Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The epiphyte Microcladia coulteri (Rhodophyta) : changes in population structure with spatial and temporal… Kendrick, Gary 1986

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

Item Metadata


831-UBC_1986_A6_7 K46.pdf [ 5.52MB ]
JSON: 831-1.0096697.json
JSON-LD: 831-1.0096697-ld.json
RDF/XML (Pretty): 831-1.0096697-rdf.xml
RDF/JSON: 831-1.0096697-rdf.json
Turtle: 831-1.0096697-turtle.txt
N-Triples: 831-1.0096697-rdf-ntriples.txt
Original Record: 831-1.0096697-source.json
Full Text

Full Text

c / T H E E P I P H Y T E M I C R O C L A D I A C O U L T E R I (RHODOPHYTA): C H A N G E S IN P O P U L A T I O N S T R U C T U R E W I T H S P A T I A L A N D T E M P O R A L V A R I A T I O N I N A V A I L A B I L I T Y OF HOST SPECIES. B Y G A R Y K E N D R I C K B .Sc , University of Western'Australia, Nedlands, 1980. A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S FOR T H E D E G R E E OF M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Botany) We accept this thesis as conforming to the required standard. T H E U N I V E R S I T Y OF BRITISH C O L U M B I A A P R I L 1986 © Gary Kendrick 1986 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an advanced degree a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I agree t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by t h e head o f my department o r by h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department o f The U n i v e r s i t y o f B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 E-6 (3/81') ii A B S T R A C T A comparison of the population structures of the epiphyte, Microcladia coulteri and the three hosts: Prionitis lanceolata, Iridaea cordata and Odonthalia floccosa, was made at Beaver Point, Saltspring Island, British Columbia. The three host species had distinct seasonal patterns in density and size class distribution. By the use of A N O V A , the partitioning of variation in epiphyte population structure with the seasons, between host species and, within host species variations in size of thalli, reproductive status and spatial distribution was performed. Small percentages of total determined variation were accounted for by seasonality in the abundance of size class and reproductive components of the epiphyte population and distribution of the epiphyte between host species. Larger percentages were due to variations in epiphyte population structure with within host species variations in size and reproductive status and the spatial variation in availability of host substrata. It was concluded that persistence of M. coulteri in relatively stable populations, both temporally and spatially, was due to differential use of available host substrata combined with continuous reproductive output and recruitment of the epiphyte. iii T A B L E OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES vi LIST OF FIGURES viii ACKNOWLEDGEMENTS ix CHAPTER 1. INTRODUCTION 1 CHAPTER 2. LITERATURE REVIEW: EPIPHYTES 2.1. Introduction to epiphytic communities 3 2.2. Factors affecting epiphyte community organization 4 2.3. Processes in epiphyte colonization 6 2.3.1. Dispersal and settlement 7 2.3.2. Attachment 10 2.3.3. Germination 11 2.3.4. Colonization 11 2.3.5. Development on the host 13 2.3.6. Reproduction 14 2.3.7. Summary 15 2.4. The genus Microc lad ia 16 2.4.1. The epiphyte M ic roc lad ia coulter i 17 2.4.2. Host specificity in M icroc lad ia coulter i 19 IV 2.4.3. Prionitis J . Agardh 1851 19 2.4.4. Iridaea Bory 1826 21 2.4.5. Odonthalia Lyngbye 1819 22 C H A P T E R 3. M E T H O D O L O G Y 3.1. Description of the study location 23 3.2. Methods of sampling 26 3.3. Data analysis 32 C H A P T E R 4. T E M P O R A L VARIATION IN P O P U L A T I O N S T R U C T U R E OF T H E H O S T SPECIES 4.1. Introduction 38 4.2. Results 40 4.2.1. Odonthalia floccosa 40 4.2.2. Iridaea cordata 44 4.2.3. Prionitis lanceolata 44 4.3. Discussion 49 C H A P T E R 5. T E M P O R A L A N D S P A T I A L VARIATION IN P O P U L A T I O N S T R U C T U R E OF M I C R O C L A D I A C O U L T E R I 5.1. Introduction 53 5.2. Results 54 5.2.1. Population structure of Microcladia coulteri 49 5.2.2. P C A and A N O V A 57 5.2.3. Within host variation in size and spatial distribution 69 5.3. Discussion 76 5.3.1. Population structure of Microc lad ia coulter i 76 5.3.2. Inferences from A N O V A results 76 5.3.3. Within host species variation in size 77 5.3.4. Spatial variation 78 C H A P T E R 6. PARTITIONING OF R E P R O D U C T I V E SIZE C L A S S E S OF M I C R O C L A D I A C O U L T E R I A M O N G T H E T H R E E HOST SPECIES 6.1. Introduction 81 6.2. Results 81 6.3. Discussion 95 C H A P T E R 7. G E N E R A L C O N C L U S I O N S A N D S U M M A R Y 98 L I T E R A T U R E CITED 100 A P P E N D I X 1 113 A P P E N D I X 2 116 vi L I S T O F T A B L E S 1 Distribution of Microcladia coulteri on various host species along the west coast of North America 20 2 Percent variation determined from sum of squares from nested ANOVA model for epiphyte population structure 68 3 Correlations (Pearsons product-moment) between total epiphyte density and host size 70 4 Mean, variance and k value (from negative binomial) for density of epiphytes on host species over the sampling months, 1984-1985 71 5 Y-intercept (a), slope fa), coefficient of determination (r^), X-intercept (xp and sample size (N) from total numbers of host thalli versus total epiphytes per thallus for the seasons. Regression equation is: y = a + blnx 75 6 Percent variation determined from sum of squares from the nested ANOVA model for the reproductive component of the epiphyte population 94 7 Correlations (Pearsons product-moment) between reproductive epiphyte density and host wet weight 96 8 Correlations (Pearsons product-moment) between reproductive epiphyte density, host species and time of year (months) 96 A l The size class, number and relative percent of Iridaea thalli containing epiphytic thalli 23 days after tagging epiphyte free thalli 115 V l l LIST OF FIGURES 1 Location of study area 24 2 Location of transect within study area 25 3A Topography of transect 28 3B Change in slope of rock substratum along transect 28 4 The size classes of Microcladia coulteri 31 5 Cumulative frequency distribution of Mahalanobis squared distance estimates for the epiphyte population growing on Prionitis lanceolata 34 6 Change in density of erect host thalli/m during the sampling period 41 7 Change in size class distribution of vegetative, carposporangial and tetrasporangial thalli of Odonthalia floccosa during the sampling period 42 8 Change in size class distribution of vegetative, carposporangial and tetrasporangial thalli of Iridaea cordata during the sampling period 45 9 Change in size class distribution of vegetative, carposporangial and tetrasporangial thalli of Prionitis lanceolata during the sampling period 47 10 Change in percent of host plants epiphytized over the sampling period for the three host species 55 11 Change in density of epiphytes per host thallus over the sampling period for epiphytized thalli of the host species 56 12 Change in density/m of Microcladia coulteri on the three host species during the sampling period 58 13 Change in density of the three size classes of Microcladia  coulteri per host thallus from epiphytized thalli of the three host species. A - Iridaea cordata 60 B - Odonthalia floccosa 60 C - Prionitis lanceolata 61 viii 14 Eigenvector plot for variables for vegetative and reproductive epiphyte thalli from all host species 62 15 Plot of monthly means of the PCA scores of the epiphyte variables for each of the host species. A - Prionitis lanceolata 65 B - Odonthalia floccosa 66 C - Iridaea cordata 67 16 Distribution of epiphytes on host thalli as a function of the number of host thalli with increasing numbers of epiphytes per host thallus. P - Prionitis lanceolata 0 - Odonthalia floccosa 1 - Iridaea cordata 72 17 Linear regression of the number of host thalli with increasing numbers of epiphytes per host thallus natural log transformed. Regression equation is y = a + blnx 73 18 Seasonal change in mean number of carposporangial, spermatangialand tetrasporangial epiphytes per thallus from Prionitis lanceolata 83 19 Seasonal change in mean number of carposporangial, spermatangial and tetrasporangial epiphytes per thallus from Iridaea cordata 85 20 Seasonal change in mean number of carposporangial, spermatangial and tetrasporangial epiphytes per thallus from Odonthalia floccosa 87 21 Eigenvector plot of the variables for reproductive epiphytic thalli for all host species 88 22 Plot of monthly means of PCA scores of the reproductive epiphyte variables for each of the host species. A - Prionitis lanceolata 91 B - Odonthalia floccosa 92 C - Iridaea cordata 93 A2.1 Total combinations of random placement of 8 quadrats without replacement over 15 months of sampling 1. - x/y plot 2. - x/log y plot 119 ix ACKNOWLEDGEMENTS I am grateful to Dr. M . W. Hawkes for his encouragement guidance and support. His excitement towards Phycology has been inspirational. This research was supported by N S E R C Operating Grant U - 0 3 1 8 to Dr. Hawkes. Dr R.F. Scagel is gratefully acknowledged for his support of my application to graduate school, financial aid and -use of laboratory facilities. Thanks to Dr. Jack Maze, Rob Scagel, Dr. Gary Bradfield, John Spence and Kali Robson for support and discussion on data analysis. Special thanks to Drs. P. G. Harrison and R. De Wreede for directing my excitement in benthic marine ecology and for excellent technical advice during the early development of my study program. Thanks to all the members of B .G.S.A. whose banter, discussion and arguements made for an enriched academic environment. To my wife, Amrit must go the ultimate acknowledgements. She braved the ice and snow, wind, sleet and rain, hypothermia and the denizens of the deep off Beaver Point to assist in a year round sampling experience. Thank you Amrit for encouragement and comradeship throughout. To my cousins, Rick Blacklaws, Carol Trauman and Brydon also goes heartfelt thanks and without whose use of house, cars, bathtub and potbellied stove this thesis would not have been possible. 1 CHAPTER 1. INTRODUCTION Epiphytes constitute a significant component of nearshore benthic algal communities (Kain 1971; Ballantine and Humm 1975; Harlin 1980; Whittick 1983). Studies of epiphyte community structure (Ballantine and Humm 1975; Ballantine 1979; D'Antonio 1985), effect of epiphytes on benthic macrophytes (Sand-Jensen 1977; Harkin 1981; Bulthuis and Woekerling 1983 ), and autecology of epiphyte-host interactions (Harlin 1970; Harlin and Craigie 1975; Lewis 1982; Apt 1984) have indicated a wide spectrum of levels of interactions between host and epiphyte. Very few studies have investigated temporal variation in population structure of epiphyte species (Kain 1982; Whittick 1983) nor questioned how the population structure of the epiphyte is affected by temporal and spatial variation in availability of host substrata. The structuring of epiphyte populations is important in determining persistence of epiphyte species in a temporally and spatially patchy environment. Studies that investigate interactions between host and epiphyte at the individual level (nutrient transfer- Evans et al. 1973; Harlin and Craigie 1975; Turner and Evans 1977; Goff 1979: pit connections- Wetherbee 1979; Wetherbee and Scott 1980; Goff 1982: host specificity-Rawlence 1972; Nonomura 1979; Goff 1982; Apt 1984) contribute little towards understanding the persistence of populations of diverse algal epiphytes. Similarly, studies of epiphyte communities (Ballantine and Humm 1975; Ducker et al. 1977; May et al. 1978; Ballantine 1979) indicate the diversity and relative abundance of epiphyte species, but are usually insensitive to change in size structure and reproductive output for single species of epiphytes. Studies of population structure of benthic (Hansen and Doyle 1976; Norall et al. 1981; Kain 1982; De Wreede 1984; Hannach and Santelices 1985) and epiphytic algal species (Kain 1982; Whittick 1983) have resulted in information on size class allocation of 2 abundance and reproduction temporally, and sometimes spatially (Norall et _al. 1981). No studies of epiphyte populations have attempted to correlate structure of the epiphyte population with host population structure. Such an investigation would allow the partitioning of variation of epiphyte population structure amongst different host species and, within host variations in shape, size, surface texture, and reproduction. Partitioning of variation in this way would allow for the generation of hypotheses that test important interactions between specific epiphytes and their hosts. With the above in mind, it was my intention to study the spatial and temporal changes in population structure of the epiphyte, Microc ladia coulteri Harvey on the hosts: Pr ioni t i s lanceolata (Harvey) Harvey, Iridaea cordata (Turner) Bory and Odonthalia floccosa (Esper) Falkenberg. The goals of the study were: 1. To describe the spatial and temporal variation in population structure of the host species. 2. To describe the spatial and temporal variation in population structure of the epiphyte. 3. To analyze the observed variation in the epiphyte population structure in relation to spatial and temporal changes in host substrata. From this investigation hypotheses based on important interactions between the epiphyte and the hosts were generated. 3 C H A P T E R 2. L I T E R A T U R E R E V I E W : E P I P H Y T E S . 2.1. Introduction to epiphytic communities A rich flora of ephemeral algae are found to colonize benthic marine algae and seagrasses. Some species are found to be highly host specific (Phaeocolax kajimura Hollenberg, host specific to Lobophora variegata (Lamouroux) Womersley (Apt 1984), Polysiphonia lanosa (Linnaeus) Tandy found only on Ascophyllum nodosum (Linnaeus) Le Jolis (Rawlence 1972; Harlin and Craigie 1975; Turner and Evans 1977, 1978; Lobban and Baxter 1983) Ceramium codicola J . G . Agardh, on Codium (Lewis 1982). Most epiphytes are not substratum specific. Humm (1964), Ballantine and Humm (1975), Ducker et al. (1977), and May et al. (1978) all reported the presence of algae common to other plant hosts and non-living substrates as epiphytes on seagrasses. Markham (1969), Ballantine (1979) and Whittick (1983) also indicated the same for epiphytes growing on macrophytic brown and red algae. Harlin (1975) wrote that the occurrence of seagrasses in soft bottom communities increases the availability of surfaces on which algae could grow. D'Antonio (1985) noted that organisms which otherwise might grow on rocky sustrates are epiphytic in rocky intertidal environments where space is often a limiting resource. Seasonal variations in abundance of epiphytic algae have been well documented in both temperate and tropical waters. Humm (1964), Van der Ben (1969), Ducker et al. (1979) and Heijs (1985) emphasized considerable seasonal variation in epiphyte species composition and abundance found on seagrasses. Two components of the epiphyte flora were recognized on the basis of seasonal behaviour; the ephemeral and the perennial component. Humm (1964) and Ducker et al. (1979) suggested that the ephemeral component was adapted to leaf colonization, leaves of seagrasses being 'turned over' every 30 to 60 days. Kain (1982) and Whittick (1983) found that the seasonality of epiphyte abundance on species of Laminaria had no year round component, possibly because of the ephemeral nature of the algal host thallus. 4 Epiphytes constitute a large proportion of total biomass in seagrass (Silberstein 1980; Bulthuis and Woelkerling 1983) and algal communities (Harkin 1981). They have negative effects on seagrass productivity (Sand-Jensen 1977; Silberstein 1980), but constitute a major supporter of foodwebs within seagrass beds (Kitting et jd. 1984) and rocky intertidal shores (D'Antonio 1985). In summary, epiphytic communities are widespread both on seagrasses and benthic algae. They are an important contributor to foodwebs, although they have been found to reduce growth and photosynthesis in the host. Epiphyte species vary from being highly host specific to occurring on both biological and non-living substrata. 2.2. Factors affecting epiphyte community organization. Epiphytic communities found on marine algae and seagrasses are the product of: 1. Interactions of the host with the macro-environment. This includes the flow environment around the host (Cattaneo 1976; Vogel 1981; Nowell and Jumars 1984), release of antibiotics or volatile chemicals from the host (Fenical 1975; Hornsey and Hide 1976; Al-Ogily and Knight-Jones 1977; Glombitza et_al. 1982; Harrison 1982; Hay 1984) and release of organics from the host that may attract epiphyte colonization (Sieburth 1969; Penhale and Smith 1977). 2. Growth characteristics of the host. Russell (1983a, b) found that colonization of epiphytes on Laminaria spp. was proximal to the intercalary meristem of the host otherwise they would be sloughed off with senescent host blade tissue before reaching maturity. Seagrasses also have a high turnover of leaf blade tissue with blades being replaced every 20 to 40 days in Amphibolis sp. (Walker pers. comm.) and Posidonia sp. (Silberstein 1980). 3. Interaction between epiphytic organisms. Well-defined competitive interactions in benthic communities may not occur when the competitive species are found epiphytically (Seed and O'Connor 1981; D'Antonio 1985). Seed and O'Connor (1981) noted 5 "the importance of resource generation (host growth) as an alternative to the disturbance and competitive network paradigms" in the structuring of epiphytic faunal communities. Substratum generation can be viewed as a periodic disturbance, the epiphytic community structure being analagous to algae found in seasonally devastated cobble substrata (Lieberman et al. 1979) or sand-influenced rocky intertidal areas (Taylor and Littler 1982). The 'periodic disturbance' view of structuring in epiphytic communities characterizes disturbances as: 1. Constant loss of older substratum at the senescent region of the host. 2. Constant occurrence of new substratum for colonization (post-disturbance patch) the amount being determined by the growth rate of the host. 3. Limited time for growth and reproduction of epiphytes on the available substratum. The above processes are compounded by: 1. Sloughing off of cuticle by marine algal hosts (e.g. Halidrys siliquosa (Linnaeus) Lyngbye (Moss 1982), Ascophyllum nodosum (Linnaeus) Le Jolis (Frillion-Myklebust and Norton 1981). Some epiphytes have means of surviving cuticle sloughing; e.g. Elachista scutulata (J.E. Smith) Areschoug and Herponema velutinum (R.K. Greville) J . G . Agardh exploit cuticle breaks and cryptostomata in the thallus surface of Himanthalia elongata (Linnaeus) S.F. Gray (Russell and Veltkamp 1984). 2. Exudation of antibiotics and allelopathic compounds by many groups of marine algae. Antibiotic compounds (haloketones, brominated phenols and mono- sesqui- and diterpentenes) were described from 5 orders of the Rhodophyta by Fenical (1975) and reference to their effects on other organisms have produced massive amounts of literature (Allen 1967; Fenical 1975; Hornsey and Hyde 1976; Al-Ogily and Knight-Jones 1977; Glombitza et al. 1982). As noted by Hay (1984), unique metabolites were characteristic of almost all of the herbivore resistant algae in his study, but some of the herbivore susceptible species also contain compounds proposed as herbivore deterents. Steinberg (1985) studied the effect of phenolic compounds from brown algae on the feeding behavior of Tegula funebralis and noted that the gastropod did not feed on algae containing large amounts of phenolic compounds. From this 6 study he inferred that phenolic compounds in brown algae are effective feeding deterents against many invertebrate herbivores. He tested only one invertebrate. There is a need to stop making such global inferences about the activity of exuded organic compounds on the reduction of both epiphyte growth and herbivore grazing. 3. Grazing pressure by herbivores. Epiphytes were found to provide food for littorine snails and gammarid amphipods on the rocky intertidal alga Neorhodomela l a r i x (Turner) Masuda (as Rhodomela l a r i x in D'Antonio 1985), and to be the primary basis for the foodweb in seagrass meadows (Kitting e_t j i l . 1984) where herbivores included the snails Hydrobia and Modulus , foraminifera, marine nematodes, amphipods and polychaete worms (Orth and Van Montfrans 1984). Grazing pressure was found to vary both spatially and temporally within seagrass beds and rocky intertidal environments. In summary, the development of epiphytic communities on benthic macroalgae and seagrass are affected by: the interactions of the host with the environment, the growth characteristics of the host, and the interactions between epiphytic organisms. Epiphytic communities are structured by 'periodic disturbances' consisting of sloughing off of older host tissue and growth of new tissue with epiphyte development occurring between these events. This pattern is compounded by sloughing of the cuticle by marine benthic algae, exudation of allelopathic chemicals, and herbivore grazing pressure on the epiphytic community. 2.3. Processes i n epiphyte colonization For an epiphyte species to persist in the dynamic community discussed in section 2.1 it is necessary for the species to successfully establish itself on the host species to counteract host defences and competitive interactions with other epiphytes, to persist to a reproductive size against cuticle sloughing, allelochemicals and herbivory pressure, and to optimize dispersal of spores and gametes. Stages in this process can be broken down into: establishment, consisting of settlement, attachment, germination and colonization, and 7 development and reproduction. As this thesis is on a rhodophyton epiphyte much of this discussion will pertain to the Rhodophyta in general. 2.3.1. Dispersal and settlement The successful establishment of marine algae depends on dispersal, settlement, attachment and germination of spores and gametes. These spores and gametes represent the planktonic phase of algae and the interaction of the algal spore with the ambient environment determines the success of dispersal (Coon et al. 1972). In motile zoospores and gametes of the Phaeophyta and Chlorophyta, recognition of compatable surfaces has been shown to occur through recognition of released volatile, low molecular weight hydrocarbons (e.g. chemical lures produced by female gametes of Cutleria and Desmarestia - Boland et al. 1984, sexual pheromone specificity in Laminariales - Muller et al. 1985). In the non-motile spores of Rhodophyta settlement onto available substrata appears to be random. The importance of surface recognition of algal chemoreceptors or other hydrocarbons by rhodophyton spores seems highly unlikely. Spore settlement appears to be mainly affected by: 1. Rate of spore settlement. Density and size of spores plays an important role in the settlement rates of algal spores. Okuda and Neushul (1981) found that the rate of sinking of red algal spores was directly proportional to spore size. They attributed the successful colonization of particular species in recruitment studies by Neushul etjd. (1976) as being due to both spore size and the microhabitats into which spores settle. Any adaptive advantage to large spore size is extremely doubtful because of the considerable variation in spore size found in populations of temperate (Coon_et al. 1972) and tropical red algal spores (Ngan and Price 1979). Surface mucilage has been shown to double the diameter of red algal spores (Ngan and Price 1979). Released red algal spores are commonly surrounded by a mucilage layer (McBride 1972; Chamberlain and Evans 1973: Fletcher 1979) or are sometimes collectively released in a mass of mucilage (Boney 1975). Experimental evidence, in moving water, suggests that upon release spores may aggregate or be transferred to adjacent thalli by mucilage strands 8 [e.g. Ti f fanie l la snyderiae (Farlow) Abbott (Fetter and Neushul 1981)] which would reduce their randomness of settlement. 2. Flow regime around the substratum. As spores settle from the water column they are being carried by uni-directional currents, bi-directional oscillatory wave motion and upon nearing substratum, an area of decreased water motion known as the boundary layer. A l l of these affect the rate of spore settlement (Neushul 1972). The theoretical mechanical forces that they undergo are discussed by Nowell and Jumars (1984) and Vogel (1981). Actual measurements of velocities of breaking ocean waves have been made by Jones and Demetropoulis (1968) and found to be 14 meters sec 1 . Mechanical adaptations of macroscopic algae to breaking oceanic waves have been investigated for E isenia arborea Areschoug (Charters et al. 1969), Nereocystis luetkeana (Mertens) Postels et Ruprecht (Koel and Wainwright 1977) and Fucus sp. (Norton et al. 1982). Tensile strengths were found to be an order of magnitude less in seaweeds than other biological materials, but elasticity was very high. Few studies have directly investigated the effect of water motion on the distribution of settling spores. Munteanu and Maly (1981) and Cattaneo (1978) both discussed the micro-distribution of diatoms in relation to water flow. Charters et_al. (1972) devised experiments to test the effect of water motion on algal spore adhesion. 3. Surface characteristics of substratum. As indicated in studies using artificial substrata of varying textures (Linskens 1966; Niehuis 1969; Risk 1973; Harlin and Lindberg 1977) and from field observations (Den Hartog 1972; Harlin 1974) the surface texture of the substratum affects the settlement and community structure of marine algae. In studies on the settlement of spermatia of Aglaeothamnion neglectum Feldmann-Mazoyer, Magruder (1984) found that external ornamentation such as hyaline hairs may aid in the entrapment of spermatia. Vogel (1981) discusses the effect of surface ornamentation (e.g. bumps, hairs, spines) in reducing the boundary layer and in formation of within boundary layer turbulent flow. Studies on distributions of settling diatoms (Munteanu and Maly 1981; Medlin 1983) and macroalgae (Harlin and Lindberg 1977) indicate that the areas of densest settlement and population 9 growth are associated with areas of turbulent flow. Harlin and Craigie (1975) investigated transfer of nutrients from the host Aseophyl lum nodosum to the epiphyte Polys iphonia  lanosa and concluded that the surface characteristics of the host were critical in the settlement of the epiphyte spores. They went on to infer that the surface characteristics of the host could explain much of the apparent 'obligate' nature of Polys iphonia lanosa. The shape of the substratum is also important in determining the distribution of organisms on it. Ballantine (1979) found that multiple branched host algae were more suitable for epiphyte colonization than bladed algae. Epiphytic diatoms on Potomogeton (Cattaneo 1978) and on marine algae (Medlin 1983) were found to have high densities on branch apices of the hosts. Studies using artificial seagrass made from strips of plastic (Harlin 1970; Silberstein 1980; Bell et al. 1985) determined that the species composition of the epiphyte flora on the plastic strips was similar to that found growing on natural seagrass substrata. Silberstein (1980) attributed this similarity to the hydrodynamic environment being similar around these substrata. Surface chemistry of the substratum is a major factor in the settlement of invertebrate larvae onto surfaces (Crisp 1965). Chemical stimuli for settlement of algal spores onto specific substrata have been investigated for the 'parasitic red algae' (Gonzalez and Goff 1980; Nonomura and West 1981). Excretion of organic compounds occurs in all groups of micro- and macro-algae, in healthy cells (Tolbert and Zill 1956; Wetzel 1969; Kroes 1970) and in senescent cells (Tukey 1970). In seagrasses, excretion of organic compounds has attracted colonization by epiphytes (Penhale and Smith 1977). Recent investigations of marine bacteria have indicated that they are capable of assimilating amino acids excreted by algae and recycling nitrogenous compounds (Amano et_al. 1982). Brown algae are known to take-up and metabolize exogenously supplied glucose (Drew 1969). Niehof and Loeb (1974) illustrated that dissolved organic material in seawater affects the electric charge of immersed surfaces. The strength of attachment and growth of rhizoids of Enteromorpha intestinalis (Linnaeus) Link, was found to be affected by surface charge (Fletcher and Baier 1984). How the surface 10 chemistry and associated charge affects the settlement and attachment of algal spores has not been investigated. 2.3.2. Attachment Specific recognition reactions during attachment, possibly, initiated by compounds on the surface of the host thallus, have not been shown for red algal spores. Specialized appendages on spermatia of Aglaeothamnion negleetum were shown to specifically bind to trichogynes (Magruder 1984) although the mechanism of binding was not shown. Specific recognition by contact with compounds on the surface of the host have been shown to occur in diverse symbioses: Symbionts of Didemnum and species of of Hydra (Pardy and Muscatine 1973, Chlorella cytobionts of Paramecium bursaria (Weis 1980), and fungal pathogens on angiosperms (Allen 1976). Release of intracellular adhesion vesicles may also involve stimulation from a substance on the surface of the host thallus. It has been suggested that the mucilage surrounding red algal spores assists in initial adhesion to surfaces (Evans and Christie 1970; McBride and Cole 1971; Kugrens and West 1972; Scott and Dixon 1973; Chamberlain and Evans 1973; Kugrens 1974; Fletcher 1979; Cole and Sheath 1980; Tveter-Gallagher et al. 1980). Release of an adhesive compound, from vesicles within the spore, upon attachment, may further aid in adhesion (Chamberlain and Evans 1973). Preliminary results from adhesion studies of the parasitic red alga Asterocolax gardneri (Setchell) Feldmann et Feldmann suggest that the process is random since the red algal spore attaches equally well to the host and non-hosts in the Delesseriaceae, as well as to brown and green algae (Goff 1982). There is a need to study adhesion of the spores of the red algae with an emphasis on chemical and physical stimuli for attachment. 11 2.3.3. Germination Host specific epiphytic algae and parasitic red algae show a range of germination responses to the presence of a suitable host for spore germination. The highly pigmented parasitic red alga, Janczewskia morimotoi Setchell et Guernsey (Nonomura 1979) does not seem to need the presence of a compatible host for spore germination. The spores of the less pigmented, Harveyella mirabilis (Reinsch) Schmitz et Reinke (Goff 1975) are not able to germinate in the absence of their specific host. In some red algal species (e.g. Plocamiocolax sp. (Goff 1982), the frequency of spore germination increases in the presence of compatible hosts. The interaction of stimulants, self-inhibitors and external restraints determine the success of germination of fungal pathogens (Allen 1976). From the indirect evidence presented above, stimulatory compounds on host surfaces seem to be associated with successful germination in the parasitic red algae. The cuticle of red algae is rich in proteins (Hanic and Craigie 1969) as well as halogenated phenolics and mono-, sesqui- and diterpenes (Fenical 1975) . Phenolics and terpenes were found to stimulate spore germination in fungi (Allen 1976) , therefore it is highly likely that the speciation of these compounds result in host specificity in at least the parasitic red algae. Pigmented parasitic red algae and specific epiphytes germinate more freely in the absence of host material possibly because of synthesis of self-stimulatory germination compounds through photosynthesis. 2.3.4. Colonization The type of colonization of algal epiphytes can be divided into two types: 1. Superficial colonization. This occurs with many generalist epiphytes on seagrasses (Ducker et al. 1979). 2. Penetration colonization. This can be defined as the penetration of host cuticle and development of rhizoids within the cortex and medulla of the host. Host penetration can occur 12 by penetration through damaged tissues (e.g. Harveyel la mirabi l is - Goff and Cole 1976a; Choreocolax polysiphoniaea Reinsch - Kugrens and West 1973a as Leachie l la pacifica Kugrens), and penetration of healthy tissues (Polysiphonia lanosa - Rawlence and Taylor 1970; Ectocarpus sp. - Russell 1983a, b). The mechanisms proposed for primary penetration of host tissues by parasitic red algae are a combination of mechanical and enzymatic processes (Goff 1982). Mechanical penetration is inferred from vacuole enlargement in germinating spores (e.g. Janczewskia morimotoi - Nonomura and West 1980). In Garder ie l la tubifera Kyl in and Asterocolax gardner i (Goff 1982) there is loss of staining ability of the host cell wall around the algal parasite, and a distinct halo surrounding the advancing tip of the parasitic rhizoid suggesting enzymatic degradation of the host cell wall. No attempt has been made to isolate cell wall degrading enzymes from germinating parasitic red algal spores. Rawlence and Taylor (1970) and Rawlence (1972) suggested that the rhizoids of Polys iphonia  lanosa were capable of penetrating healthy tissues of the host A s c o p h y l l u m nodosum. Lewis (1982) described the apices of rhizoids of Microc lad ia coulteri , and from their thickened cell walls and lance-like shape inferred that they were capable of penetrating healthy host tissues. Although the indirect evidence given above suggests the probability that both mechanical and enzymatic processes are important in the penetration of the host cell wall by both specific epiphytes and parasitic red algae, definitive studies isolating and characterizing enzymes excreted during rhizoidal penetration, studying mobilization and growth of vacuoles in germinating spores, and determining the energy needed to mechanically penetrate the host, need to be undertaken. Research on fungal pathogens of Angiosperms has also indicated that the mechanism of primary penetration is both partially mechanical and enzymatic (Allen 1976). Some of the wall degrading enzymes isolated from fungi include cutinases, pectinases and cellulases (Cooper 1983; Kolattukudy and Koller 1983). Angiosperms have a cuticle composed of insoluble polymers embedded in hydrophobic materials as well as a cell wall containing polysaccharides and lesser amounts of glycoproteins. The algal cell wall and cuticle 13 is a laminated structure containing rich proteinaceous bands embedded in polysaccharides with no insoluble polymers embedded in hydrophobic materials surrounding it (Hanic and Craigie 1969). Cell wall porosity in Angiosperms ranges from 35 to 70 Angstroms. This is probably greater in the algae due to the need to exchange ions and nutrients with the ocean through the surface of the algal thallus. Thus, the types of wall degrading enzymes in epiphytes and parasitic red algae may be different and in lower concentrations than those produced by fungal pathogens on Angiosperms. Variations in the structure and composition of cell wall polymers in algae may be another factor adding to the host specificity of epiphytes. 2.3.5. Development on the host. Once penetration of the host thallus has occurred the epiphyte has: 1. Access to develop intercellular connections with the host. In parasitic red algae secondary pit connections form between parasite and host cells; the ultrastructure of development and effects to the host are discussed by Wetherbee (1979), Wetherbee and Scott (1980), Goff (1982), and Wetherbee et al. (1984). 2. To effectively counter host defences. Goff (1976a) and Nonomura (1979) reported a thickening of host cell walls in response to infection by parasitic red algae. No studies have been made to determine if any inhibitory substances are synthesized by hosts flsoth algal and seagrass) with the onset of infection of a "penetrating type" epiphyte or parasite. Inhibitory substances are produced by algae and seagrasses (Lucas 1947; Fenical 1975; Allen 1976; Hornsey and Hyde 1976; Al-Ogily and Knight-Jones 1977; Ragan and Jensen 1979; Glombitza et jil. 1982; Hartmann and Lohr 1983). Whether the synthesis of these compounds can be mediated in response to infection, as in Angiosperms infected by fungal pathogens (Allen 1976), in not known. 3. Access to organic compounds synthesized by the host. Many studies on translocation of organic compounds between host and epiphyte have been attempted. Linskens (1963, 1966, 1968) separated the spatial and functional interactions between host and epiphyte, and 14 showed that contact is not necessary for exchange of organic compounds between the two organisms. Nutrient transfer has been reported between epiphytes and seagrasses (Harlin 1973; McRoy and Goering 1974) and marine macroph3'tic algae (Linskens 1963; Floc'h and Penot 1970, 1971, 1972; Penot 1974; Harlin and Craigie 1975; Turner and Evans 1978) and parasitic red algae and algal hosts (Evans et al. 1973; Goff 1979). A l l of these studies describe an apparent biochemical exchange. The i n si tu excretion of organic compounds both in healthy and senescent tissues in marine algae and seagrasses (Tolbert and Zill 1965; Muscatine 1967; Tukey 1970; Penhale and Smith 1977, Ragan and Jensen 1979) casts doubt on the importance of direct translocation of photo-assimilates from host to epiphyte. Results from translocation experiments demonstrate transfer of organic compounds between host and epiphytes, but offer no adequate explanations of the interactions between hosts and epiphytes, of why host-epiphyte relationships exist, nor how epiphytes become established on specific hosts. 2.3.6. Reproduction Unlike cytobionts where transmission occurs with the host, the epiphytic algae need mechanisms that will increase the probability of contacting suitable hosts during dispersal. These include: 1. Producing many spores and gametes. 2. Seasonal occurrence of various reproductive phases of the epiphyte (Goff and Cole 1976b). 3. Utilizing a combination of spore size and density combined with timing release to determine spore settling rates (Okuda and Neushul 1981). 4. Forming spore aggregates attached to mucilage strands (Fetter and Neushul 1981). 5. Simultaneous production and release of spores and gametes that represent different phases in the life history of the epiphyte (Goff 1982, West and Hommersand 1981). 15 6. Compartmentalization of two genotypes into the one spore (binucleated spore; Goff 1982). 7. Asexual reproduction (Goff 1982, West and Hommersand 1981). Studies on the ectocarpoid epiflora on Laminaria digitata (Linnaeus) Lamouroux (Russell 1983a,b) indicate that successful completion of the life history of these epiphytes occurs because they attain reproductive maturity prior to blade shedding in the host. The generalist epiphytes Palmaria, Phycodrys and Membranoptera made use of the substratum Laminaria hyperborea (Gunnerus) Foslie because their peak sporing season spanned the period when stipes of the kelp were growing (Kain 1982). These same species were found to be early colonizers of areas where the Laminaria canopy was removed (Hawkins and Harkin 1985). The success of reproductive output in the perpetuation of generalist epiphyte populations seems to be a mixture of reproductive strategy and the fortuitous availability of substratum. 2.3.7. Summary The process of colonization of a host by an epiphyte consists of 3 phases: 1. Dispersal and settlement; 2. Attachment, germination and colonization, and; 3. Development and reproduction. There is an immense amount of missing information on the process of successful establishment of an epiphytic alga on host algae and seagrasses. Some of the major areas that need further investigation are: 1. Dispersal distance of epiphyte spores. 2. Interaction of epiphyte spores with the hydrodynamic environment around the host thallus. 3. Surface chemistry and associated charge of surfaces and their effect on settlement and attachment of epiphyte spores. 16 4. Determination if germination stimulatory compounds are found on the surface of host thalli. 5. Isolation and characterization of enzymes involved in the penetration of host thalli. 6. Studies on mobilization and growth of vacuoles in germinating spores combined with the determination of the energy needed for physical penetration of the host cuticle by an epiphyte. 7. Description of the reproductive phenology of epiphytes. How important these areas of study are in terms of variation in population structure of individual epiphyte species is unknown. 2.4. The genus Microcladia The genus Microcladia (Ceramiaceae) was established by Greville (1830) based on a species collected from Britain, Microcladia glandulosa. This taxon was based on Fucus  glandulosus Turner (1808). The genus stands very close to Ceramium, differing only in that the cortical filaments are ternately and quarternately branched in a forward direction to form a continuous cortex of greater than two cells thick (Hommersand 1963). At present, eleven species, many of them epiphytes, are placed in the genus Microcladia. These can be divided ' into two groups: 1. Bilaterally compressed species: Microcladia glandulosa from England, M . moselyi and M . alternata from the south Pacific, M . dentata and M . elegans from Japan, and M . exserta from South Africa (Wynne 1985). 2. Radially symmetrical species: M . pinnata and M . novae-zelandia from New Zealand, M . capensis from South Africa, M . coulteri, M . borealis and M . californica from the Pacific coast of North America. 17 All but two species, M . borealis and M . novae-zelandia, are found predominantly as epiphytes. The benthic species also commonly occur as epiphytes (Lewis 1982). 2.4.1. The epiphyte Microcladia coulteri Harvey 1853 Microcladia coulteri was established by Harvey (1853) from collections made by Dr. Coulter, in the Monterey Bay area, California. The species has been reassessed by Hommersand (1963) and Abbott and Hollenberg (1976). Abbott and Hollenberg (1976) found similarities in carposporophyte development with the type species, M . glandulosa, but also found shared characteristics (e.g. cortical rhizoids) with the Japanese genus, Campylaephora J . G . Agardh (1851). Microcladia coulteri has been reported on a variety of red, brown and green algae along the shores of the west coast of North America. It is predominantly an epiphyte rarely found on hard substratum. It has the triphasic diplohaplontic life history common to many red algae. Coon et al. (1972) measured tetraspores of M . coulteri and found plant to plant variation in spore size to be great perhaps associated with rapid germination. Charters etjal. (1972) found tetraspores to be sticky, becoming attached to surfaces almost immediately upon settling. Possible stimulation of germination may occur upon release from the sporangium of M . coulteri. Whether stimulatory compounds for germination from the host play an important role in determining the epiphytic nature of the alga remains to be shown. Penetration colonization occurs with ramifying rhizoids developing from the cortical cells. Transfer of 1 4 C between the host Grateloupia and M . coulteri was recorded by Harlin (1970). Although penetration colonization and nutrient transfer occurs between the epiphyte and hosts, the relationships to growth and reproduction of M . coulteri, and specific hosts have not been documented. Microcladia coulteri is an epiphyte on other algae. The holdfast is an irregularly branched rhizoid that develops into an anastomosing network within the host tissue. Lower 18 cortical cells also develop into rhizoids. The mature thallus ranges in size from 30 to 400 mm and is pyramidal in outline. There is an alternate branching pattern, with the branch tips incurved and forcipate, similar to Ceramium. In addition to this branching pattern, adventitious branches are frequently produced from the main axis, especially in older portions of the thallus. Because of the adventitious branching there is considerable variation in the habit of the older thalli. Male, female and tetrasporangial thalli look similar and show the same variability in shape. Male thalli are generally smaller in stature. Female and carposporangial thalli show greater amounts of adventitious branching. The gametophytes are dioecious with male and female structures occuring on separate thalli. Spermatangia in the male gametophyte cover the last three or four orders of branching, 3 or 4 spermatangial mother cells, giving rise to 2 or 3 spermatangia, develop from outer cortical cells. Each spermatangium produces one spermatium. Therefore 6 to 12 spermatia are developed from each outer cortical cell. Procarps are formed abaxially in branch apices in the last 3 to 4 orders of branching, in female gametophytes. A fertilized carpogonium may have one to several spermatia attached to the trichogyne and in cytoplasmic connection with it. Polyspermy at the nuclear level has never been traced. The carposporophyte that develops from the fertilized carpogonium is involucrate, with 3 to 5 involucra. As many as 200 carpospores can be produced from one carposporophyte. The carpospores develop into independant tetrasporophytic thalli. In the tetrasporophyte, the tetrasporangia are immersed in the thallus in the last 3 to 4 orders of branching of the thallus and are produced randomly from pericentral and cortical cells. Once the tetraspores are released the resulting gap in the cortex is filled by neighbouring cells. This morphological and developmental summary was derived from Hommersand (1963), Scagel (1967), Widdowson (1974), and Abbott and Hollenberg (1976). 19 2.4.2. Host specificity of Microcladia coulteri Microcladia coulteri is found epiphytic on large foliose red algae and occasionally on brown algae in the lower intertidal and upper subtidal zones of the Pacific coast of North America. It is found from Baja California to northern Vancouver Island, British Columbia (Abbott and Hollenberg 1976; Scagel 1951,1967). It has been reported predominantly on species of Gigartina, Prionitis and Grateloupia (Scagel 1967). From herbarium searches at the Universities of Southern California (AHFH) and British Columbia (UBC), M. coulteri was also found epiphytic on the rhodophytes Iridaea, Opuntiella, Callophyllis, Odonthalia, Carpopeltis and Laurencia and the brown algae, Egregia and Laminaria (Table 1). On collection excursions to Bamfield, Vancouver Island in the winter 1983-84, M. coulteri was also found on hard substratum (rock). Microcladia coulteri appears to be a generalist epiphyte capable of colonizing any available substratum within the tidal range it commonly occupies. As shown in Table 1, M. coulteri has been reported on species of Prionitis, Gigartina, Odonthalia and Iridaea in the Strait of Georgia which was the general location for this study. Initial investigations at the study location indicated that Gigartina was not a major component of the intertidal community. The following discussion will be restricted to the systematics and morpholog3' of the hosts: Prionitis, Iridaea and Odonthalia. 2.4.3. Prionitis J. Agardh 1851 Until recently, Prionitis was a member of the Cryptonemiales, but Kraft and Robbins (1985) suggest incorporation of the Cryptonemiales into the Gigartinales because of unsatisfactory distinctions between the two on the basis of "accessory" auxiliary cell filaments. The vegetative thallus is characterized by a tightly packed cortex of rows of cells and a medulla of densely interwoven filaments (Abbott and Hollenberg 1976). Prionitis lanceolata (Harvey) Harvey is the species found epiphytized by M. coulteri at the study site. The morphology of _P. lanceolata is highly variable. It has a characteristic intertidal thallus form that is irregularly branched and 200 to 300 mm tall, and a subtidal form that is 20 BRITISH COLUMBIA 0 J S V w 0 N S B C F T A A R , A 0 u G N S E C c J S C E C H G A A A T A 0 . . . L L RHODOPHYTA PRIONITIS X X X X X X X X GIGARTINA X X X X X X X X IRIDAEA X X X ODONTHALIA OPUNTIELLA X CALLOPHYLL IS X GRATELOUPIA X CARPOPELTIS X LAURENC IA X PHAEOPHYTA EGREGIA X LAMINARIA X X EPILITHIC ROCK X Table 1. Distribution of the epiphyte Microcladia coulteri on various host species along the west coast of North America. Data collected from the U.B.C. and A.H.F.H. herbaria. Geographic areas: OCOST - Outer coast of Vancouver Island J F U C A - Juan de Fuca Strait STGEO - Strait of Georgia VANC . - Vancouver Harbour WASH . - Washington State OREG. - Oregon N.CAL - Northern California S.CAL - Southern California B A J A - Baja California 21 regularly, dichotomously branched and can reach a height of 800 mm (Abbott and Hollenberg 1976). Major branches have pinnately arranged proliferous branchlets along the lateral margins of the thallus (Chiang 1970). Thalli are solitary or in tufts and are dull brown to reddish purple in color. Attachment to the substratum is by a small discoid holdfast. The thallus is multiaxial and growth is from marginal apical cells (Chiang 1970). 2.4.4. Iridaea Bory 1826 Iridaea is a member of the Gigartinales, family Gigartinaceae and is characterized by a tightly packed cortex of branched filaments and a medulla of interwoven filaments with much intercellular space (Abbott and Hollenberg 1976; K i m 1976). There are taxonomic difficulties in delineating Iridaea f laccida (Setchell and Gardner) Silva and Iridaea cordata (Turner) Bory. Abbott (1971) distinguished between the two species on the basis that the thallus o f l . flaccida was greenish in colour and that in the tetrasporangial thallus there is a visible margin at the edge of the blade free of tetrasporangia. Other workers have found neither character clearly distinguishable. Foster (1982) found that the greenish_I. f laccida would take on the purple coloration of_I. cordata when transfered into deeper water. The delination between the two species appears to be on distributional grounds. Iridaea flaccida has a mid to low intertidal distribution and_L cordata is more common subtidally (Abbott and Hollenberg 1976). Abbott (1972) found that vegetative shape can not be used to differentiate between the two species. Size thickness and color of blades were largely dependant on amounts of exposure. In Iridaea , development of upright thalli from perennating prostrate crusts is initiated by intercallary cells on the prostrate crusts (Norris and K i m 1972). Upright thalli are multiaxial and grow by means of a multiaxial system of filaments having many marginal apical cells. Growth occurs at the margins and immediately behind it where cell enlargement occurs. Although there are no vegetative differences delineating_I. flaccida from_I. cordata there are suggestions that they are different entities (Abbott 1971), exactly how different is different enough for them to be called different species has not been clearly indicated in the 22 literature (Abbott 1971, 1972). For that reason the study entities will be called Iridaea  cordata. 2.4.5. Odonthalia Lyngbye 1819 Odonthalia is a member of the Ceramiales, family Rhodomelaceae and is more closely related to the epiphyte, Microcladia coulteri (Ceramiales, Ceramiaceae) than the other two hosts. Odonthalia is polysiphonous, producing 4 pericentral cells, the axes are clothed in a dense pseudoparenchymatous cortex (Scagel 1955, 1967). Odonthalia floccosa (Esper) Falkenberg is dark brown in color, up to 400 mm tall with alternating distichous branchlets from major cylindrical branches (Abbott and Hollenberg 1976). It is abundant in restricted areas on intertidal rocks subjected to surf, from Aleutian Is., Alaska to Santa Barbara county, California (Garbary et_al. 1986). Tetrasporangial branchlets are borne in the uppermost portion of unmodified branchlets and are frequently aggregated, 2 opposite pericentral cells of each primary segment produces tetrahedrally divided tetrasporangia. The carposporophytes are borne on unmodified branchlets in clusters or alternately and distichously in subultimate ramuli. Spermatia cover the surface of unmodified branchlets in the last 3 orders of branching (Smith 1969; Masuda 1982). 23 C H A P T E R 3. M E T H O D O L O G Y 3.1 Descr ipt ion of the Study Locat ion. This study was carried out at Beaver Point on Saltspring Island, British Columbia (48° 47 'N, 123° 22'W). Saltspring is one of the largest Gulf Islands in the Strait of Georgia (Figure 1). Beaver Point is a prominent peninsula that juts eastward into Swanson Channel. The study location has a south east aspect leaving it unprotected in winter storms. The exposed aspect is compounded by the wake produced by the large (>300 car carrying capacity) ferries that regularly pass close by. Location of the study site is shown in relation to the coastline and access road in the inset map in Figure 2. The transect (Figure 2) was 50 meters in length, located in the lower intertidal on a coastline that trends north-south and dips into the ocean at slopes between 5 and 45°. The study area was selected after reconnaissance surveys of several potential locations on Saltspring and other lower Gulf Islands during April and June, 1984. The Beaver Point site was chosen because it contained extensive populations of the host species, P r i o n i t i s lanceolata, Iridaea cordata and Odonthalia floccosa. Three methods were used to mark the location of the permanent transect: 1. Temporary attachment of fluorescent surveyors tape to large algae found along the transect. 2. Cementing of stainless steel bolts into natural cracks in the bedrock with underwater epoxy. 3. Driving of pitons into natural cracks in the bedrock. The pitons were the most successful and most durable of the methods used. Cemented bolts came off with the abrasive action of waves. Figure 1. Location of Study Area. 25 Figure 2. Location of Transect within Study Area. The mid-littoral zone of Stephenson and Stephenson (1972) is shaded. Solid line is boundary between midlittoral and infralittoral fringe. Dashed line, midlittoral and supralittoral zones. Boulders are also shown. Four shaded circles are permanent markers (pitons), 0 and 50 represent the extent of transect. Inset shows location of study shown in relation to coastline and access road (land hatched). 26 Differences in elevation along the transect were determined using a device for measuring changes in water level. The device consisted of a clear plastic hose 30 meters long, attached at one end to a tripod (that was levelled) and to a measuring staff (in cm) at the other. A known volume of water was poured into the hose and the difference in height between the two menisci measured (and doubled) to determine changes in vertical height between locations along the transect. This method is inexpensive, accurate and can be managed by one person. Weaknesses in the method are loss of volume in the hose caused by evaporation and, trapping of air in the hose during filling with water. Variations in height above datum (altitudinal differences) with position along the transect are shown in Figure 3A. The datum at distance 0 was a permanent transect marker (piton). The wave-like pattern shown in Figure 3A represents the block fracturing and erosion marks on the bedrock seen at the study site. Maximum variation in height along the transect was approximately 0.12 meters, one quarter of the length of the 0.50 x 0.50 meter (0.25 m ) quadrats used. Slope perpendicular to the waterline was determined by the use of a compass mounted on a flat wooden clip board. The range of slopes (Figure 3B) was from 5 to 45°, the pattern being similar to the variation in height above datum. The variation in slope appears to be an expression of the dip and strike of the block fracturing in the bedrock. 3.2 Methods of Sampling Each month, random placement of eight 0.25 m quadrats along the transect was obtained by the use of a random numbers table (for N = 100; Lewis 1966). Sampling was without replacement. The sampling design was nested with 2 placements at each location, for 100 locations. The placement of 200 quadrats was therefore possible along the transect, 104 being sampled during the course of the thirteen month sampling program. Sampling without replacement resulted in the total number of quadrat placements F i g u r e 3 A . Topography of Transect. Figure 3B . Change in slop? of rock substratum along transect. SLOPE degrees HEIGHT ABOVE DATUM meters o i O o i O c n O m O u n I I I I I I I I L_ O cn I O _L_ Ul o _ l _ cn _L_ CJ o _l ! i ! i ! I 09 • • CO 29 along the transect being reduced by 8, each month. It could be argued that although there was a reduction in the total population size from which the monthly samples were taken, the sampling process was still random. Also, sampling without replacement represents a more linear representation of the population because redundancy of information caused by sampling the same part of the population twice cannot occur. Unfortunately, for a truly random sampling, the assumption of a homogeneous distribution of the population from which samples are taken is invalid for intertidal benthic communities. The intertidal environment has been said to be in a dynamic state of disturbance and recovery (Dethier 1984) with the resulting mosaic of species assemblages reflecting the disturbance history of the area (Sousa 1979). Thus reduction in the total population over time could represent a major departure from random sampling. This problem and a solution are summarized in Appendix 2. From the eight quadrats sampled each month all host thalli were harvested and the density of each host species determined. The mean and standard error were calculated for density for each host species from each monthly sample. The standard error of sampling was very large indicating a clumped (contagious) distribution for the host species (Zar 1974). When the variance to mean ratio is greater than 1, as in the case for these data, the summary distribution of a random sample from the population is the negative binomial (Pielou 1977). In the laboratory, random subsamples of 10 host thalli were taken from the plants collected in each quadrat. Sampling was randomized by collecting thalli by number from a random numbers table (Lewis 1966). This was found to be necessary as size bias otherwise occurred, larger sizes being sampled more than small ones. The distribution of epiphytes within subsamples was contagious with many thalli having zero or few epiphytes and a few thalli with many epiphytes. For each host thallus sampled, the epiphyte thalli were removed with forceps and sorted by size class and reproductive status. 30 Three size classes of epiphytes were recognized (Figure 4): size class 1 - less than 8 mm in height size class 2 - 8 to 20 mm in height size class 3 - greater than 20 mm in height Size classes 1 and 2 differed in reproductive status, size class 1 only being represented by vegetative thalli whereas size class 2 was also represented by reproductive gametophytic and sporophytic thalli. Reproduction was recorded by the occurrence of spermatangial, carposporangial and tetrasporangial epiphyte thalli. The carposporangial thalli were used to indicate syngamy, it was assumed that apomixis by apogamy was not occurring. With the removal of all epiphytes the host plants were wet weighed and sectioned for determination of reproductive status. Size classes were determined from wet weight. The same scale of size classes was used for all host species examined as this made possible a comparison of substratum availability. As epiphyte colonization was patchy upon individual thalli the use of alternative measurements such as surface area was decided against. Size classes used were the same as those determined by Hansen (1977) and Hansen and Doyle (1976) for Iridaea cordata on the central Californian coast. The six size classes were: size class 1 - less than 1.0 g wet weight size class 2 - 1.1 to 2.0 g wet weight size class 3 - 2.1 to 5.0 g wet weight size class 4 - 5 . 1 t o l 0 . 0 g wet weight size class 5 - 10.1 to 15.0 g wet weight size class 6 - greater than 15.1 g wet weight The justification for using size rather than age was: the fate of the individual can be predicted more accurately from its size (Werner 1975; Hughes and Jackson 1980); 1 ***** :?V) Figure 4. Size classes of the epiphyte Microcladia coulteri. 32 aging the algae accurately would be difficult; age is not a constant representation of availability of substratum for epiphyte colonization. Representative specimens of Microcladia coulteri and the three host species have been deposited in the phycological herbarium of the University of British Columbia (UBC). 3.3 Data Analysis Both host and epiphyte data were recorded in matrix format. The rows (cases) in the matrix were individual host thalli and the columns (variables) were: 1. Host variables -date, species of host, host wet weight, host size class, host reproductive status; 2. Epiphyte variables - total number of epiphytes, number of size class 1 epiphytes, number of vegetative, size class 2 epiphytes, number of carposporangial size class 2 epiphytes, number of spermatangial size class 2 epiphytes, number of tetrasporangial size class 2 epiphytes, number of vegetative size class 3 epiphytes, number of carposporangial size class 3 epiphytes, number of spermatangial size class 3 epiphytes, number of tetrasporangial size class 3 epiphytes, and; 3. Collection location along transect (quadrat number). A total of 2674 cases were observed throughout the thirteen months of sampling; 1221 (46%) were recorded with epiphytes. Data on epiphytized thalli from the host Prionitis lanceolata (N = 635) were examined for univariate (Zar 1974) and multivariate normality (Campbell 1980). Deviations from normality indicated that the data needed to be transformed before proceeding with further statistical analysis. The reasons for deviations from normal distribution were: 1. many zeros in some variables, and, 2. a contagious distribution (variance to mean ratio greater than 1) of the epiphyte variables. The data were tested for univariate normality by using the "normal probability transformation" (MIDAS: Statistical Research Lab., Univ. of Michigan) for each of the variables and observing the distributions of cumulative frequency plots of the transformed variables. If the frequency distribution appears as a straight line then the variable distribution is approximately normal. The 33 normal probability transformation is equivalent to plotting the sample cummulative distribution of a variable against a standardized normal co-ordinate axis. Standardized transformation of a normal distribution gives a mean of 0 and variance of 1 and once applied to a normal probability transformation results in the cummulative frequency approximating a straight line. No untransformed variables were normally distributed. The multivariate normal distribution is more robust than the univariate distribution. For multivariate data, observations (or cases) are often only found to be not normally distributed when the value for each variable is considered in relation to the other variables (Campbell 1980). Plotting the Mahalanobis squared distance against the order statistic for a chi-squared distribution combines examination of the distributional assumption of a multivariate normal distribution with detection of atypical observations. The cumulative frequency distribution of Mahalanobis squared distance estimates of means and covariances of the study variables (Campbell 1980) indicated that the data were severely skewed (Figure 5). . Since the analytical tools used in this study, namely, principal components analysis (PCA) based on the correlation matrix [R], and analysis of variance (ANOVA) which are both more robust when the normal probability distribution is not violated, it was necessary to transform the raw data before analysis. In the raw data, variations from normality was caused by: (1) variation in sample size (N) (Box 1953), and; (2) unequal sample variances (Underwood 1981). Of those transformations attempted, namely logarithmic, arc-sine, square root, and polynomial extensions of these, the best fit to both univariate and multivariate normal distributions was the quadratic root of X+ 1 (4/x + 1). The square root transformation has been found to be successful in stabilizing the variances from quadrat samples or samples where the variance to mean ratio is 1 or greater (Underwood 1981). If the variance to mean ratio is 1 then the distribution is Poisson (Random) and if it is greater than 1 then the distribution is a negative binomial. The value X + 1 was used because of the great number of zeros in some of the variables. The 34 Figure 5. Cumulative frequency distribution of Mahalanobis squared distance estimates for the epiphyte population growing on Pr ion i t i s lanceolata. Mahalanobis squared distances are: Dm 2 / 3 =[(Xm-X) T V- 1 (Xm-X'5? untransformed — transformed 4 X + 1 35 transformation approximated a normal distribution as shown in the multivariate probability plot (Figure 5). Principal components analysis (PCA) was performed on the correlation matrix [R] determined from the +1 transformed data. Why should one transform data for use in PCA? There are no distributional assumptions underlying the computations of P C A but normality is assumed for a probablistic interpretation of the P C A results. The multivariate normal distribution for multiple measures can be often assumed without testing because the multiple measurements are sums of small independent effects (Anderson 1958). This is the same central limit theorum argument that leads to the assumption of normality for a univariate measurement (Daultrey 1976). But, P C A performed on the correlation matrix [R] uses as a sample estimate Pearsons product-moment correlation coefficient (r). As noted by Kowalski (1972) " the distribution of r may be quite sensitive to non-normality and non-linearity, and the correlation analysis should be limited to where (X,Y) is normal." Therefore, it is necessary to at least have an understanding of the distributional properties of the data in relation to the multivariate normal distribution. P C A was performed from the P R I N C O M command from the M I D A S statistical package (Statistical Research Lab.: Univ. of Michigan). Tests for independence and equicorrelation (Morrison 1976) on the data set (N = 1221) indicated that the original variables were correlated with each other. If the tests indicated total independence and no equicorrelation then all original variables would be independent, suggesting P C A unecessary (Green 1976). The percent of total variation accounted for by the individual P C A axes was also calculated. For each component (axis) the eigenvectors for original variables were given and the component (axis) scores calculated. As P C A is primarily a transformation process, from a set of correlated variables to a derived set of uncorrelated variables, the main strength in the method comes in the furthur analysis of the component scores. In this study, component scores were analyzed 36 in a nested analysis of variance using the program U B C : A N O V A R . The question of which components to analyze was determined by eliminating all components whose eigenvalues are less than 1.0, on the grounds that these components were accounting for less of the total variation than any one original transformed variable (Pocock and Wisharf 1969). Note that when there is a low correlation between variables the above method may lead to the elimination of important dimensions of variability in the original data. The first three P C A axes were used for the A N O V A on the total data set (Chapter 5); P C A axes 1 and 2 were used in the A N O V A of the reproductive variables only (Chapter 6). The nested A N O V A model with 5 host variables and an error term was used. The A N O V A was undertaken to determine relative importance of different sources of variation in the data. There is too much emphasis on hypothesis testing in the modern usage of A N O V A as opposed to an estimation of source of variation, or as an information summary as proposed by the inventor, R. A . Fisher. The usage of A N O V A in this exercise is as an information summary. The A N O V A used in Chapters 5 and 6, with slight variations, was: Y l , Y 2 , Y3 = A + B(A) + C(AB) + D(AB) + F(AB) + E where Y n = P C A axis n A = time of year B = species of host C = size class of host D = reproductive status of host F = location along transect E = error The factor A measures variation due to the seasonality in the population structure of Microc ladia coulter i . The combination B(A) measures the variation due to between host species variation in epiphyte population structure. The combinations C(AB) and D(AB) measure variation accounted for by within host species variations in size class structure in host thalli and reproduction. The combination F(AB) measures variation accounted for by 37 the spatial distribution of host species. The E or error term consists of all unaccounted variation in the above A N O V A model. The F ratios (signal+noise/noise) which are the formal significance tests of the adequacy of a model fit are strongly affected by variations in sample size (N) and therefore are not given. 38 C H A P T E R 4 . T E M P O R A L V A R I A T I O N I N P O P U L A T I O N S T R U C T U R E O F T H E H O S T S P E C I E S . 4.1 Introduction This study addresses the question "how is the population structure of the epiphyte, Microc ladia coulteri maintained in a temporally and spatially patchy environment, where the patchiness is determined by availability of host thalli?" To answer this question it is first necessary to study how the population structure of the hosts vary in time and space. Temporal changes in abundance of Pr ioni t i s spp. in tidepools of Washington State showed little seasonality other than losses in populations during the autumn and winter from increased wave drag and storm damage (Dethier 1982). Seasonal fluctuations in the population density of Iridaea cordata has been reported as only occurring in the juvenile stages initiated in winter. Total density was reported as almost constant over the seasons, with observable changes in biomass being associated with the fluctuations in amount of juveniles (Hansen and Doyle 1976). Seasonal changes in size class distributions of Iridaea  cordata populations has been well documented (Hansen 1977). Blade initiation begins in winter with greatest growth of blades during the spring; mature plants dominate the population during summer-autumn with senescence during autumn-winter. Odonthalia  floccosa persists intertidally throughout the winter (November-January) as hollow basal branches. Growth is initiated in late winter and early spring from the distal end of the overwintering basal branches, with maximum size being reached by autumn (Goff and Cole 1976b). Seasonal changes in reproduction for Pr ion i t i s lanceolata are poorly documented. Abbott and Hollenberg (1976) noted that reproductive structures for mid and low intertidal zone populations of_P. lanceolata have been reported for most of the year; for subtidal 39 populations mainly in the summer. Both gametangial and tetrasporangial phases of_I. cordata were reported as being present over the entire year (Hansen and Doyle 1976). The tetrasporangial phase was dominant for most of the year except for spring when all reproductive phases were equally present. No information was available on the seasonality of reproductive phases of Odonthalia floccosa. With such a paucity of information on the seasonality in growth and reproduction and their expression on the population structure of the host species under study, no assumptions can be generated on the seasonal availability of substratum for colonization by Microc ladia  coulteri . Jernakoff (1985) tested the hypothesis that seasonal abundances of intertidal algae have general patterns that are temporally consistent. He found that seasonal patterns of abundance varied between years as well as on small spatial scales. No generalized patterns were found to exist. His observations infer that population structure at every location for every time is a unique event and not a general expression of a more global seasonal pattern. Considering the above, a study of variation in size structure and reproduction of the three host species was undertaken. The study consisted of: 1. Determining the density of erect host thalli for each month over the thirteen month sampling period; 2. Determining the size structure of the populations of the three hosts for each month of the sampling period; 3. Determining the reproductive size structure of the populations of the three host species for each sampling month. This approach allowed direct comparison of population structure of Microc ladia coulteri to the population structure of the host species. 40 4.2 Results Densities of erect host thalli per m 2 are show in Figure 6. The error bars shown (plus and minus one standard error) indicate there are large variances associated with the monthly mean densities. Variance to mean ratios are greater than one inferring that the species sampled have a contageous (Zar 1980) or clustered distribution. 4.2.1. Odonthalia floccosa o Odonthalia floccosa showed maximum densities of 600 to 800 erect thalli/m from n March to July and minimum densities of 120 to 160 erect thalli/m during September 1984 to January 1985 (Figure 6). It shows the greatest seasonal variation in thallus density of the three host species. Size class analysis of O. floccosa (Figure 7) indicated large fluctuations in the structure of the population with season. July 1984 and June 1985 have similar size class structure with a predominance of size class three thalli. Both reproductive phases shown (tetrasporophyte and carposporophyte) were present over the first 5 size classes. Carposporophytes were the dominant component of the reproductive thalli. Carposporophytes are used in this analysis to represent successful syngamy. It has been assumed that apogamous development of the carposporophyte is not occurring. No attempt has been made to account for the time between syngamy and carposporophyte development. By October the reproductive population was minimal and restricted to carposporophytes. Size class 1 thalli were the major component of the population. The predominance of size class 1 thalli continued until March 1985. Tetrasporangial thalli appeared in February and were followed by carposporophytic thalli in March and June. Many of the carposporophytic thalli recorded were immature until as late as May, 1985. Spermatangia were observed in February. Dominance of size class 1 thalli was replaced by greater occurrences of size class 2 and 3 thalli from March 1985 until the end of sampling. 1000 -. Figure 6. Change in density of erect host thalli/m during the sampling period. I - Iridaea O - Odonthalia P - Prionitis Figure 7. The change in size class distribution of vegetative, carposporangial and tetrasporangial thalli of Odonthalia floccosa during the sampling period. 0 -Vegetative | 'Tetrasporangial H -Carposporangial 43 40 -J 20 J 40-20-40 20 JS3 m «-I LU 3 60-1 40 H S 20-E 6 0 -4 0 -2 0 -r " r _UBI 80-60-| 40 2 0 - | J U L A U G " T - 1 1 SEP i i i O C T T 1 N O V i i — i 1 r D E C 1 1 — — 2 3 4 5 6 S I Z E C L A S S 1 2ZTa—r T 1 r r~—i r IS J A N 20 H i 1 r F E B 40H 20-M A R 20 A A P R 20-M A Y 20 J • J U N 20-J U L 20-- i r p 1 2 3 4 5 6 S I Z E C L A S S 44 4.2.2. Iridaea cordata o Iridaea cordata varied in density from 20 to 160 thalli/m over the entire year with no apparent seasonal pattern. Size class analysis of the_I. cordata population (Figure 8) indicated that the population size structure was relatively stable over the entire year. A n abundance of larger size classes occurred in summer followed by a loss in proportion of the larger size classes in winter (December, January, February). Reproductive thalli occurred as a component in varying proportions over the entire year. There was a strong alternation in the pattern of dominance between tetrasporangial and carposporangial (the measure of the female gametophytic population) thalli during the study period. Tetrasporophytes were a dominant reproductive component from July to November 1984 and April and May, 1985. Carposporangial thalli became more abundant by November, 1984 and was a dominant phase of reproduction through to May, 1985. The patterns of dominance between the two reproductive phases seem to be a month or two apart. The movement of reproductive phases through the size classes in Iridaea seems to be from larger size classes to smaller ones indicating that reproductive thalli were mature and senescing (Hansen and Doyle 1976). 4.2.3. Pr ioni t i s lanceolata o Densities of Pr ioni t i s lanceolata varied from less than 20 to 160 erect thalli/m with a winter peak in densities from November 1984 through January 1985. Pr ioni t i s lanceolata showed a different seasonal pattern in size class distribution than the other two host algae (Figure 9). Size class 1 was dominant over the entire year with a more even distribution of size classes only occurring during July and August, 1984 and not repeated in the following year. Very few size class 5 and 6 thalli occurred throughout the year. Reproductive components occurred throughout the year in small numbers. Reproductive thalli were a small proportion of the total thalli studied. This was unlike the other two host species investigated. Patterns in size class distribution and abundance of reproductive thalli were also hard Figure 8. The change in size class distribution of vegetative, carposporangial and tetrasporangial thalli of Iridaea cordata during the sampling period. gj) -Vegetative B - Tetrasporangial Q - Carposporangial 46 J U L 40-20-40-20-40-20-i I 1 A U G S E P O C T 40-K 20 > O z 40-N O V 20-i i JO. D E C 40 20-40-20-J A N M rn .n r * - r -2 3 4 5 6 SIZE C L A S S F E B I i . Ik. 20-1 r~—r MAR I i 20-A P R .3 i 20-M A Y J U N J Z L £2 20-I O , J U L 20-2 3 4 5 6 SIZE C L A S S 47 Figure 9 . The change in size class distribution of vegetative, carposporangial and tetrasporangial thalli of Prionitis lanceolata during the sampling period. ^ - Vegetative • " Tetrasporangial 0 - Carposporangial 48 2 3 4 I SIZE C L A S S SIZE C L A S S 49 to discern. The major trend was the loss of tetrasporangial thalli in July, 1984 and June and July, 1985. 4.3 Discuss ion "Availability of substratum" can be defined as the density of host species and their size class distribution. This is modified both by the changes at the surface of the host (e.g. loss of surface tissues, changes in surface of host caused by reproductive organs), and by the suitability of the particular host as a substratum. Greatest seasonal changes in population densities occurred in Odonthalia floccosa, which showed a summer maximum and winter minimum. Pr ion i t i s lanceolata occurred in higher densities in winter with minimums in June and July. No distinct seasonal pattern in density was observed in the Iridaea population. The density of the host thalli component in the 'availability of substratum' for epiphyte colonization, was weighted in favor of Odonthal ia in summer and shared equally between the three host species during winter. Odonthal ia floccosa is predominantly a summer alga, with both larger size classes and more reproductive thalli occurring in the summer months. During the rest of the year it was mainly represented by size class 1 vegetative thalli, overwintering basal branches which regenerate vegetatively in spring (Goff and Cole 1976b). Iridaea cordata occurred throughout the year but was characterized by blade initiation in winter, fast growth in spring, maturity in summer and senescence in autumn. P r i o n i t i s lanceolata had winter maxima in density and occurred predominantly as smaller size classes. Size class distribution was probably an artifact of the size class divisions chosen indicating little relationship between them and the actual maximum size and rates of growth of Pr ion i t i s . Modification of surface texture of the host can be caused by senescence, surface damage from wave action and herbivory, and formation of reproductive structures. A size class analysis of reproductive thalli was performed on the host species to investigate the 50 suitability for epiphyte colonization of changing surface structure caused by thalli becoming reproductive. Reproduction occurs on mature thalli which represent substrata that have been available for colonization by epiphytes for longer time periods than vegetative thalli. Reproductive thalli were not a major component in the population of jO. floccosa studied. In contrast with O. floccosa , Iridaea cordata had a predominance of reproductive thalli. Reproduction in Pr ion i t i s lanceolata was small but continuous throughout the entire year. No distinct patterns of seasonality in the abundance and size class distribution of reproductive thalli were found. A l l three host species exhibit a triphasic diplohaplontic life history with isomorphic alternation of generations. Diplohaplonts generally seem to show a great variation in reproductive pattern and it is often questionable if in nature they alternate regularly between a haploid, gamete-producing and a diploid, spore-producing generation (Lovlie and Bryhni 1978). Variations through apomeiosis (Kim 1976), from direct alternation of haploid and diploid generations have been well documented in the Gigartinales. Tetraspores of an isolate of Iridaea cordata var. splendens from Washington State germinated to form tetrasporophytes, whereas other isolates of this species from Washington and nearby British Columbia had regular alternation of generations. There are conflicting reports in the literature on frequencies of gametophytes to tetrasporophytes of I. cordata. Hansen and Doyle (1976) reported a higher frequency of tetrasporophytes to gametophytes in populations of_I. cordata from the central California coast. Abbott (1972) reported a very low frequency of tetrasporophytes (20%) of_I. cordata (as_I. flaccida) from the same area. Note that phase frequency differences may result from many causes not related to life history differences (differential rates of metabolism, growth, and herbivory are but some; West and Hommersand 1981). Patterns of distribution of predominantly tetrasporangial populations of algae at the extremes of their geographical range have been used to infer higher adaptiveness of diploid over haploid algae. The North Atlantic species Bonnemaisonia hamifera Harvey was found to recycle apomeiotically through the diploid tetrasporophyte phase at the northern extremes 51 of its distribution (Dring 1982). Hannach and Santelices (1985) question the adaptivity of diploidy over haploidy in Chilean species of Iridaea. Gametophytes dominated populations higher in the intertidal and grew faster in laboratory tests than the tetrasporophytes. The resulting picture of the algal life history is basically a triphasic diplohaplontic cycle with various degrees of vegetative reproduction and apomixis complexing the distributions of phases in the resultant population. Complex life histories are thought of as evolutionarily unstable (Istock 1967) because natural selection acts on the various phases independantly. The diversity of complex life history strategies found in red algae would suggest that either the phylum is evolutionarily unstable or that the diversity of complex life histories indicate that evolutionary theories which predict complexity (Brooks and Wiley 1986) may be more appropriate to the Rhodophyta. The theory of Brooks and Wiley (1986) also offers an explaination for the differences reported for Iridaea cordata by K i m (1976), Hansen and Doyle (1976) and Abbott (1971), since such differences reflect variation, one aspect of complexity. Vegetative propagation from basal crusts also determines the population structure shown for the three host species. Basal crusts are longer lived than the upright thallus but have a slower growth rate (Littler and Littler 1980). They also appear more resistant to grazing (Dethier 1982) although recent studies (Padilla 1985) have found crusts to take less energy to graze than upright forms. Once a post disturbance patch has been colonized by the basal crust then the population structure of the upright thalli have a predetermined ploidy level. There is evidence that basal crusts can consist of both haploid and diploid tissues. Thus, vegetative propagation from basal crusts confuses patterns of reproductive output. The combination of complex life history, different morphologies of the same phase within the life history, levels of herbivory, and interactions with other species and the abiotic environment make the seasonal patterns seen in host species population structure unique to 52 the time frame and location of the present study. The substrata (host thalli) are both temporally and spatially patchy. 53 C H A P T E R 5. T E M P O R A L A N D S P A T I A L V A R I A T I O N IN P O P U L A T I O N S T R U C T U R E O F M I C R O C L A D I A C O U L T E R I 5.1 Introduction As shown in studies by Kain (1982) and Whittick (1983) the expression of epiphyte population structure is intimately tied to the availability of host substratum during initial settlement and recruitment. Studies of epiphytes on seaweeds (Ballantine 1979, D'Antonio 1985) have indicated the importance of host shape and surface texture as a determinant of epiphyte community structure. Studies of community interactions in algal epifaunas (Seed and O'Connor 1981) infer no density dependant, interspecific competition within these communities resulting in a habitat where the community expression is a product of settlement (recruitment) patterns of epiphytes and availability of colonizable space. Availability of space suggests a resultant epiphyte community that is regulated by density dependant growth, however this appears not to be the case. Settlement of spores from moving waters (as described by Neushul 1972) resulting in clumped epiphyte distributions is density independant. Whether or not the population structure of Microcladia coulteri is regulated by density dependant interactions (intra - and interspecific competition and predation) or by differential availability of substratum for recruitment is not known. Patchiness in algal epiphyte distribution as shown by growth of Smithora naiadum concentrated along margins of seagrasses (Harlin 1970), distribution of epiphytes on branch apices of branched algal thalli (Medlin 1983, Ballantine 1979), and heavy growth proximal to the base of the thallus of Neorhodomela larix (D'Antonio 1985, as Rhodomela larix) suggest that clumped distributions of algal epiphytes are common. In this chapter the patterns of distribution of Microcladia coulteri over the seasons and on individual host species was investigated. Within host species variation in size, 54 reproductive status, and spatial abundance was investigated. The variation in the epiphyte population structure was then partitioned between these variables. 5.2 Results 5.2.1. Population structure of Microcladia coulteri The population structure of Microcladia coulteri was determined from the percentage of epiphytized host plants and the density of epiphytes on those host plants. The percentage of epiphytized host plants sampled each month (Figure 10) for each host species indicated differences in distribution of epiphytes between the host species. Prionitis lanceolata had the highest percentages of epiphytization (76 to 98.5% of host plants sampled) throughout the year. The highest percentage of host thalli containing M . coulteri occurred during winter (October 1984 to February 1985). Odonthalia floccosa was less heavily epiphytized (0 to 71%), with a distinct summer-autumn maximum (July to November 1984, June 1985) and winter-spring minimum (December 1984 to May 1985). Iridaea cordata had intermediate percentages of epiphytized host thalli (7 to 77%), with an autumn and late spring maximum (August to November 1984, May 1985) and a summer and winter-early spring minimum (July 1984 and June 1985, December 1984 to April 1985). There are differences in mean density of epiphytes per host thallus (Figure 11) between the three host species. Seasonal patterns in epiphyte population structure on the three host species were not interpretable due to large standard errors. However, the late summer-autumn (July to November 1984, June and July 1985) increase in epiphyte density on Odonthalia floccosa was distinct from winter-spring minimum densities (December 1984 to May 1985). Prionitis  lanceolata had the highest mean densities of epiphytes Qoetween 20 and 55 per host thallus), Iridaea cordata was intermediate (5 to 20 per host thallus), and Odonthalia floccosa was lowest (0 to 10 per host thallus). 55 Figure 10. Change in percentage of host plants epiphytized over the sampling period for the three host species. I - Iridaea O - Odonthalia P - Prionitis 56 80-i Figure 11. Change in density of epiphytes per host thallus over the sampling period for epiphytized thalli of the host species. I - Iridaea O - Odonthalia P - Prionitis 57 By multiplying the mean densities of Microc ladia coulteri per host thallus by the density of the host species per m 2 , a density of M . coulteri per m 2 was determined (Figure 12). Density of M . coulteri per m 2 is a function of both the degree of colonization of the host and of availability of host thalli (substrata). The population of M . coulteri on Pr ioni t i s  lanceolata was a major determinant of the total density of M . coulteri per m 2 for most of the year. For the summer months (July and August 1984, June and July 1985) the epiphyte population on Odonthalia floccosa was larger, due to high densities of the host, during the summer (Figure 6). Mean densities of the 3 size classes of M . coulteri for each of the host species are shown in Figure 13. For Iridaea cordata increases in size class 2 epiphytes occurred from October 1984 to March 1985 and for size class 3 epiphytes, from February and March 1985 (Figure 13A). Density of size class 1 epiphytes decreased over the sampling period. There was a 4 to 20 fold decrease in density of epiphytic thalli from size class 1 to size classes 2 and 3. For Odonthalia floccosa (Figure 13B) all epiphyte size classes showed summer-autumn maxima (July to October 1984, May to July 1985) and winter-spring minima (December 1984 to Apri l 1985). There was a 3 to 6 fold decrease in density of size class 1 epiphytes compared to size classes 2 and 3. For Pr ioni t i s lanceolata no seasonal pattern in epiphyte density was discerned (Figure 13C). There was a 6 to 30 fold decrease in densities of size class 1 epiphyte compared to size classes 2 and 3. 5.2.2. P C A and A N O V A Epiphyte density data presented in Figures 10, 11, 12, and 13 were transformed to an approximate multivariate normal distribution (i/X+ 1), and analyzed using principal components analysis (PCA). The relationship between original epiphyte population structure variables and P C A axes are interpreted from the eigenvector rose diagram (Figure 14). The first axis (PCA 1) is a measure of overall epiphyte population abundance with the highest positive values denoting greater amounts of vegetative epiphyte thalli.The second axis (PCA 2) 58 8000 6 0 0 0 -£ 4000H a 2000 -1984 1985 Time (Months) Figure 12. Change in density/m of the epiphyte Microcladia coulteri during the sampling period for the three host species. O Prionitis • Iridaea • Odonthalia • Total 59 Figure 13. The change in density of the three size classes of Microc lad ia coulteri per host thallus from epiphytized thalli of the three host species. A - Iridaea cordata B - Odonthalia floccosa C - Pr ion i t i s lanceolata 1 - size class 1 epiphytes 2 - size class 2 epiphytes 3 - size class 3 epiphytes EPIPHYTE THALLI PER HOST THALLUS 05 62 F igure 14. Eigenvector plot of the variables for vegetative and reproductive M i c r o c l a d i a coulteri from all host species. Tot V - total epiphytes I V - vegetative size class 1 epiphytes 2V - vegetative size class 2 epiphytes 2T - tetrasporangial size class 2 2S - spermatangial size class 2 2C - carposporangial size class 2 3V - vegetative size class 3 3T - tetrasporangial size class 3 3S - spermatangial size class 3 3C - carposporangial size class 3 63 separates reproductive from vegetative epiphytic thalli. The greater the positive loading on P C A 2 the greater the proportion of reproductive size class 2 epiphyte thalli. P C A axis scores (Figure 15) show that the three host species are unique in terms of seasonal changes in epiphyte population abundance. Prionitis lanceolata is positively dispersed along the first axis (PCA1 - with the exception of July 1985) and negatively dispersed along the second axis (PCA2 - exceptions being October 1984 and February, March 1985). Odonthalia floccosa is negatively dispersed along P C A 1 and positively dispersed along P C A 2 . Iridaea cordata is similar to Odonthalia except for July 1984 and February 1985. Prionitis lanceolata had the highest densities of epiphyte thalli as well as the greatest proportion of vegetative thalli. Odonthalia floccosa had low densities of epiphytes, but a higher proportion of reproductive size class 2 epiphyte thalli. Iridaea cordata also exhibited a high loading of reproductive epiphyte thalli combined with low densities of epiphytes. A seasonal fluctuation in dominance from vegetative to reproductive epiphyte thalli occurred on _P. lanceolata in October and November 1984 and January and February 1985. The epiphyte population on_I. cordata occurred in high abundance in winter and low abundance in summer. The P C A axis scores were then subjected to a nested A N O V A model. Table 2 gives the partitioning of variation between terms from the A N O V A model: The first P C A axis accounted for 40.99% of the total variation in the epiphyte population structure, the second 12.57%, and the third 9.9%, all three accounting for a total of 63.46%. The actual time of year (seasonality of Microcladia coulteri) accounted for a total determined variation of 3.54 %. The host species nested by time (between host species variation in epiphyte population structure) accounted for 7.40% of the variation. Reproductive status of the host nested by time (in months) accounted for determined variation totalling 4.75%. Size class of host nested by time and species (within species variation in size) accounted for variation totalling 15.36%. Location along the transect (nested by species and size class) accounted for 11.52% of determined variation. The error term was the largest single term (21.05%). 64 Figure 15. Plot of the monthly means of the PCA scores of the epiphyte variables for each of the host species. A - Pr ioni t i s lanceolata B - Odonthalia floccosa C - Iridaea cordata 65 PCA 2 • I I , p p p o O O O O o o a > j > k > o i o J > b > b o i • PCA 2 67 PCA 2 p bo L_ • O CO I o I o to O O O _ L _ O CD _ 1 _ S O U R C E PCA1 P C A 2 P C A 3 T O T A L 40.99% 12.57% 9.9% 63.46% A 2.75 0.31 0.48 3.54 B(A) 5.70 0.82 0.88 7.40 C(AB) 3.30 0.90 0.54 4.74 D(AB) 11.15 2.65 1.57 15.37 F(AB) 7.60 2.38 1.54 11.52 ERROR 10.50 5.49 5.06 21.05 Table 2. Percent variation accounted for by various sources on the first three P C A axes as determined with the nested A N O V A model. Percent variation for P C A i = SSi(factor)/SSi(total).Xi/Xtotal. P C A 1 , P C A 2 , P C A 3 = A + B(A) + C(AB) + D(AB) + F(AB) + E where P C A n = PCA axes scores A = time (months) B = host species C = reproductive status of hosts D = size class of host F = location of host along transect 69 To summarize, the A N O V A indicated that within host species variation in reproductive status, size of thallus and spatial distribution accounted for 31.63 % of the total variation in the epiphyte population structure. 5.2.3. Wi th in host var iat ion i n size and spatial distr ibut ion. There was a significant positive correlation between epiphyte abundance and host size for all host species (Table 3). A larger positive correlation between epiphyte population structure and increase in size of thalli of P r i o n i t i s occurred than between the other two host species. Spatial patchiness in epiphyte distribution was investigated using the negative binomial distribution (Pielou 1977). Table 4 gives the mean, variance and value for k for the epiphyte density on each host species for each month. To reproduce the distribution of the epiphyte between host thalli mean, variance and value of k are necessary. Note that for most of the sampling dates the variance to mean ratio was greater than 1. The value k is a measure of clumping in the epiphyte distribution; the smaller the k the greater the clumping. The epiphyte population on Iridaea cordata in July, 1985 had a variance to mean ratio approaching one and a value of k equal to 64.80 (nearing Poisson distribution) and so can be used as a reference point for a random distribution. Values of k for_P. lanceolata ranged from '1.55 to 0.32, whereas in_I. cordata and (). floccosa a greater range in k values was obtained. These data generally indicate that M . coulteri has a clumped distribution on all three hosts throughout the year of sampling. The contagious nature of the epiphyte distribution on host thalli is graphically illustrated in Figure 16. There were exponentially fewer host thalli with higher epiphyte densities. Natural logarithmic transformation of the x-axis (numbers of epiphytes per host thallus) resulted in a linear expression of epiphyte densities on the host species (Figure 17). The x-intercept approximates the maximum densities of epiphytes found on the host species, and the slope is a 70 IRIDAEA SIZE CL. 0.25 0.24 0.17 0.09 347 0.10 0.14 ODONTHALIA i SIZE CL. 0.19 0.21 0.13 0.00 235 0.13 0.17 , PRIONITIS ! SIZE CL. 0.41 0.41 0.40 0.29 633 0.08 0.10 ALL HOST SIZE CL. 0.11 0.10 0.11 0.03 1219 0.06 0.07 TOT.EPl. SIZE CL.1 SIZE CL.2 SIZE CL.3 d.f. R 0.05 R 0.01 Table 3. Correlations (Pearsons product-moment)between total epiphyte density and host size. M E A N V A R I A N C E X S 2 k Iridaea J U L Y 14.5 553.0 0.39 A U G U S T 11.8 1187.3 0.12 S E P T E M B E R 7.6 92.3 0.68 OCTOBER 16.6 504.5 0.56 N O V E M B E R 13.4 290.6 0.64 D E C E M B E R 8.6 79.4 1.04 J A N U A R Y 9.0 131.9 0.66 F E B R U A R Y 11.1 72.9 2.01 M A R C H 5.4 43.3 0.76 A P R I L 3.4 6.3 4.11 M A Y 4.5 31.8 0.75 J U N E 6.2 120.2 0.34 J U L Y 3.6 3.8 64.80 Odonthalia J U L Y 6.0 46.3 0.90 A U G U S T 5.5 28.0 1.37 S E P T E M B E R 4.1 30.2 0.63 OCTOBER 3.4 9.0 2.06 N O V E M B E R 3.4 4.9 7.32 D E C E M B E R -J A N U A R Y 3.0 - _ F E B R U A R Y - - -M A R C H -A P R I L 1.2 0.2 M A Y - - -J U N E 4.8 31.8 0.87 J U L Y 9.5 201.7 0.47 P r i o n i t i s J U L Y 55.3 2017.7 1.55 A U G U S T 28.9 1625.5 0.52 S E P T E M B E R 32.4 1018.9 1.06 OCTOBER 47.0 4250.4 0.52 N O V E M B E R 44.8 2541.1 0.80 D E C E M B E R 46.8 2209.3 1.01 J A N U A R Y 34.4 2621.8 0.46 F E B R U A R Y 17.4 284.9 1.12 M A R C H 16.7 302.1 0.98 A P R I L 29.7 893.3 0.47 M A Y 32.5 3361.0 0.32 J U N E 30.1 1546.5 0.60 J U L Y 13.0 508.1 0.34 Table 4. Mean, variance and k-value (negative binomial) for density of epiphytes for the host species by the months of sampling (1984-1985). 72 • ! O - o o - in O - o o r 10 o o o IB - o J. o o C M to z> - J < X r-r-O I OC t o LU O r-e >-x g_ - O LU C O z ID o CO o ID o o r o I o — r -o mVHl 1S0H dO «38Wf1N Figure 16. Distribution of epiphyte thalli on host thalli as a function of the number of host thalli with increasing numbers of epiphytes per host thallus. P - Prionitis lanceolata 0 - Odonthalia floccosa 1 - Iridaea cordata 73 Figure 17. The linear regression of the number of host thalli with increasing numbers of epiphytes per host thallus natural log transformed. Regression equation is: y = a + blnx. 74 function of successful colonization of the host thallus where low slopes represent increased colonization. The x-intercept was greatest for_P. lanceolata (approx. 90 epiphytes per host thallus), least for JO. floccosa (approx. 18 epiphytes) and intermediate for_I. cordata (approx. 42 epiphytes). Low values for the slope indicated more host thalli with greater number of epiphytes per thallus: Odonthalia floccosa had the greatest slope and the lowest maximum densities of epiphytes; Pr ioni t i s lanceolata exhibited the least slope and the highest maximum densities of epiphytes; Iridaea cordata was intermediate both in the expression of maximum density of epiphytes and slope. The slopes, intercepts, coefficients of determination and sample size for the epiphyte population on the host species over the four seasons are shown in Table 5. There was no significant variation in maximum epiphyte densities with the seasons for all host species (observe the r 2). The slope showed significant seasonal variation. Pr ion i t i s lanceolata was the most colonizable substratum throughout the year of sampling with low slopes during summer, autumn and winter. The slope was high during spring probably because of losses in larger size classes of the host during this time. Coefficients of determination for Pr ion i t i s were low (r =0.50) indicating that variability in epiphyte numbers per host thallus was high. Odonthalia floccosa had a slope 2 to 6 times greater than recorded for_P. lanceolata. The seasonality of O. floccosa was emphasized by no values for the winter and spring months. Coefficients of determination for (). floccosa(r =0.69) suggest low variability in epiphyte numbers per host thallus. Iridaea cordata had slopes that were intermediate between the other two host species. For_I. cordata, slope was greatly affected by sample size and o coefficients of determination were intermediate (r =0.57). 75 Odonthalia a floccosa b r 2 x i N Iridaea a cordata b r 2 x i N Autumn Winter 25.11 -8.59 0.77 8.6 46 14.44 9.40 -3.77 -2.42 0.57 0.62 46.1 48.6 142 97 Spring Summer ; 7.39 -2.60 0.69 17.1 34 4.10 8.15 -1.11 2.2 0.60 0.53 40.2 35.7 57 45 Pr ioni t i s lanceolata a 6.28 b -1.37 r 2 0.53 X i 97.9 N 165 6.78 -1.38 0.43 137.0 158 10.42 -2.41 0.65 75.5 181 5.39 -1.25 0.56 74.6 122 Table 5 . Y-intercept (a), slope Oo), coefficient of determination (r ), X-intercept (xp, and sample size (N) from total numbers of host thalli versus total epiphytes per thallus for the seasons. Regression equation is: y = a + blnx 76 5.3 Discuss ion 5.3.1 Populat ion Structure of Microc ladia coulter i . P r i o n i t i s lanceolata had the greatest percent of epiphytized host thalli, the greatest mean densities of epiphytes per host thallus and the greatest amounts of the epiphyte size classes per host thallus. Superficial observation would suggest it to be the most important host species in the perpetuation of the population of Microc ladia coulter i . Pr ioni t i s lanceolata had the highest density of M . coulteri per m 2 , however a low summer density of this host, combined with high densities of Odonthalia floccosa resulted in (). floccosa being more important as a substratum for M . coulteri during the summer. From the size class components for epiphytes on each host (Figure 13) initial colonization appeared more successful on Pr ioni t i s lanceolata than on the other two host species; although persistence of epiphytes to larger size classes was least common on this host. The trend towards more reproductive size class weighting along P C A 2 for (). floccosa and_I. cordata (Figure 15) suggested that larger percentages of the size class 2 and 3 components on these host species became reproductive. 5.3.2 Inferences from A N O V A results The partitioning of variation in the A N O V A indicates no host specificity was exhibited by Microc lad ia coulteri ; no distinct seasonality in density of vegetative and reproductive M . coulter i was shown, and between host species variation in epiphyte population structure was less than 50 % of the variation accounted for by within host species differences in size and reproductive status. The epiphyte population structure seen in this study results from interaction between available substrata and year round reproductive output and recruitment of M . coulter i . Russell (1983a) reported a similar result for the relationship between the host L a m i n a r i a digitata, and its ectocarpoid epiflora. 77 5.3.3. W i t h i n host species variat ion i n size of t h a l l i Within host species variation in epiphyte population structure due to variation in size of host thalli is complicated by the different growth responses of the host species. Blade initiation in Iridaea cordata on the central California coast (Hansen 1977) begins in winter, with near exponential growth in spring, maturation in summer and senescence in autumn. Such a pattern of growth is supported by the variations in population structure of Iridaea observed at Beaver Point, British Columbia (Figure 7). Iridaea cordata contained a larger and more diverse population of Microc ladia coulter i during autumn and early winter with greater numbers occurring on larger host thalli (size classes 4, 5, and 6). The main seasonal trend in population structure of Odonthalia floccosa was a summer maximum in both blade density and larger sized thalli and a winter minimum in blade density with predominance of smaller size classes. The seasonal pattern in population structure of M . coulteri was towards larger, more diverse populations (in size class and reproduction) on small to mid-sized thalli of _0. floccosa (size classes 1, 2 and 3) during late summer and early autumn. Large losses of tidepool populations of P r i o n i t i s lanceolata during the autumn and winter months were reported by Dethier (1982). Lower intertidal populations of P . lanceolata in this study showed both a predominance of size class 1 thalli over the winter as well as increased densities of thalli. Winter dominance of size class 1 thalli could be caused by reduced photosynthesis and growth, and/or increased drag from wave action (Denny et al. 1985). All three host species showed a reduction in their larger size classes during the winter months. The epiphyte population had its greatest density in autumn and winter on medium to large thalli of JP. lanceolata (size class 2, 3, and 4). The large variation in epiphyte population structure due to thallus size variation within host species is a result of temporal variation in patterns of growth and maturation of host 78 species. The results of this study confirmed the observation of Ballantine (1979) that epiphytism was heaviest on branched (i.e. Prionit is) rather than bladed (i.e. Iridaea) algal hosts. Ballantine (1979) also noted that higher densities of epiphytes occured on older portions of algal host thalli. My study also found higher densities of epiphytes on larger size classes of hosts and older host tissues, and suggest: 1. Time of colonization of available host tissues by epiphytes approximates the time needed for growth of hosts into larger size classes. 2. Changes in surface texture, chemistry and charge of surfaces occur with increasing age of host tissues and may affect the rate of epiphyte colonization through increased settlement and adhesion of spores (Linskens 1966, Niehof and Loeb 1974). 3. Epiphyte abundances decrease with proximity to meristematic regions (D'Antonio 1985, Ballantine 1979). The decrease is either due to fast-growing regions not offering suitable substratum for epiphyte colonization or, the effect of antibiotic substances produced in active meristematic zones on the hosts (e.g. L a m i n a r i a digitata - Hornsey and Hide 1976, Al-Ogily and Knight-Jones 1977). 4. Increased availability of substratum occurs with increased size. Settlement and growth of the epiphyte does not seem to be affected by density dependant interactions such as intra- and inter-specific competition but still is highly clumped in distribution. Increase in surface area of the host thallus may increase opportunity for epiphyte spore settlement by its interaction with the hydrodynamic environment (Vogel 1981, Nowell and Jumars 1984). Patchy distribution of diatoms on glass slides (Munteanu and Maley 1981) and macroalgae (Medlin 1983) have been associated with areas of turbulent flow across the substratum surface. 5.3.4 Spatial var ia t ion i n distr ibution of host and epiphyte tha l l i . Both the highly clumped distribution, spatially and temporally, of the host species and M i c r o c l a d i a coulter i account for the high percentage of total variation in epiphyte population structure accounted by spatial variation in distribution of host thalli. The results from chapter 79 4 indicate both spatial and temporal patchiness in the distribution of host species. Patchiness of intertidal organisms has been interpretted as the product of dispersal mechanisms of the organisms (Hoffman and Ugarte 1985; Connell 1985), available substratum (Sousa 1984), recruitment (Roughgarden et al. 1985; Dethier 1984), and effects of competition (Huang and Boney 1984; Taylor and Littler 1982), predation (Cubit 1984; Lubchenco 1983, 1978; Lubchenco and Gaines 1981; Lubchenco and Cubit 1980) and disturbance (Sousa 1984; Paine and Levin 1981) after recruitment. The intertidal environment is in a dynamic state of disturbance and recovery (Dethier 1984). The resulting mosaic of assemblages of organisms reflects the disturbance history of the particular area (Sousa 1979). Disturbance (both biotic and abiotic) prevents the monopolization of space by competitive dominants (Dayton 1971, Paine and Levin 1981) and opens patches for early successional species (Whittaker and Levin 1977). There are differences in community response to disturbance that can be attributed to life history characteristics of the component species (Hruby and Norton 1979; Kennelly and Larkum 1983; Sousa 1984). The seasonal pattern in population structure of the host species from this study result from both the disturbance history of the area and the life history characteristics that the individual host species exhibit. Spatial distribution of Microc ladia coulteri is partly determined by the pattern of distribution of host species, both temporally and spatially, and partly by factors affecting the organization of epiphyte populations and communities. These include interactions of the host with the macro-environment (Nowell and Jumars 1984), growth characteristics of the host (Russell 1983a,b) interactions between epiphytes (Seed and O'Connor 1981) and between epiphytes and herbivores (D'Antonio 1985). The population structure of Microc ladia coulteri exhibited a highly clumped spatial distribution (Table 4). The spatial distribution observed result from: 1. Different epiphyte carrying capacities between host species, 80 2. Differential success of colonization of epiphytes between different host species with the seasons. The exponential decline in numbers of host thalli with increased epiphyte densities could be caused by initial difficulty in settlement of epiphyte spores onto available substratum. After initial colonization, continued settlement of epiphyte spores could be assisted by the proximity of reproductive epiphyte thalli (maturity of initial colonizers). 8 1 C H A P T E R 6. PART IT IONING OF R E P R O D U C T I V E SIZE C L A S S E S OF M I C R O C L A D I A C O U L T E R I A M O N G T H E T H R E E HOSTS. 6.1 Introduction Microc lad ia coulter i exhibits a triphasic diplohaplontic life history with iteroparous reproduction. The number of dispersable products resulting from syngamy is enhanced through the multiplication effect of the carposporophyte (Searles 1980) which maximizes reproductive output and possibly increases the rate of colonization of host substrata. This chapter will characterize the temporal and spatial partitioning of the reproductive size classes of M. coulter i between the three host species . The relative importance of the host species as substrata containing the reproductive components of the epiphyte can be then determined. 6.2 Results Numbers of carposporangial, spermatangial, and tetrasporangial size class 2 and 3 epiphytes per host thallus on Pr ion i t i s lanceolata, Ir idaea cordata, and Odonthal ia  f loccosa are given in Figures 18, 19, and 20. The reproductive epiphyte population on_P. lanceolata (Figure 18) was more common in winter, with a predominance of size class 3, than summer. OnJ. cordata (Figure 19) no reproductive thalli were observed in the summer months of May, June and July. Maximum densities occurred in winter (size class 2; October to December 1984) and early spring (size class 3; January to April 1985). Reproductive epiphytes on (). floccosa (Figure 20) were low in density compared with those on P. lanceolata and_I. cordata. On (). floccosa higher densities of reproductive epiphytes occurred in summer and early autumn (June, July, August and September). Principal components analysis (PCA) showed that monthly changes occurred in the reproductive components of the epiphyte population on each of the host species. The plot of the eigenvectors for reproductive variables (Figure 21) indicated that distribution along 82 F igure 18. Seasonal change in mean number of carposporangial, spermatangial and tetrasporangial epiphyte thalli per epiphytized thallus from Pr ioni t i s lanceolata. 2 - size class 2 epiphytes 3 - size class 3 epiphytes Mean Number of Epiphytes 84 Figure 19. Seasonal change in mean number of carposporangial, spermatangial and tetrasporangial epiphyte thalli per epiphytized thallus from Iridaea cordata. 2 - size class 2 epiphytes 3 - size class 3 epiphytes 85 0 . 8 0 . 6 0.4H 0 . 2 0 . 8 CO g. 0 . 6 'Q. S3 0.4H E c CO cu 0 . 2 0 . 8 i > 0 . 6 H 0.4 H 0 . 2 H 0 . 0 - 6 — Q Carposporangial Spermatangial Tetrasporangial 1984 1985 86 Figure 20. Seasonal change in mean number of carposporangial, spermatangial and tetrasporangial epiphyte thalli per epiphytized thallus from Odonthalia floccosa. 2 - size class 2 epiphytes 3 - size class 3 epiphytes 87 0.8 0.6 -0.4H 0.2 0.8 CO g. 0.6 Q . © £ 3 C CD © 0.4 H 0.2 H 0.8 -<K--0.6 0.4 0.2 H 0.0 / ^ Carposporangial Spermatangial — o — a — < > -Tetrasporangial I n <3 O-1984 1985 88 0 .8 0 . 6 -0.2 0.3 0.4 0.5 P.C.A. 1 Figure 21 . Eigenvector plot of the variables for reproductive epiphytic thalli for all host species. 2T - Tetrasporangial size class 2 2S - Spermatangial size class 2 2C - Carposporangial size class 2 3T - Tetrasporangial size class 3 3S - Spermatangial size class 3 3C - Carposporangial size class 3 89 PCA 1 was determined by density of reproductive epiphyte thalli. Distribution on PCA 2 related more to the proportion of size class 2 (positive eigenscores) to size class 3 (negative eigenscores). Monthly means of the axis scores from the PCA are shown in Figure 22 a, b, and c. Reproductive epiphyte populations onJP. lanceolata (Figure 22a) occurred in highest densities and had a higher proportion of larger reproductive thalli. Reproductive epiphyte populations on_I. cordata (Figure 22b) and _0. floccosa (Figure 22c) occurred in much lower densities and smaller reproductive thalli. Reproductive epiphytes on_P. lanceolata were predominantly size class 3 through winter and spring (December 1984 to May 1985). Iridaea cordata contained a highly seasonal reproductive epiphyte component with maximum densities of size class 2 in late autumn-early winter, a maximum in size class 3 in late winter-early spring and virtually no reproductive epiphytes in summer. Reproductive epiphytes onO. floccosa exhibited lower densities and more even distribution between size class 2 and 3 epiphyte thalli. There was a pronounced summer-autumn seasonality in occurrence of reproductive epiphytic thalli on 0». floccosa. The PCA axis scores were subjected to a nested ANOVA and the partitioning of reproductive epiphyte population variation between the terms of the model was determined. The first P.C.A. axis accounted for 37.24% of the variation in the reproductive component of the epiphyte population, the second, 17.33% (Table 6). Within host variation in size and spatial variation accounted for 21.4% of variation in the reproductive component of the epiphyte population, almost 50% of the determined variation. Between host species variations only accounted for 2.1%. There are strong similarities in the allocation of variation; between host species, within host species, and spatially, between the results of the ANOVA for reproductive components of the epiphyte population and those reported for the total population structure (Table 2). Correlation between size of host thallus and the reproductive components of the epiphyte population is significant with both reproductive size class 2 and 3 Figure 22. Plot of monthly means of P C A scores of the reproductive epiphyte variables for each of the host species. A - Pr ioni t i s lanceolata B - Odonthalia floccosa C - Iridaea cordata 9] o 94 P C A 1 P C A 2 T O T A L 37.24 17.33 54.60 A 2.40 0.50 2.90 B(A) 1.70 0.40 2.10 C(AB) 2.10 1.00 3.10 D(AB) 9.30 3.80 13.10 F(AB) 5.70 2.60 8.30 ERROR 16.00 9.10 25.10 Table 6. Percent variation determined from Sum of Squares from nested A N O V A model for the reproductive component of the epiphyte population. Percent variation for P C A i = SSi(factor)/SSi(total)>iAtotal. P C A 1 , P C A 2 = A + B(A) + C(AB) + D(AB) + F(AB) + ERROR where PCAn = axis scores A = time (months) B = species of host C = reproductive status of host D = size class of host F = location along transect E = error 95 epiphytes and host wet weight (p < 0.01; Table 7). The largest correlation was with size class 3 epiphytes. Correlation between reproductive epiphytes and the time of year was also significant (p < 0.01; Table 8). Size class 3 reproductive epiphytes again showed the higher correlation, probably related to their high densities on the host Pr ioni t i s lanceolata. 6.3 Discuss ion The partitioning of variation in the reproductive population structure of Microc lad ia  coulter i , between and within host species, was very similar to that for the whole epiphyte population. Although the seasonal patterns in density of reproductive components of the epiphyte are strikingly different from host to host, the overall variation accounted for between host species was small. Within host species variations in size and spatial distribution accounted for approximately 50 % of the total determined variation. Weighting of the variation in epiphyte population structure to within host species size variations is not surpising. For successful colonization of host substrata by epiphytes, the time between initial settlement of epiphyte spores to the growth of a mature reproductive thallus must be less than the time needed for growth, maturation and senescence of the host tissue (Russell 1983b). Greater numbers of reproductive epiphyte thalli would be expected on larger host thalli (Table 7). Variations due to spatial distribution of host species are compounded by the dispersal distances over which M . coulter i spores travel. As indicated in Chapter 5, the variance to mean ratios for epiphyte densities were greater than 1 with corresponding k values indicating a pronounced clumping in the epiphyte distribution. The clumping suggested that the location of an epiphyte on a host thallus increases the probability that subsequent epiphytes will develop on that thallus (adapted from Thomas 1977). Two processes may contribute to clumping: 96 Host Wet Weight Size class 2 Rep. 0.1763 Size class 3 Rep. 0.2610 Total Rep. Epiphytes 0.2745 N = 635 R(0.05) = 0.0778 d.f. = 633 R(0.01) = 0.1022 Table 7. Correlations (Pearsons product-moment) between reproductive epiphyte density and host wet weight. Time (months) Host species Size Class 2 Epi . 0.0584 0.0816 Size Class 3 Epi . 0.2029 0.2191 Total Rep. Epi . 0.1897 0.2109 Host Species 0.9581 N = 635 r(0.05) = 0.0778 d.f = 633 r(0.01) = 0.1022 Table 8. Correlations (Pearsons product-moment) between reproductive epiphyte density, host species and time of year. 97 1. Growth of one epiphyte thallus enhances the settlement of other spores of the epiphyte species. This could occur through development of turbulent eddies around the epiphyte (Vogel 1981, Munteanu and Maley 1981) or chemical attractants as observed in barnacles and other invertebrates (Crisp 1965); 2. The epiphyte matures and reproduces progeny which settle around the parent. Recruitment of visible size class 1 epiphyte thalli in mid-winter was approximately 25 days (Appendix 1). Growth and maturation of_I. cordata took approximately 9 to 12 months (Hansen 1977). From herbivory experiments (Dethier 1982), growth and maturation in_P. lanceolata was at least greater than 130 days and as long as 6 to 9 months (personal observations). The seasonal trend in density of reproductive Microc ladia coulteri was not only correlated to patterns in availability of host substrata. Size class 3 reproductive epiphytes had a significant correlation with time. Low densities observed for reproductive epiphytes during May and June may be associated with daytime low tides in combination with high insolation. In summary, Microc lad ia coulteri was found to be reproductively active throughout the entire year. Reproductive epiphytes were mainly found on Odonthalia floccosa during the autumn months (August to November) when the host thalli were mature and senescent. On Iridaea cordata reproductive epiphytes were abundant during winter; maximum densities occurring on mature and senescent thalli. On the host Pr ioni t i s lanceolata reproductive epiphytes occurred in high densities over most of the year. 98 C H A P T E R 7. G E N E R A L C O N C L U S I O N S A N D S U M M A R Y The distribution and abundance of an epiphyte in the intertidal environment has been shown to be an interaction between occurrence of suitable host substrata for colonization and the patterns of epiphyte recruitment. Microcladia coulteri is a generalist epiphyte whose population structure at Beaver Point resulted from a combination of continuous reproductive output and the fortuitous availability of host substrata. Prionitis lanceolata had the highest carrying capacity for the epiphyte. In PCA analyses of the total and reproductive components of the epiphyte population, P. lanceolata was found to support a large proportion of all vegetative size classes and size class three reproductive thalli throughout the year. Seasonal reduction in the availability of this host occurred during the summer months. Odonthalia floccosa showed the least epiphyte carrying capacity of the three host species, but reproductive populations of the epiphyte developed on this host during the late summer and early autumn. The highly seasonal epiphytization of _0. floccosa occurred during a period when the other hosts were less available for colonization. Iridaea cordata contained a seasonal epiphytic flora that occurred during late autumn and winter. Carrying capacity, colonizability and size of the reproductive components of the epiphyte were intermediate between the other two host species. Partitioning of the variation in the epiphyte population structure in an analysis of variance indicated that differences in epiphyte colonization between host species did not account for a large part of total variation. Although the patterns of epiphyte population structure among host species are visibly different, within host species variations in size of thallus and spatial distribution of host thalli accounted for nearly 50% of the total variation. Analysis of the epiphyte population by size class of host species indicated that the size 99 distribution of the host was a major determinant of the density and population structure of the resultant epiphyte population. Size and age of hosts have been shown to affect species richness of epiphytic communities (Ballantine 1979) and the relative density of reproductive components in epiphyte populations (Russell 1983a, b). A highly clumped (negative binomial) distribution of epiphytes was found to occur on all host species. Many host thalli contained zero or a few epiphytes whereas a few host thalli contained many epiphytes. It was inferred from this that the epiphyte population structure was the result of the dispersal and settlement of epiphyte spores. Further study should center on two major hypotheses that can be generated from the results obtained from my research. First, 'patterns i n dispersal and settlement of the epiphyte are major determinants of the h ighly c lumped spatial distr ibution of Microc lad ia coulteri ' . Studies on the settlement of spores onto host thalli in moving waters would be necessary to characterize the effect of differential settlement on the resulting epiphyte distribution. Second, 'availabil i ty of host substrata for colonization is a major factor affecting the resultant population structure of Microc ladia coulteri ' , and possibly many other generalist epiphytes. From the data on epiphyte density obtained in this study, a model determining expected epiphyte densities from the density and size class distribution of host species can be generated. Testing the expected epiphyte densities obtained from the model, against field populations of the host species would either support or refute the hypothesis. 100 LITERATURE CITED Abbott, LA. 1971. On the species of Iridaea (Rhodophyta) from the Pacific Coast of North America. Syesis_4: 51-72 Abbott, LA. 1972. Field studies which evaluate criteria used in separating species of Iridaea (Rhodophyta). p265-275. In: Abbott LA. and M. Kurogi (eds.) Contributions to the Systematics of Benthic Marine Algae of the North Pacific. Japanese Society of Phycology: Kobe, Japan. Abbott, LA. and G.J. Hollenberg 1976. Marine Algae of California. Stanford University Press: Stanford. 827 pp. Abugov, R. 1985. Is there a cost of meiosis in life history. J.Theor.Biol. 116: 613-624. Agardh, J.G. 1851. Species, genera et ordines algarum. Lund: Vol. 2, pt. 2, sect. 1: 337-505. Allen, M.B. 1967. Antibiotic properties of Marine Algae. Proc. Seminar on Sea, Salt and Plants (V. Krishnamurthy ed) Allen, P.J. 1976. Spore germination and its regulation. In: Pirson, A and M.H. Zimmermann (eds.), Physiological Plant Pathology. Springer-Verlag: Berlin. Al-Ogily, S.M. and E.W. Knight-Jones 1977. Anti-fouling role of antibiotics produced by marine algae and bryozoans. Nature (Lond.) 265: 728-729 Amano, M., S. Hara and N. Taga 1982. Utilization of dissolved amino acids in seawater by marine bacteria. Mar. Biol. 68: 31-36. Anderson, T.W. 1958. An Introduction to Multivariate Statistical Analysis. Wiley: N.Y. 704 pp. Apt, K.E. 1984. The morphology of Phaeocolax kajimurai (Rhodomelaceae, Rhodophyta), a host specific epiphyte on Lobophora variegata (Phaeophyta). Phycologia 23: 239-247. Ballantine, D.L. 1979. The distribution of algal epiphytes on macrophyte hosts offshore from La Parguera, Puerto Rico. Bot.Mar. 22: 107-110. Ballantine, D.L. and Humm H.J. 1975. Benthic algae of the Ancote Estuary: Epiphytes on seagrass leaves. Fla. Sci. 38: 150-162. Bell, J.D., A.S. Steffe and M. Westoby 1985. Artifical seagrass: How useful is it for field experiments on fish and marine invertebrates. J. Exp. Mar. Biol. Ecol. 90: 171-177. 101 Boland, W., L. Jzenicke, D . G . Muller and A. Peters 1984. Differentiation of algal chemoreceptors: A comparative receptor study with two seaweed pairs: Cutleria multifida / Syringoderma phinneyi and Desmarestia aeuleata / Ectocarpus siliculosus (Phaeophyceae). Eur. J . Biochem. 144: 169-176. Boney, A . D . 1975. Muscilage sheaths of spores of red algae J . Mar. Biol. Assoc. U . K . 55: 511-518. Box, G.E.P. 1953. Non-normality and tests of variance. Biometrika 40: 318-335. Brooks, D.R. and E.O. Wiley 1986. Evolution as entropy: Toward a unified theory of biology. Univ. Chicago Press: Chicago, xiv + 335pp. Bulthuis, D.A. and W.J. Woelkerling 1983. Biomass accumulation and shading effects of epiphytes on leaves of the seagrass, Heterozostera tasmanica, in Victoria, Australia. Aquatic Bot. 16: 137-148. Campbell, N.A. 1980. Robust proceedures in multivariate analysis. I. Robust covariance estimation. App. Stat. 29: 231-237. Cattaneo, A. 1978. The Microdistribution of epiphytes on the leaves of natural and artificial macrophytes. Br. Phycol. J . 13: 183-188. Chamberlain, A . H . L . and L . V . Evans 1973. Aspects of spore production in the red alga, Ceramium. Protoplasma 76: 139-159. Charters, A . C , M . Neushul and C. Baulotti 1969. The functional morphology of Eisenia  arborea. Proc. Int. Seaweed. Symp._6: 89-105. Charters, A . C , M . Neushul and D . A . Coon 1972. Effect of water motion on algal spore attachment. Proc. Int. Seaweed Symp._7: 243-247. Chiang, Y . M . 1970. Morphological Studies of Red Algae of the Family, Cryptonemiaceae. Univ. Calif. Publ. Bot. 58. Cole, K. and R.G. Sheath 1980. Ultrastructural changes in major organelles during spermatial differentiation in Bangia (Rhodophyta). Protoplasma 102: 253-279. Connell, J . H . and R.O. Slayter 1977. Mechanisms of succession in natural communities and their roles in community stability and organization. Am. Nat. I l l : 1119-1144. Coon, D . A . , M . Neushul and A . C . Charters 1972. The settling behavior of marine algal spores. Proc. Int. Seaweed Symp._7: 237-242. Cooper, R.W. 1983. The mechanisms and significance of enzymatic degradation of host cell walls by parasites. pplOl-135 In: J .A. Callow (ed.) Biochemical Plant Pathology. John Wiley and Sons: N.Y. Crisp, D.J . 1965. Surface Chemistry, a factor in the settlement of marine invertebrate larvae. Proc. Eur. Mar. Biol. Symp._5: 51-65. 102 Cubit, J.D. 1984. Herbivory and the seasonal abundance of algae on a high intertidal rocky shore. Ecology 65: 1904-1917. D'Antonio, C. 1985. Epiphytes on the rocky intertidal red alga, Rhodomela la r ix (Turner) C. Agardh: negative effects on the host and food for herbivores? J. Exp. Mar. Biol. Ecol. 86: 197-218. Daultrey, S. 1976. Principal Components Analysis. Concepts and Techniques in Modern Geography_8. Geo. Abstracts Ltd. Norwich, ppl-51. Dayton, P.K. 1971. Competition, disturbance and community organization: the provision and subsequent utilization of space in a rocky intertidal community. Ecol. Monogr. 41: 351-389. Denny, M.W., T.L. Daniel and M.A.R. Koehl 1985. Mechanical limits to size in wave swept organisms. Ecol. Monogr. 55: 69-102 Dethier, M.N. 1982. Pattern and process in tidepool algae: Factors influencing seasonality and distribution. Bot. Mar. 25: 55-66. Dethier, M.N. 1984. Disturbance and recovery in intertidal pools: maintenance of mosaic patterns. Ecol. Monogr. 54: 99-118. De Wreede, R.E. 1984. Growth and age class distribution of Pterygophora cal i fornica (Phaeophyta). Mar. Ecol. Prog. Ser. 19: 93-100. Drew, E.A. 1969. Uptake and metabolism of exogenously supplied sugars by brown algae. New Phytol. 68: 35-43. Dring, M.J. 1982. Geographic distribution of marine plants, pp 139-155. In: Dring M.J. The Biology of Marine Plants. Edward Arnold: London. Ducker, S.C, N.J. Foord and R.B. Knox 1977. Biology of Australian seagrasses: the genus A m p hibo l i s C. Agardh (Cymodoceae). Aust. J. Bot. 25: 67-95. Evans, L.V. and A.O. Christie 1970. Studies on the ship fouling alga, Enteromorpha. I. Aspects of the fine structure and biochemistry of swimming and newly settled zoospores. Ann. Bot. 34: 451-466. Evans, L.V., J.A. Callow and M.E. Callow 1973. Structural and physiological studies of the parasitic red alga, Holmse l la . New Phytol. 72: 393-402. Fenical, W. 1975. Halogenation in the Rhodophyta: a review. J. Phycol. JLl: 245-259. Fetter, R. and M. Neushul 1981. Studies on developing and releasing spermatia in the red alga, T i f fanie l la snyderae (Rhodophyta). J. Phycol. _17: 141-159. Fletcher, R.L. 1979. Studies on attachment in the red algal genus, Po lys iphonia . Br. Phycol. J. 14: 123(abstract). Fletcher, R.L. and R.E. Baier 1984. Influence of surface energy on the development of the green alga, Enteromorpha. Mar. Biol. Letters_5: 251-254. 103 Floc'h, J .Y. and M . Penot 1970. Mise en evidence d'une repartition preferentielle de divers cations le long du thalle des Laminaires. C. R. Acad. Sci. 271: 288-291. Floc'h, J .Y. and M . Penot 1971. Repartition preferentielle du phosphore et movements de redistribution du 3 2 P chez quelques laminaires. C. R. Acad. Sci. 273: 1100-1103. Floc'h, J .Y. and M . Penot 1972. Transport du 3 2 P et du 8 6 Rb chez quelques algues brunes: Orientation des migrations et fostervoices de conduction. Physiol. Veg. 10: 677-686. Foster, M.J . 1982. Factors controlling the intertidal zonation of Iridaea flaccida (Rhodophyta). J . Phycol. 18: 285-294. Frilion-Myklebust, C. and T . A . Norton 1981. Epidermis shedding in the brown seaweed, Ascophyllum nodosum (L.) Le Jolis and its ecological significance. Mar. Biol. Letters_2: 45. Garbary, D.J . , L . Golden, M.W. Hawkes and R.F. Scagel 1986. A Synopsis of the Benthic Marine Algae of British Columbia, northern Washington and southeast Alaska. Phycological Contribution N o . l , Botany Dept., Univ. of B.C.: Vancouver. Glombitza, K.W., M . Forster and W.F. Farnham 1982. Antibiotics from algae: Pt 25. Polyhydroxyphenyl Ethers from the brown algae, Sargassum (Yendo) Fensholt (Part II). Bot. Mar. 25: 449-453. Goff, L . J . 1975. A developmental, physiological and structural study of the rhodophycean alloparasite, Harveyella mirabilis (Choreocolaceae, Cryptonemiales). Ph.D. Thesis. University of British Columbia: Vancouver. 239pp. Goff, L . J . 1979. The biology of Harveyella mirabilis (Cryptonemiales, Rhodophyceae) VI. Translocation of photo-assimilated 1 4 C . J.Phycol. 15: 82-87. Goff, L . J . 1982. The biology of parasitic red algae. Prog. Phyc. Res J.: 289-369. Goff, L . J . and K. Cole 1976a. The biology of Harveyella mirabilis (Cryptonemiales, Rhodophyceae) III. Spore germination and subsequent development within the host, Odonthalia floccosa (Ceramiales, Rhodophyta). Can. J . Bot. 54: 268-280. Goff, L . J . and K. Cole 1976b. The biology of Harveyella mirabilis (Cryptonemiales, Rhodophyceae) IV. Life history and phenology. Can. J . Bot. 54: 281-292. Gonzalez, E . A . and L.J . Goff 1980. Erythrocystis saccata: A biochemically obligate symbiont? J . Phycol. H>(suppl.):pl4 (abstract). Green, R .H. 1976. Sampling Design and Statistical Methods for Environmental Biologists. Wiley Interscience PubhN.Y. pp 1-257. Greville, R.K. 1830. Algae Britannicae. Edinburgh. LXXXVIII + 218pp., 19 pis. Hanic, L . A . and J.S. Craigie 1969. Studies on the algal cuticle. J . Phycol._5: 89-102. 104 Hannach, G. and B. Santelices 1985. Ecological differences between the isomorphic reproductive phases of two species of Iridaea (Rhodophyta, Gigartinales). Mar. Ecol. Prog. Series 22: 291-303. Hansen, J .E . 1977. Ecology and natural history of Iridaea cordata (Gigartinales, Rhodophyta): Growth. J . Phycol. 13: 395-402. Hansen, J .E . and W.T. Doyle 1976. Ecology and natural history of Iridaea cordata (Rhodophyta, Gigartinaceae): Population structure. J. Phycol 12} 273-278. Harkin, E . 1981. Fluctuations in epiphyte biomass following Laminaria hyperborea canopy removal. Proc. 10th Int.Seaweed Symp. Harlin, M . M . 1970. Epiphytic Marine Algae: Interactions with their Hosts. Ph.D. Thesis. University of Washington: Seattle. 132pp. Harlin, M . M . 1973. Transfer of products between epiphytic marine algae and host plants. J . Phycol._9: 243-248. Harlin, M . M . 1974. The surfaces that seaweeds grow on may be a clue to their control. Maritimes (Uni.R.I. Grad. Sch. Oceanogr. 18: 7-8. Harlin, M . M . 1980 Seagrass epiphytes, pp 151-171. In: R.C. Phillips and C P . McRoy (ed.) Handbook of Seagrass Biology: An Ecosystem Perspective. Garland S.T.P.M. Press :N.Y. Harlin, M . M . and J.S. Craigie 1975. The distribution of photosynthates in Ascophyllum  nodosum as it relates to epiphytic Polysiphonia lanosa. J . Phycol. 11: 109-113. Harlin, M . M . and J . M . Lindberg 1977. Selection of substrata by seaweeds: Optimal surface relief. Mar. Biol. 40: 33-40. Harrison, P.G. 1982. Control of microbial growth and of amphipod grazing by water soluble compounds from leaves of Zostera marina. Mar. Biol. 67: 225-230. Hartmann, T. and E . Lohr 1983. Decarboxylation of environmental L-Leucine by marine red algae. Mar. Biol 78: 7-12. Hartog, Den C. 1972. Substratum: plants-multicellular plants. In: Kinne O. (ed.) Marine Ecology Vol 1. Environmental factors. Wiley Interscience: London. Harvey, W . H . 1853. Nereis Boreali - Americana II. Rhodospermeae. Smithonian Contributions to Knowledge. 5(Art. 5) 258pp. pis. 13-36. Hay, M . E . 1984. Predictable spatial escapes from herbivory: How do these affect the evolution of herbivore resistance in tropical marine communities. Oecologia 64: 396-407. Heijs F . M . L . 1985. The seasonal distribution and community structure of the epiphytic algae on Thalassia hemprichii (Ehrenb.) Ashers. from Papua New Guinea. Aquatic Botany 21: 295-324. 105 Hoffmann, A . J . and R. Ugarte 1985. The arrival of propagules of marine macroalgae in the intertidal zone. J . Exp. Mar. Biol. Ecol. 92: 83-95. Hommersand, M . 1963. The morphology and classification of some Ceramiaceae and Rhodomelaceae. U.C.(Berkley) Publ. Bot. 35: 214-223. Hornsey, I.S. and D. Hide 1976. THe production of antimicrobial compounds by British marine algae III Distribution of antimicrobial activity within the algal thallus. Br. Phycol. J . 11: 175-181. Hruby, T. and T . A . Norton 1979. Algal colonization on rocky shores of the Firth of Clyde. J . Ecol. 67: 65-77. Huang, R. and A . D . Boney 1984. Growth interactions between littoral diatoms and juvenile marine algae. J . Exp. Mar. Biol. Ecol. 81: 21-45. Hughes, T.P. and J .B.C. Jackson 1980. Do corals lie about their age? Some demographic consequences of partial mortality, fission and fusion. Science 209: 713-715. Humm, H.J . 1964. Epiphytes of the seagrass, Thalassia testudinum in Florida. Bull. Mar. Sci. Gulf Caribb. 14: 306-341. Hurlburt, S.H. 1984. Pseudoreplication and the design of ecological field experiments. Ecol. Monogr. 54: 187-211. Istock, C.A. 1967. The evolution of complex life cycle phenomina: An ecological perspective. Evolution 21 : 592-605. Jernakoff, P. 1985. Temporal and small scale spatial variability of algal abundance on an intertidal rocky shore. Bot. Mar. 28: 145-154. Jones, W.E. and A. Demetropoulos 1968. Exposure to wave action: Measurement of an important ecological parameter on rocky shores of Anglesey. J . Exp. Mar. Biol. Ecol_2: 46-63. Kain, J . M . 1971. Synopsis of biological data on Laminaria hyperborea. F . A . O . Fish. Synopses 87: 1-74. Kain, J . M . 1982. The reproductive phenology of nine species of Rhodophyta in the subtidal region of the Isle of Man. Br. Phycol. J . 17: 321-331. Kennelly, S.J. and A . W . D . Larkum 1983. A preliminary study of temporal variation in the colonization of subtidal algae on an Ecklonia radiata community. Aquat. Bot. 17: 275-282. Kim, D . H . 1976. A study on the development of cystocarps and tetrasporangial sori in Gigartinaceae (Rhodophyta, Gigartinales). Nova Hedwigia 27: 1-146. Kitting, C.L. , B. Fry and M . D . Morgan 1984. Detection of inconspicuous epiphytic algae supporting food webs in seagrass meadows. Oecologia 62: 145-149. Koehl, M.A.R. and S.A. Wainwright 1977. Mechanical adaptations of a giant kelp. Limn. Oceanog. 22: 1067-1071. 106 Kolattukudy, P.E. and W. Roller 1983. Fungal penetration of the first line defensive barriers of plants. p79-100. In: J .A . Callow (ed.) Biochemical Plant Pathology. John Wiley and Sons: N.Y. Kowalski, C.J. 1972. On the effects of non-normality on the distribution of the sample product-moment correlation coefficient. App. Stat. 21: 1-12. Kroes, H.W. 1970. Excretion of mucilage and yellow brown substances by some brown algae in the intertidal zone. Bot. Mar. 13: 107-110. Kraft, G.T. and P.A. Robins 1985. Is the order Cryptonemiales (Rhodophyta) defensible? Phycologia 24: 67-77. Kugrens, P. 1974. Light and electron microscopic studies on the development and liberation of Janczewskia gardneri Setch. spermatia (Rhodophyta). Phycologia 13: 295-306. Kugrens, P. and J .A. West 1972. Ultrastructure of spermatial development in the parasitic red algae Levringiella gardneri and Erythroeystis saccata. J . Phycol. 8: 331-343. Kugrens, P. and J . A . West 1973. The ultrastructure of an alloparasitic red alga, Choreocolax polysiphoniae. Phycologia 12: 177-186. Lewis, A . E . 1966. Biostatistics. Reinhold Publ.Corp: New York, pp 1-227. Lewis, R.J. 1982. The Biology of the Host-Specific Ephytic Red Alga, Ceramium codicola and some other epiphytes of Codium spp. in British Columbia. M.Sc. Thesis. University of British Columbia: Vancouver. 75pp. Lieberman, M . , D . M . John and D. Lieberman 1979. Ecology of subtidal algae in seasonally devastated cobble substrates of Ghana. Ecology 60: 1151-1161. Linskens, H . F . 1963. Beitrag zur Frage der Beziehungen zwischen Epiphytund Basiphyt bei marinen Algen. Publ. Staz. Zool., Napoli 33: 274-293. Linskens, H.F . 1966. Adhaesion von Fortpfamzungszellen benthontischer Algen. Planta(BerL) 68: 99-110. Linskens, H . F . 1968. Host pathogen interaction as a special case of interrelations between organisms. Neth. J . PI. Path. 74: 1-8. Littler, M . M . and D.S. Littler 1980. The evolution of thallus form and survival strategies in benthic marine macroalgae: Field and laboratory tests of a functional form model. Amm. Nat. 1_16: 25-44. Lobban, C S . and D . M . Baxter 1983. Size class distribution of the red algal epiphyte, Polysiphonia lanosa on its brown algal host, Ascophyllum nodosum in the Bay of Fundy, Canada. Bot. Mar. 26: 533-538. Lovlie, A. and E. Bryhni 1978. On the relationship between sexual and parthenogenetic reproduction in haplodiplontic algae. Bot. Mar. 21: 155-163. 107 Lubchenco, J . 1978. Plant species diversity in a marine community: importance of herbivore food preference and algal competitive abilities. Am. Nat. 112: 23-39. Lubchenco, J . 1983. L i t tor ina and Fucus : Effects of herbivores, substratum and plant escapes during succession. Ecology 64: 1116-1123. Lubchenco, J . and J .D. Cubit 1980. Heteromorphic life histories of certain marine algae as adaptations to variations in herbivory. Ecology_6: 676-687. Lubchenco, J . and S.D. Gaines 1981. A unified approach to marine plant-herbivore interactions: I. Populations and communities. Ann. Rev. Ecol. System. 12: 405-437. Lucas, C E . 1947. The ecological effects of external metabolites. Biol. Rev. 22: 270-295. McBride D.L. 1972. Studies of some British Columbian representatives of the Erythropeltidaceae (Rhodophyceae: Bangiophycidae). Ph.D. thesis. University of British Columbia: Vancouver, xi + 108pp, 40pl. McBride, D.L . and K. Cole 1971. Electron microscopic observations on the differentiation and release of monospores in the marine red alga, Smithora naiadum. Phycologia 10: 49-61. McRoy, C P . and J.J. Goering 1974. Nutrient transfer between the seagrass Zostera  mar ina and it's epiphytes. Nature 248: 123-124. Magruder, W . H . 1984. Specialized appendages on spermatia from the red alga, Aglaeothamnion neglectum (Ceramiales: Ceramiaceae) specifically bind with trichogynes. J . Phycol. 20: 436-440. Markham, J .D. 1969. Vertical distribution of epiphytes on the stipes of Nereocystis  luetkeana (Mertens) Postels & Ruprecht. Syesis_2: 227-240. Marshall, W. 1960. An underwater study of the epiphytes of L a m i n a r i a hyperborea (Gunn.)Fosl. Br. Phycol. Bull._2: 18-19. Masuda, M . 1981. Taxonomic notes on Odonthalia l y a l l i i (Harvey) J. Agardh and related species (Rhodophyta). J.Fac.Sci., Hokkaido Univ., ser. 5(bot.) 12: 147-158. May, V . , A . J . Collins and L . C Collett 1978. A comparative study of epiphytic algal communities on two common genera of seagrasses in eastern Australia. Aust. J . Ecol._3: 91-104. Medlin, L .K . 1983. Community analysis of epiphytic diatoms from selected species of macroalgae collected along the Texas coast of the Gulf of Mexico. Ph.D. thesis, Texas A & M University:College Station. 136pp. Morrison, D.F. 1976. Multivariate Stastical Methods. 2nd Edition. McGraw Hill: New York, pp 1-415. Moss, B.L. 1982. The control of epiphytes by H a l i d r y s si l iquosa (L.)Lyngb. (Phaeophyta:Cystoceraceae) Phycologia 21: 185-191. 108 Muller, D . G . , I.Maier and G. Gassmann 1985. Survey on sexual pheromone specificity in Laminariales (Phaeophyceae). Phycologia 24: 475-477. Munteanu, N . and E.J . Maley 1981. The effect of current on the distribution of diatoms settling on submerged glass slides. Hydrobiologia 7_8: 273-282. Muscatine, L. 1967. Glycerol excretion by symbiotic algae from corals and Tridachna and it's control by the host. Science 156: 516-518. Neushul, M . 1972. Functional interpretation of benthic marine algal morphology, pp 47-73. In: L A . Abbott and M.Kurogi (eds.) Contributions to the Systematics of Benthic Marine Algae of the North Pacific. Jap. Soc. Phycol: Kobe. Neushul, M . , M.S. Foster , D.A. Coon, J.W. Woessner and B.W.W. Harger 1976. An in situ study of recruitment, growth and survival of subtidal marine algae: techniques and preliminary results. J. Phycol. 12: 397-408. Ngan, Y. and I.R. Price 1979. Systematic significance of spore size in the Florideophyceae (Rhodophyta). Br. Phycol. J . 14: 285-303. Niehof, R. and G. Loeb 1974. Dissolved organic matter in seawater and the electric charge of immersed surfaces. J . Mar. Res. 32: 5-12. Niehuis, P .H. 1969. The significance of the substratum for intertidal algal growth on the artificial rocky shore of the Netherlands. Int. Revue. Ges. Hydrobiol. 54: 207-215. Nonomura, A . M . 1979. Development of Janczewskia morimotoi (Ceramiales) on its host Laurencia nipponica (Ceramiales, Rhodophyceae). J . Phycol. 15: 154-162. Nonomura, A . M . and J .A. West 1981. Host specificity of Janczewskia (Ceramiales, Rhodophyta). Phycologia 20: 251-258. Norall, T . L . , A . C . Mathieson and J .A. Kilar 1981. Reproductive ecology of four subtidal red algae. Exp. Mar. Biol. Ecol. 34: 119-136. Norris, R.E. and D . H . Kim 1972. Development of thalli in some Gigartinaceae. p 253-264. In: L A . Abbott and M . Kurogi (eds.) Contributions to the Systematics of Benthic Marine Algae of the North Pacific. Jap.Soc.Phycol: Kobe. Norton,T.A.,A.C. Mathieson and M . Neushul 1982. A review of some aspects of form and function in seaweeds. Bot. Mar. 25: 501-510. Nowell, A . R . M . and P.A. Jumars 1984. Flow environments of aquatic benthos. Annual Review of Ecology and Systematics 15: 303-328. Okuda, T. and M . Neushul 1981. Sedimentation studies of red algal spores. J . Phycol. 17: 113-118. Orth, R.J. and J . von Montfrans 1984. Epiphyte-seagrass relationships with an emphasis on the role of micrograzing: A review. Aquatic Bot. 18: 43-69. 109 Padilla, D.K. 1985. Structural resistance of algae to herbivores. A biomechanical approach. Mar. Biol. 90: 103-109. Paine, R.T. and S.A. Levin 1981. Intertidal landscapes: disturbance and the dynamics of pattern. Ecol. Monogr. 51: 145-178. Pardy, R.L. and L . Muscatine 1973. Recognition of symbiotic algae by H y d r a v i r i d i s : A quantitative study of the uptake of living algae by aposymbiotic H . v i r i d i s . "Biol. Bull. 145: 565-579. Penhale, P.A. and W.O. Smith 1977. Excretion of dissolved organic carbon by eelgrass (Zostera marina L.) and its epiphytes. Limnol. Oceanogr. 22: 400-407. Penot, M . 1974. Ionic exchange between tissues of A s c o p h y l l u m nodosum (L.) Le Jolis and Polys iphonia nodosum (L.) Tandy. Z. Pflanzenphysiologie 73: 125-131. Pielou, E .C. 1977. Mathematical Ecology. Wiley-Interscience Publications, New York, pp 1-385. Pocock, D.C.D. and D. Wisharf 1969. Methods of deriving multi-factor uniform regions. Trans. Inst. British Geographers 47: 73-98. Ragan, M . A . and A. Jensen 1979. Quantitative studies on brown algal phenols. III. Light mediated exudation of polyphenols of A s c o p h y l l u m nodosum (L.) Le Jolis. J . Exp. Mar. Biol. Ecol. 36: 91-101. Rawlence, D.J. 1972. An ultrastructural study of the relationships between rhizoids of Polys iphonia lanosa (L.) Tandy (Rhodophyceae) and tissue of A s c o p h y l l u m nodosum (L.) Le Jolis (Phaeophyceae). Phycologia 11: 279-290. Rawlence, D.J. and A.R.A. Taylor 1970. The rhizoids of Po lys iphonia lanosa. Can. J. Bot. 48: 607-611 Risk, M.J . 1973. Settling plates of cold-cure acrylic plastic replicated from natural surfaces. Limnol. Oceanog. 18: 801-802. Roughgarden, J . Y. Isawa and C. Baxter 1985. Demographic theory for an open marine population with space limited recruitment. Ecology 66 : 54-67. Russell, G. 1983a. Formation of an ectocarpoid epiflora on blades of L a m i n a r i a digitata. Mar. Ecol. Prog. Series 11: 181-187. Russell, G. 1983b. Parallel growth patterns in algal epiphytes and L a m i n a r i a blades. Mar. Ecol. Prog. Series 13: 303-304. Russell, G. and C.J . Veltkamp 1984. Epiphyte survival on skin shedded macrophytes. Mar. Ecol. Prog. Series 18: 149-153. Sand-Jensen, K. 1977. Effect of epiphytes on eelgrass photosynthesis. Aquat. Bot._3: 55-63. 110 Scagel, R.F. 1951. An annotated list of the marine algae of British Columbia and Northern Washington. National Museum of Canada Bull. 150. Biol. Ser. 52. pp 1-289. Scagel, R.F. 1955. A morphological study of some dorsiventral Rhodomelaceae. Univ. Calif.(Berkley) Publ. Bot. 27: 1-108. Scagel, R.F. 1967. Guide to common seaweeds of British Columbia. B.C. Provincial Museum Handbook 27: 3-330. Scott, J .L . and Dixon P.S. 1973. Ultrastructure of spermatium liberation in the marine red alga, Pt i lota densa. J . Phycol._9: 85-91. Searles, R.B. 1980. The strategy of the red algal life history. Am. Nat. 115: 113-120. Seed, R. and R.J. O'Conner 1981. Community organization in marine algal epifaunas. Ann. Rev. Ecol. Syst. 12: 49-74. Sieburth, J . McN. 1969 Studies on algal substances in the sea III the production of extracellular organic matter by littoral marine algae. J . Exp. Mar. Biol. Ecol._3: 290-309. Silberstein, K. 1980. The effect of epiphyte shading on the primary productivity of Posidonia austral is and_P. sinuosa. M . Sc. University of Western Australia: Nedlands. 106pp. Smith G . M . 1969. Marine Algae of the Monterey Peninsula: Secound Edition. Stanford University Press: Stanford. 752 pp. Sousa, W.P. 1979. Disturbance in marine intertidal boulder fields: The non-equilibrium maintanance of species diversity. Ecology 60: 1225-1239. Sousa, W.P. 1984. The role of disturbance in natural communities. Ann. Rev. Ecol. Syst. 15: 353-391. Steinberg, P.D. 1985. Feeding preferences of Tegula funebralis and chemical defences of marine brown algae. Ecol. Monogr. 55: 333-349. Stephenson, T . A . and A. Stephenson 1972. Life between tidemarks on rocky shores. W . H . Freeman and Co: San Francisco. 425 pp. Taylor, P.R. and M . M . Littler 1982. The roles of compensatory mortality, physical disturbance, and substrate retention in the development and organization of a sand influenced rocky-intertidal community. Ecology 63: 135-146. Thomas, R.W. 1977. An introduction to quadrat analysis. Concepts and techniques in Modern Geography 12: 18-25. Tolbert, N.E . and L.P. Zill 1965. Excretion of glycolic acid by algae during photosynthesis. J . Biol. Chem. 222: 895-906. Tukey, H.B. Jr. 1970. The leaching of substances from plants. Ann. Rev. Plant. Physiol. 21: 305-324. I l l Turner, C.H.C. and L.V. Evans 1977. Physiological studies on the relationship between A s c o p h y l l u m nodosum and Po lys iphonia lanosa. New Phytol. 79: 363-371. Turner. C.H.C. and L.V. Evans 1978. Translocation of photoassimilated 1 4 C in the red alga, Polys iphonia lanosa. Br. Phycol. J. 13: 51-55. Turner. D. 1808. Fuci,sive Plantarum Fucorum Generi a botanicis adscriptarium descriptiones et historia Vol.1 pits 1-71. Tveter-Gallagher, E.,A.C. Mathieson and P. Cheney 1980. The ecology and developmental morphology of male plants of Chondrus crispus (Gigartinales, Rhodophyta). J. Phycol. 16: 257-264. Underwood, A.J. 1981. Techniques of analysis of variance in experimental marine biology and ecology. Oceanogr. Mar. Biol. Ann. Rev. 19: 513-605. Van Der Ben, D. 1969. Les epiphytes des feuilles de Pos idonia oceanica sur les cotes francais de la Mediterranee. Proc. Int. Seaweed Symp._6: 79-84. Vogel, S. 1981. Life in moving fluids: The physical biology of flow. Willard Grant Press: Boston. 352 pp. Weis, D. 1980. Free maltose and algal cell surface sugars are signals in the infection of Paramec ium bursar ia by algae. In: W. Schemmler and H.E.A. Schenk (eds.) Endocytobiology: Endosymbiosis and Cell Biology. Vol.1 Walter de Gruyter and Co.: Berlin-NY. Werner, P.A. 1975. Predictions of fate from rosette size in Teasel (Dipsacus fuHonum L.) Oecologia 20: 197-201. West, J.A. and M.H. Hommersand 1981. Rhodophyta: Life Histories. ppl33-193. In: Lobban C. and M. Wynne (eds.) The Biology of Seaweeds. University of California Press: Berkeley Wetherbee, R. 1979. "Transfer Connections": Specialized pathways for nutrient translocation in a red alga. Science 204: 858-859. Wetherbee, R. and F.J. Scott 1980. The fine structure and distribution of transfer connections in the red alga, Po lys iphonia . Micron 11: 509-510. Wetherbee, R., H.M. Quirk, J.E. Mallett and R.W. Ricker 1984. The structure and formation of host-parasite pit connections between the red algal alloparasite, Harveyel la mirabi l i s and it's red algal host, Odonthalia  floccosa. Protoplasma 119: 62-73. Wetzel, R.G. 1969. Excretion of dissolved organic compounds by aquatic macrophytes. Bioscience. 19: 539-540. Widdowson, T.B. 1974. The marine algae of British Columbia and Northern Washington: revised list of keys part III Rhodophyceae (red algae). Syesis_7: 143-186. Whittaker, R.H. and S.A. Levin 1977. The role of mosaic phenomenon in natural communities. Theor. Pop. Biol. 12: 117-139. 112 Whittick, A. 1983. Spatial and temporal distributions of dominant epiphytes on the stipes of laminaria hyperborea (Gunn.) Fosl. (Phaeophyta: Laminariales) in S.E. Scotland. J. Exp. Mar. Biol. Ecol. 73: 1-10. Wynne, M.J . 1985. Microcladia exserta sp. nov. (Ceramiaceae, Rhodophyta) from the East Coast of South Africa. Br. Phycol. J . 20: 193-200. Yamanouchi, S. 1906. The life history of Polysiphonia violacea. Bot. Gaz. 42: 401-449. Zar, J . H . 1974. Biostatistical Analysis. Prentice-Hall: New Jersey. 620 pp. 113 A P P E N D I X 1 It has been shown that the fate of an individual can be predicted more accurately from its size than its age because for many organisms the growth rates of individual organisms within a cohort can vary enormously. This variability is probably due to a combination of genetic and environmental differences and chance (Werner 1975; Hughs and Jackson 1980; Highsmith et al. 1980). Correlation between age and size classes of Microc lad ia coulter i would be advantageous as this would increase understanding on colonization of the epiphyte on the hosts. An i n s i tu colonization experiment was devised to estimate the time needed to colonize a substrate and for growth from one size class to another.The results are far from definitive since many environmental and biological factors were not controlled. Colonization experiments were done in October and December of 1984, and June, 1985. Results are only available for December, 1984, due to diffuculties with methodology and weather. The experiment consisted of punching holes in blades of epiphyte free erect Iridaea sp. with a leather punch. The exact location of the punched thallus was determined from a compass bearing and distance from a permanent marker (piton). The punched thalli were retrieved approximately 25 days later and the number and size class of epiphytes on the thalli were determined. All the host thalli were of size class 1 and 2 and determined to be free of epiphytes by a visual inspection before being punched. Problems encounted with this experimental technique included: 1. Loss of large numbers of Iridaea thalli from autumn and winter storms. This was especially the case in October 1984; 2. Identification difficulties with tagged thalli caused by heavy grazing pressure. The grazing pressure in June 1985 resulted in hundreds of 'leather punch-like' holes in Iridaea thalli; 3. Selective punching of size class 1 and 2 thalli of Iridaea. These size classes didn't contain a 114 large proportion of the epiphyte populations observed in the monthly samples (Chapter 5); 4. In addition spores may have been on thalli determined as being epiphyte free. Other host species were not investigated due to the difficulties of seeing small epiphytes on them, even in the laboratory. Attempts made to surface sterilize Iridaea thalli using an I,KI solution (Russell 1983) resulted in death of cultured thalli. Surface sterilization in the field was not attempted. Results from the December 1984 experiment were that 25% of all size class 1 and 2 Iridaea thalli retrieved after 23 days were colonized by size class 1 epiphytes within 23 days (Table A. 1). No size class 2 and 3 epiphytes were observed. It should be noted that from +200 host thalli punched only 52 were retrieved. Only 12.5% of size class 1 Iridaea thalli were epiphytized compared to 36% of the size class 2 thalli. These results can be interpreted in two ways. Either they represent the timing between spore settlement and growth of the epiphyte into size class 1 thalli; or it indicates a preferential settlement and growth of epiphytes on size class 2 Iridaea thalli. Use of in situ tagging of epiphyte free Iridaea thalli to monitor growth on a temporal basis was not a success. A more distinctive tagging method must be developed for ease of identification. Monthly sampling to observe seasonal changes in the population structure of Microcladia  coulteri appears long enough to observe changes in the size structure of the epiphyte. The experiment indicated that it took 25 days for a size class 1 epiphyte to be discernable on Iridaea thalli. 115 Colonization Experiment 1 plants punched 3/12/84 plants retrieved 26/12/84 No of days 23 Size Class (Iridaea sp.) No Epiphytes Epiphytes #1 21 40% 3 6% #2 18 35% 10 19% Table A l . The size class, number and relative percent of Iridaea thalli containing epiphytic thalli 23 days after tagging epiphyte free thalli. 116 A P P E N D I X 2 The sampling program was initially set up to randomly sample 8, 0.25 m 2 quadrats destructively for each month, over a period of 13 months. A buffer zone between each sampling location of 0.25 m 2 was to be left to counteract any effects on algal community structure caused by destructive sampling. The effect of sampling without replacement, on the total combinations of randomly placing 8 quadrats in 100 locations with time was not considered. Without replacement, N, the total number of quadrat placements available, is reduced by 8 every month. A large loss in the population size was visible by the second month and the sampling design was altered to a random design with 2 quadrat placements nested for 100 sampling locations. For the sampling design chosen, the probability of random placement of 8 quadrats a month was measured. The total combinations possible by placing 8 quadrats within an initial N of 200 was determined using the following formula ni x ni-1 x ... ni-8 8! where ni = initial N at the begining of each sampling time The combinations were determined and plotted (figure A2-1). The results indicate an exponential decrease in probability of random placement with sampling times. Reduction in the randomness of sampling with removal, over the 13 months is approximately 0.5% that of the beginning. Although the total combinations of placing 8 quadrats at each sampling period starts at 5.51 x 10 1 3 and only is reduced to 2.58 x 10 1 1 by the thirteenth month, this reduction could result in not fulfilling basic assumptions surrounding the use of normal distribution statistics. Namely that the sample is a random sample. 117 From this exercise I would strongly recommend that nested sampling designs be used when sampling is done without replacement. Nested designs also help reduce pseudoreplication (Hurlbert 1984). How large nesting need be can be determined from the duration of the sampling program. For my study, the ideal design would be 13 quadrat placements (for 13 months) nested at 100 sampling locations. This would result in no loss in N over the sampling period. Figure A .2.1. Total combinations of random placement of 8 quadrats without replacement over 15 months of sampling. 1. - x/y plot 2. - x/log y plot 119 6 0 0 0 1 1 — 1 - I 1 0 5 10 15 n u m b e r o f s a m p l i n g s 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items