chapter 9,these values have been compared to the output of the model.—49`30"NOM/—49°35'NSechelt12335W* LOCATION OFSECHELT INLETCanadaVancouverPacificOcean4945NSealeIt:'Rapids:•..^.NMISecheItInletSalmon Inlet.^... Nine Mile Pt•• PorpoiseBoy123 55W162Fig. A3.1. Sechelt Inlet Complex with station locations (1-6).1635040302010a 0 phytoplankton-^• ciliates•••■■••••05000co 4000300020000 100005000ci 40004030200^" 00 N DIJIFIMI AiMjJ1JiA S1989^. 1990Time (months)Fig. A3.2. The seasonal cycle of ciliate and small (<10 Am) phytoplankton abundance (numbers rnL.as anaverage from integrating a 30 m water column) at 6 sites in Secbelt Inlet (a-f = 1-6, respectively; see Fig. A3.1).Open circles represent phytoplankton numbers. Solid circles represent ciliate numbers. Values above or besidepeaks off the panel indicate the magnitude of these measurements.r--iP-1 3000 -O 2000• 10005000 ^40003000^02000 -1000 - / alikvioos° A:14'J J IA'S5000400030002000100005000I^40003000" 200010000CD 5000c) 40003000C7:5 200010004030201005040..401030200 150-I 40302010• •  n14388 1/4'50^ 50-I 40- 30- 2010 I050164.-4------.^0^I^50  50C._) 20 -)--,to 0^ 0^1_1E 30 •.--• •1 40 - b ^1 4010^X.,- • ov •./0  ^• ^S.^20I^I 3010 '-150 •^ 50^1,--140 - C - 4020^- 30 C.-)C/1 30-ril^• 111---..^- 20c'd tall54 0/1 \CD- n Olb 10 —^it.4-CCI---- e--e--14 `-et----------2M^0C)--4)(:)..77." 0o 50  50•r■4^ ril^,-.0 40 40 71^30 30 CdO 20^ 20-4-) 10 10^0ke'^ • r—i^0  ^ 0^,Ct^50  50cti,--.1 40 - e - 40^Q.)30-^ •^ - 30 cd• •0^ • r••••4_,..) 20 - \ ZN^• -20 f•.•1•••••^ • t•I^›..110• •^1 0 0oQ-4 502o -^ \lo4030 •5040302010-a 0 phytoplan.kton.-^• ciliates•- „ •\ •^\OC)   50- 40• 30").^20s,<:„0,toste. 0^  09 9 50403020100/^10 •V •^ 0/50o •AIMJ•J A SOND J FMIA'NeJ J 'A 1 S1989 Time (months) 1990Fig. A3.3. The seasonal cycle of ciliate and small (<10 um) phytoplankton biomass (ng, carbon mL as anaverage from integrating a 30 m water column) at 6 sites in Sechelt Inlet (a-f = 1-6, respectively; see Fig. A3.1).Open circles represent phytoplankton numbers. Solid circles represent ciliate numbers. Values above or besidepeaks off the panel indicate the magnitude of these measurements.•1 0^1: r = 0.5.—•^ 47 .^a7—.--, 1 0 2 r • 1 •.. 1 •^•t. j'A.0 An•• . •. #.•","• ...•• lbet•• •^•• III• • • V •••• •■• qt..INN• • • • ••I•^• • =1,0 •• • ••031 0 sOl 3.••% • • .• •0 0 iO4V3 Oa••▪ •• • • % •■•••▪ •• • •• • #• • . •••• .:•••• •• •• •^•• • •101^102^1 03^1 04^105Prey abundance (mL )102^iO3^104^o5^o5Prey biomass (ng C mL )= 0.410• . • ••. 46, ' •• .•7: •1+0 a. •••• • •• • • •^ •I 1 1• • ASE...e.:1:::•• • • •• NO MIIIIMMI•M•• 4WD•• • • •••••^•• • • • MD• a^• • • •• ••. ••••: _2- r = 0.383••• *OP • •••• ••■• • • •^Ow..^•• • •• .•11.%^ • •••• •• as• •• •^•4. • • •• • ••••• • •^•• •••■ ••• • • •••^• Oa • • 11. •• • • • •. .^I•. •.1 •165Fig. A3.4. Four correlations between ciliates and their putative prey (<10 gm phytoplankton) using data fromevery sample taken over the 18 month sampling period (n = 1056). a, the abundance of all oligotrich ciliates vs.the prey abundance; b, the abundance of 25-50 pm (diameter) oligotrich ciliates vs. the prey abundance; c, thebiomass of all oligotrich ciliates vs. the prey biomass; the biomass of 25-50 pm (diameter) olieotrich ciliates vs.the prey biomass The r2 values for each regression are presented in the appropriate panels; all four correlationswere significant at a = 0.05.o phytoplankt on• ciliates6000E 40002000"-**,---44 ^01000080006000400020000100008000 - d166100806040200100SO604020T0100806040 "."-'20^C)0^C-)100C-1r-I•crict500Le,10000(-6 80006000oP.4 4000-4-3 2000010000P-1 80006000400020000TimeFig. A3.5. The seasonal cycle of ciliate and small (<10 m) phytoplankton abundance (numbers ml--1 as anaverage from integrating 16 m water column) at 6 sites in Sechelt Inlet (a-f 1-6, respectively; see Fig. A3.1)Open circles represent phytoplankton numbers. So lid circles represent ciliate numbers. Values above or besidepeaks off the panel indicate the magnitude of these measurements.1000080006000400020000000080008060004000200060-L 40.4 0OA •ori^ 1A M J 'J A S'0 N D'JIF19898060V4IY^20• 4 • 0M1 A'^J J 'A SI1990(months)1200014)\1 81000• il 40^ 04 3 5 9 10001675040302010o phytoplankton• ciliateso• ci,"0*050^40-/126 ip^30 -2010 -/0\trif49^3020co 100• 5040▪ 3020O 100ct30-20100 ^50 ^4030-20 -^/0\1(0) -•___•••o4/-cAMJIJiASIO N'DIJ'F'Ml A10080604020• 0100806040.^40 ,..4.."'"%%! 2o I• 0^1-4,..100806040 tzi)204,t 0”rD•rw91980emO? 60\.q 40./.••^Sio 20J1J'AISI05040 - e^ 10960I 402001041^-^ 10068004020^ 01001989^Time (months) 1990Fig. A3.6. The seasonal cycle of ciliate and small (<10 gm) phytoplankton biomass (ne carbon mL as anaverage from integrating a 6 m water column) at 6 sites in Secheit Inlet (a-f = 1-6, respectively; see Fig. A3.1).Open circles represent phytoplankton numbers. Solid circles represent ciliate numbers. Values above or besidepeaks off the panel indicate the magnitude of these measurements.40000_;_) 2000010000Q-4 8000P-4168252015• <°C;^ .47tP...,40ty^ • 0- 10525200 /^15/ ‘0C 10^,--4<em.-744/ \01,• 5025201510502520 tO151050^czA60004000 • \ /i2000 r^_go Ofkic.20,4_..4/ • IFoA M J 'J AISIO'N1DIJ'FMAMJiJiAIS1989^ 1990Time (months)Fig. A3.7. The seasonal cycle of 25-50 gm (diameter) ciliate and small ( <10 Arn) phytoplankton abundance(numbers mL as an average from integrating a 6 m water column) at 6 sites in S=helt Inlet (a-f = 1-6,respectively; see Fig. A3.1). Open circles represent pbytoplankton numbers. Solid circles represent ciliatenumbers. Values above or beside peaks off the panel indicate the magnitude of these measurements.100008000600040002000080006000r--4 400020000•'-1000080000 60004000CO 20000100008000• 6000r..4 4000200001000080006000phytoplanktonciliates252015105252015105Ck.)•■■)CO•r•■•4• •••••4L.)169501-4 40302010tzt3 05040ri-/ 30 -rnCt 20 - 0o 4-4^10• 5030-• 20 --d 55 \0• •in 40-4-) 10V 410-i50 ^40 -eC-1130010▪ 2005040 - f3020105040302010o ^AM JIJIAIS 0 N'DIJa••• 5• I 0iFMAMJJAS• 0 phytoplankton• ciliates•109• • \ 20,• • \ P 15e 102 5025/\60r\-::: 2015c•104 2'52015105AI *I"  0310\ • 25\IL 207 15OK^101989^ 1990Time (months)Fig. A3.8. The seasonal cycle of 25-50 A m (diameter) ciliate and small (<10 A m) phytoplankton biomass (ngcarbon^as an average from integrating a 6 m water column) at 6 sites in Sechelt Inlet (a-f = 1-6,respectively; see Fig. A3.1). Open circles represent phytoplankton numbers. Solid circles represent ciliatenumbers. Values above or beside peaks off the panel indicate the magnitude of these measurements.170LITERATURE CITEDAndersen, P., Sorensen, H. M. (1986). Population dynamics and trophic coupling in pelagicmicroorganisms in eutrophic coastal waters. Mar. Ecol. Prog. Ser. 33: 99-109Dodge, J. D. (1982). Marine dinoflagellates of the British Isles. Her Majesty's StationeryOffice, LondonHaigh, R., Taylor, F. J. R. (1991). Mosaicism of microplankton communities in the northernStrait of Georgia, British Columbia. Mar. Biol. 110: 301-3 14Haigh, R., Taylor, F. J. R., Sutherland, T. E. (1992). Phytoplankton ecology of Sechelt Inlet,a fjord system on the British Columbia coast. I. General features of the nano- andmicroplankton. Mar. Ecol. Prog. Ser. 89: 117-134Hasle, G. R. (1978). Using the inverted microscope. In: Sournia, A. (ed.) PhytoplanktonManual. Monographs on Oceanographic Methodology 6, UNESCO, p. 191-196Lazier, J. R. N. (1963). Some aspects of the oceanographic structure in the Jervis Inletsystem. M. Sc. thesis, Univ. British ColumbiaLynn, D. H., Montagnes, D. J. 5. (1991). Global production of heterotrophic marineplanktonic ciliates. In: Reid, P. C., Turley, C. M., Burkill, P. H. (ed.) Protozoa andtheir role in marine processes, NATO ASI publication, Springer Verlag, New York p.281-307Martin, A. J., Montagnes, D. J. 5. (1992). Winter ciliates and their putative prey in a BritishColumbian fjord: with descriptions of six species. J. Euk. Microbiol. in pressMontagnes, D. J. S., Berges, J. A., Harrison, P. J., Taylor, F. J. R. (1993). Estimatingcarbon, nitrogen, protein and chlorophyll a from cell volume in marine phytoplankton: acomparison of microscopic and electronic particle counting techniques and the effect offixation with acid Lugol's iodine. Limnol. Oceanogr. (submitted)Pickard, G. L. (1961). Oceanographic features of inlets in the British Columbia mainlandcoast. J. Fish. Res. Board Can. 18: 907-999Putt, M., Stoecker, D. K. (1989). An experimentally determined carbon:volume ratio formarine "oligotrichous" diiates from estuarine and coastal waters. Limnol. Oceanogr. 34:333-355Revelante, N., Gilmartin, M. (1987). Seasonal cycle of the ciliated protozoan andmicrometazoan biomass in a Gulf of Maine Estuary. Estuar. Coast. Shelf Sci. 25: 58 1-598Stoecker, D. K., Davis, H. L., Anderson, D. M. (1984). Fine scale spatial correlationsbetween planktonic diiates and dinoflagellates. J. Plank. Res. 6: 829-842Sutherland, T. A-., Leonard, C., Taylor, F. J. R. (1992). A segmented integrating pipesampler for profiling the upper water column. J. Plankton Res. 14: 915-923Throndsen, J. (1988). Cymbomonas Schiller (Prasinophyceae) reinvestigated by light andelectron microscopy. Arch. Protistenkd. 136: 327-336Verity, P. G., Vernet, M. (1992). Microzooplankton grazing, pigment, and composition ofcommunities during late spring in two Norwegian fjords. Sarsia 77: 263-274171Watanabe, L. N. (1976). Theoretical and practical aspects of nanophytoflagellate taxonomyand ecology on the British Columbia coast. Unpubl. Manuscr.172APPENDIX 4THE CODE FOR A BASIC MODEL TO INVESTIGATE THE ROLE OF CILIATES ASGRAZERS OF SMALL PHYTOPLANKTONModel DescriptionThere were five types of organisms in the system I have modeled (Fig. 9.1, Chapter9): phytoplankton (prey), ciliate A, B, C, (see Table 8.4, Chapter 8) and a copepod species(see Chapter 9). The prey were small (8 Am in diameter) and during the simulation had aconstant growth rate of A = 0.7 d4 (near one division day-1). Algal mortality was bypredation from the ciliates (see section 2.0, Chapter 8) and the copepod (see Chapter 9).Ciliate growth rates were functions of prey concentration (see Chapters 8 and 9). For all threeciliates, mortality was due to starvation (a function of prey concentration, Table 8.4) and topredation by the copepod (a function of ciliate concentration and size). There was no growthor mortality of the copepods over the simulation period. For a full explanation of the modelsee Chapter 9.The Model in BASIC codeDECLARE SUB model ()DIM SHARED x(20), d(20), Z1(20), z2(20), Z3(20), Z4(20), Z5(20), z6(20),Z7(20), p(15), zooDIM z(15, 900), zm(5)SCREEN 9COLOR 1, 63INPUT "enter time step, as fraction of a day (eg .05):", dtOPEN "c:\user\david\link3.dat " FOR OUTPUT AS #10'this sets the time step and numerical integration constantsIF dt = 0 THEN dt = .1z2 = dt / 2: z6 = dt / 6zoo = 0: icount = 0: p(12) = .6'number of variablesnd = 18:'this opens a file and importas growth and grazing parametersg$ = COMMANDSIF g$ = " THEN LOCATE 1, 1: INPUT "enter ciliate parameter file name (e.g.ciliate.dat):", g$OPEN g$ FOR INPUT AS 1FOR i = 1 TO 11: INPUT #1, p(i), id$: NEXT173CLOSE 1LOCATE 2, 1: INPUT "enter number of days to simulate for (e.g. 18 or 36):",ndaysstart:LOCATE 24, 1: PRINTOf .LOCATE 24, 1: INPUT "enter prey p, initial algae, ciliates and copepods (e.g.0.6,1000,1,1,1,0.001):", p(12), x(5), x(1), x(2), x(3), zoolabu = p(12): labp = x(5): labl = x(1): lab2 = x(2): lab3 = x(3)IF p(12) = 0 THEN END'this set initial state of (ciliates=1, algae=input value * this value) formaximum value output and gross phytoplankton outputzm(1) = 0: zm(2) = 0: zm(3) = 0: zm(5) = 0: x(4) = 0: x(6) = 0: x(7) = 0: x(8)= 0: x(9) = 0: x(10) = 0: x(11) = 0: x(12) = 0x(13) = 0: x(14) = 0: x(15) = 0d(1) = 0: d(2) = 0: d(3) = 0: d(4) = 0: d(5) = 0: d(6) = 0: d(7) = 0: d(8) =0: d(9) = 0: d(10) = 0: d(11) = 0: d(12) = 0d(13) = 0: d(14) = 0: d(15) = 0'this integrates the equations for "ndays" days using the 5th order Runge-Kutta methodFOR time = 0 TO ndays / dtt = time * dt: Z9 = t: modelFOR i = 1 TO nd: Z1(i) = d(i): Z7(i) = x(i): z2(i) = x(i) + Z1(i) * z2: x(i)= z2(i): NEXT: t = Z9 + z2: modelFOR i = 1 TO nd: Z3(i) = d(i): Z4(i) = Z7(i) + Z3(i) * z2: x(i) = Z4(i):NEXT: modelFOR i = 1 TO nd: Z5(i) = d(i): z6(i) = Z7(i) + Z5(i) * dt: x(i) = z6(i):NEXT: t = Z9 + dt: modelFOR i = 1 TO nd: x(i) = Z7(i) + (Z1(i) + 2 * Z3(i) + 2 * Z5(i) + d(i)) * z6:NEXT'this prevents prey numbers from becoming negative and should not be necessaryif the time step is small enoughIF x(1) < 0 THEN x(1) = 0IF x(2) < 0 THEN x(2) = 0IF x(3) < 0 THEN x(3) = 0IF x(5) < 0 THEN x(5) = 0'this places the data into a matrix z (1,2,3,5 time)z(1, time) = x(1): z(2, time) = x(2) : z(3, time) = x(3): z(4, time) = x(4):z(5, time) = x(5)z(6, time) = x(6): z(7, time) = x(7) : z(8, time) = x(8): z(9, time) = x(9):z(10, time) = x(10)z(11, time) = x(11): z(12, time) = x (12): z(13, time) = x(13): z(14, time) =x(14): z(15, time) = x(15)'this prints the daily output to the screenIF icount > 1 / dt THEN PRINT z(1, time); z(2, time); z(3, time); z(4, time);z(5, time)icount = icount + 1: IF icount > 1 / dt + 1 THEN icount = 0'this records the maximum number [zm(i)] of prey and ciliatesIF x(1) > zm(1) THEN zm(1) = x(1)IF x(2) > zm(2) THEN zm(2) = x(2)174> zm(3) THEN zm(3) = x(3)> zm(4) THEN zm(4) = x(4)> zm(5) THEN zm(5) = x(5)IF x(3)IF x(4)IF x(5)NEXT time'this is the section that plots the resultsCLS : LINE (0, 0)-(300, 200)„ BLOCATE 1, 40: PRINT "max. prey--"; INT(zm(5) * 100) / 100;LOCATE 2, 40: PRINT "prey production 00000: #"; INT(z(4, (ndays / dt)));LOCATE 3, 42: PRINT "plus initial prey, ng C"; INT(((z(4, (ndays / dt)) +labp) * .027778) * 100) /LOCATE 4, 40: PRINT "max.LOCATE 5, 40: PRINT "max.LOCATE 6, 40: PRINT "max.LOCATE 7, 40: PRINT "constant prey growth rate of (p) ="; labu;LOCATE 8, 40: PRINT "initial prey conc. #/ml="; labp;LOCATE 9, 40: PRINT "initial ciliate A conc. #/ml="; labl;LOCATE 10, 40: PRINT "initial ciliate B conc. #/ml="; lab2;LOCATE 11, 40: PRINT "initial ciliate C conc. #/ml="; lab3;LOCATE 12, 40: PRINT "constant copepod conc. #/ml="; zoo;LOCATE 13, 40: PRINT "prey eaten by copepods = #"; INT(z(6, (ndays / dt)));LOCATE 14, 62: PRINT "ng C"; INT((z(6, (ndays / dt)) * .027778) * 100) / 100;LOCATE 15, 40: PRINT "ciliate A eaten by copepods="; INT((z(7, (ndays / dt)))* 100) / 100;LOCATE 15, 1: PRINT "IIIIIIIIIIIIIIIIII 1^"LOCATE 16, 40: PRINT "ciliate B eaten by copepods="; INT((z(8, (ndays / dt)))* 100) / 100;LOCATE 16, 17: PRINT ndays; "days"LOCATE 17, 40: PRINT "ciliate C eaten by copepods="; INT((z(9, (ndays / dt)))* 100) / 100;LOCATE 18, 43: PRINT "total ciliates eaten ng C"; INT(((17.368 * z(7, (ndays /dt))) + (16.302 * z(8, (ndays / dt))) + (2.91 * z(9, (ndays / dt)))) * 100) /100LOCATE 18, 1: PRINT "ciliate production as ng C/ml"LOCATE 19, 1: PRINT "total ciliate A="; ((z(13, (ndays / dt))) + labl) *17.368;LOCATE 20, 1: PRINT "total ciliate B="; ((z(14, (ndays / dt))) + lab2) *16.302;LOCATE 21, 1: PRINT "total ciliate C="; ((z(15, (ndays / dt))) + lab3) * 2.91;LOCATE 22, 1: PRINT "TOTAL CILIATES PRODUCED="; (((z(13, (ndays / dt))) +labl) * 17.368) + ((z(14, (ndays / dt)) + lab2) * 16.302) + ((z(15, (ndays /dt)) + lab3) * 2.91)LOCATE 19, 40: PRINT "prey remaining= #"; INT(z(5, (ndays / dt))); ",ng C";INT((z(5, (ndays / dt)) * .027778) * 100) / 100LOCATE 20, 40: PRINT "prey eaten by ciliate A="; INT(z(10, (ndays / dt)));LOCATE 21, 40: PRINT "prey eaten by ciliate B="; INT(z(11, (ndays / dt)));LOCATE 22, 40: PRINT "prey eaten by ciliate C="; INT(z(12, (ndays / dt)));LOCATE 23, 45: PRINT "total prey eaten, ng C"; INT(((z(10, (ndays / dt)) +z(11, (ndays / dt)) + z(12, (ndays / dt))) * .027778) * 100) / 100'pptocp = INT(100 * ((z(6, (ndays / dt))) / (z(4, (ndays / dt)) + labp)))'LOCATE 18, 33: PRINT "1"; pptocp'pptoci = INT(100 * (((z(10, (ndays / dt)) + z(11, (ndays / dt)) + z(12,(ndays / dt)))) / (z(4, (ndays / dt)) + labp)))'LOCATE 19, 33: PRINT "2"; pptoci'citocp = INT(100 * (((17.368 * z(7, (ndays / dt))) + (16.302 * z(8, (ndays /dt))) + (2.91 * z(9, (ndays / dt)))) / ((((z(13, (ndays / dt))) + labl) *100;ciliate A +++++:"; INT(zm(1) * 100) / 100;ciliate B ^ .", INT(zm(2) * 100) / 100;ciliate C 00000:"; INT(zm(3) * 100) / 100;17517.368) + ((z(14, (ndays / dt)) + lab2) * 16.302) + ((z(15, (ndays / dt)) +lab3) * 2.91))))'LOCATE 20, 33: PRINT "3"; citocp'link = INT(100 * (((17.368 * z(7, (ndays / dt))) + (16.302 * z(8, (ndays /dt))) + (2.91 * z(9, (ndays / dt)))) / ((17.368 * z(7, (ndays / dt))) +(16.302 * z(8, (ndays / dt))) + (2.91 * z(9, (ndays / dt))) + (z(6, (ndays /dt)) * .027778))))'LOCATE 21, 33: PRINT "4"; linktcopeat = INT((17.368 * z(7, (ndays / dt))) + (16.302 * z(8, (ndays / dt))) +(2.91 * z(9, (ndays / dt))) + (z(6, (ndays / dt)) * .027778))ccopeat = INT(((17.368 * z(7, (ndays / dt))) + (16.302 * z(8, (ndays / dt))) +(2.91 * z(9, (ndays / dt)))) * 100) / 100pcopeat = INT((z(6, (ndays / dt)) * .027778) * 100) / 100cileat = INT(((z(10, (ndays / dt)) + z(11, (ndays / dt)) + z(12, (ndays /dt))) * .027778) * 100) / 100pp = INT(((z(4, (ndays / dt)) + labp) * .027778) * 100) / 100FOR i = 1 TO 5FOR time = 0 TO ndays / dtt = time * dtx = t / ndays * 300y = 200 * (1 - z(i, time) / (zm(i) + .00001))IF time = 0 THEN PSET (x, y) ELSE LINE -(x, y)IF i = 1 THEN CIRCLE (x, y), 1IF i = 2 THEN CIRCLE (x, y), 2IF i = 3 THEN CIRCLE (x, y), 3IF i = 4 THEN CIRCLE (x, y), 4NEXTNEXTPRINT #10, zoo, labp, ((z(13, (ndays / dt))) + labl) * 17.368, ((z(14, (ndays/ dt))) + lab2) * 16.302, ((z(15, (ndays / dt))) + lab3) * 2.91, (((z(13,(ndays / dt))) + labl) * 17.368) + ((z(14, (ndays / dt)) + lab2) * 16.302) +((z(15, (ndays / dt) _) + lab3) * 2.91)'PRINT #10, zoo, labp, pp, tcopeat, ccopeat, pcopeat, cileat'PRINT #10, zoo, labp, pptocp, pptoci, citocp, link, INT(((z(4, (ndays / dt))+ labp) * .027778) * 100) / 100'this section allows the data to be printed to a file'LOCATE 24, 3: INPUT "type the file name for output or -0' for nooutput"; fileout$'OPEN fileout$ FOR OUTPUT AS #2'PRINT #2, "time", "ciliate A", "ciliate B", "ciliate C", "prey"'FOR outi = 0 TO ndays / dt'PRINT #2, outi, z(1, outi), z(2, outi), z(3, outi), z(5, outi)'NEXT outi'CLOSE 02GOTO start'CLOSE #10SUB model'this section calculates the derivatives of ciliate [x(1), x(2) and x(3) etc.)'it then place these derivatives in d(1), d(2) and d(3)'Note that p(i)'s are from the ciliate parameter file (e.g. cili.dat)'these are the formulae for ciliate A'Ciliate A growth rate as function of prey x(5)xcl = x(5) - p(2)176ucill = (p(1) * xcl) / (p(3) + xcl)'Ciliate A grazing rate as function of its growth rategcill = x(1) * (((ucill * 625) + (176 / p(10))) / p(11))IF gcill < 0 THEN gcill = 0IF x(5) <= 0 THEN gcill = 0'these are the formulae for ciliate B'Ciliate B growth rate as function of algae x(5)xc2 = x(5) - p(5)ucil2 = (p(4) * xc2) / (p(6) + xc2)'Ciliate B grazing rate as function of its growth rategoil2 = x(2) *^* 587) + (170 / p(10))) / p(11))IF goil2 < 0 THEN goil2 = 0IF x(5) <= 0 THEN goil2 = 0'these are the formulae for ciliate C'Ciliate C growth rate as function of algae x(5)xc3 = x(5) - p(8)ucil3 = (p(7) * xc3) / (p(9) + xc3)'Ciliate C grazing rate as function of its growth rategoil3 = x(3) *^* 105) + (48 / p(10))) / p(11))IF goil3 < 0 THEN goil3 = 0IF x(5) <= 0 THEN goil3 = 0xcop = x(5) - 767: IF xcop < 0 THEN xcop = 0DEN = 1 + (( 1 / 785) * xcop) + ((1 / 45) * x(1)) + ((1 / 45) * x(2)) + ((1 /100) * x(3))'these are the formulae for the copepods grazing prey as a function of preyx(5) concentrationgcopp = ((3312000 / 785) * xcop) / DEN'this is the formula for the copepods grazing ciliate A as a function ofciliate A x(1) concentrationgcopcl = ((14400 / 45) * x(1)) / DEN'this is the formula for the copepods grazing ciliate B as a function ofciliate B x(2) concentrationgcopc2 = ((14400 / 45) * x(2)) / DEN'this is the formula for the copepods grazing ciliate C as a function ofciliate C x(3) concentrationgcopc3 = ((30000 / 100) * x(3)) / DEN'these are the derivatives combining growth and grazing ratesd(1) = (ucill * x(1)) - (d(7))d(2) = (ucil2 * x(2)) - (d(8))d(3) = (ucil3 * x(3)) - (d(9))d(4) = p(12) * x(5)d(5) = (p(12) * x(5)) - ((gcill) + (gci12) + (gci13) + (gcopp * zoo))IF x(5) >= 50000 THEN d(5) = 0 - (gcill) - (gci12) - (gci13) - (gcopp* zoo)IF x(5) >= 50000 THEN d(4) = 0d(6) = (gcopp * zoo)d(7) = (gcopcl * zoo)d(8) = (gcopc2 * zoo)d(9) = (gcopc3 * zoo)d(10) = gcilld(11) = gcil2d(12) = gcil3d(13) = (ucill * x(1)): IF d(13) < 0 THEN d(13) = 0d(14) = (ucil2 * x(2)): IF d(14) < 0 THEN d(14) = 0d(15) = (ucil3 * x(3)): IF d(15) < 0 THEN d(15) = 0END SUB177

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0053328/manifest

Comment

Related Items