UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Intertidal community structure, dynamics and models : mechanisms and the role of biotic and abiotic interaction Kim, Jeong Ha 1995

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

Item Metadata

Download

Media
831-ubc_1995-05988x.pdf [ 3.63MB ]
Metadata
JSON: 831-1.0088097.json
JSON-LD: 831-1.0088097-ld.json
RDF/XML (Pretty): 831-1.0088097-rdf.xml
RDF/JSON: 831-1.0088097-rdf.json
Turtle: 831-1.0088097-turtle.txt
N-Triples: 831-1.0088097-rdf-ntriples.txt
Original Record: 831-1.0088097-source.json
Full Text
831-1.0088097-fulltext.txt
Citation
831-1.0088097.ris

Full Text

Intertidal community structure, dynamics and models: mechanisms and the role of biotic and abiotic interactions by Jeong Ha Kim B.Sc., Sung Kyun Kwan University, 1984 M.Sc., Western Illinois University, 1988  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Botany) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA June 1995 © Jeong Ha Kim, 1995  ___  ____________________________  In presenting this thesis in partial fulfilment of the requirements for degree at the University of British Columbia, I agree that the Library freely available for reference and study. I further agree that permission copying of this thesis for scholarly purposes may be granted by the department  or  by  his  or  her  representatives.  an advanced shall make it for extensive head of my  It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department  of  The University of British Columbia Vancouver, Canada Dateb  DE-6 (2/88)  (9  11  ABSTRACT  This study is a comprehensive and experimental approach to understand the dynamics producing structure in an intertidal algal community. Both biological factors (competition and herbivory) and physical factors (disturbance and physical stress) were investigated through field manipulative experiments, non-manipulative monitoring and laboratory experiments for the last 3 years in Barkley Sound, Vancouver Island. The community studied contains three dominant perennial macroalgae, Mazzaella cornucopiae (Rhodophyta), Fucus distichus and Pelvetiopsis limitata (Phaeophyta), and some ephemeral algae as well as coexisting invertebrates such as barnacles, limpets and snails. Biological interactions and their mechanisms for all these organisms turned out to be complicated and non-hierarchical interaction networks which include reversal of dominance and indirect interaction. There was evidence for both negative and positive interactions existing between the same pair of species. Competitive dominance was changed depending on the developmental stages of the competitors and due to differences in morphology, and also the outcomes of interactions were modified by physical stresses (i.e., desiccation and wave action). Snails preferentially grazed the red alga, and limpets reduced the abundance of the ephemeral algae (the early colonizers) which otherwise inhibited the settlement of the later successional species such as the fucoids. Barnacles facilitated the colonization of Fucus and ephemeral algae after disturbance. Responses of the three dominant algae and ephemerals to the different sizes and time of disturbance were species-specific and depended on the alga’s life history and reproductive characteristics. The study shows that both the biological and physical factors influenced the structure of this community, and that the non-hierarchical interactions among the major species and their well-balanced responses to disturbance support a high likelihood of maintaining the current diversity in this algal community.  111  TABLE OF CONTENTS Page Abstract  .  ii  Table of Contents  iii  List of Tables  vii  List of Figures  viii  Acknowledgments  xi  CHAPTER 1: Thesis introduction  1  General Introduction  1  The study site and community  4  Organization of the thesis  6  REFERENCES  7  CHAPTER 2: Patterns of interactions among neighbor species in a high intertidal algal community  10  1NTRODUCTION  10  METHODS  14  Experiment 1: The relationship between Mazzaella cover and fucoid recruitment  14  Experiment 2: Effects ofpruning Mazzaella turfonfucoid recruitment 15 Experiment 3: Effects of Mazzaella turf on post-recruitment survivorship offlicoids  16  Experiment 4: Effects offucoids on Mazzaella comucopiae  17  RESULTS  21  The relationship between Mazzaella cover and fucoid recruitment  21  Effects ofpruning Mazzaella turf on fucoid recruitment  21  Effects of Mazzaella turf on post-recruitment survivorship offucoids  ...  22  iv Effects offucoids on Mazzaella cornucopiae  .  DISCUSSION  23 24  The complexity of algal interactions among the three dominants  24  The mechanism ofpositive interaction and neighbor distance  27  Coexistence of the major species: Biological perspectives  30  REFERENCES  33  CHAPTER 3: Preferential feeding by a littorinid: implications for its role in a high intertidal algal community  50  INTRODUCTION  50  MATERIALS AND METHODS  52  Sampling site and organisms  52  Natural densities of snails  53  Single-diet experiments  54  Multiple-diet experiments  56  Snail behaviors and tide/light effects  57  RESULTS  58  Natural densities of snails  58  Single-diet experiments  58  Multiple-diet experiments  59  Snail behaviors and tide/light effects  60  DISCUSSION  61  REFERENCES  65  CHAPTER 4: The effect of barnacles and limpets on algal succession  75  INTRODUCTION  75  METHODS  78  Experimental design  78  Data analysis  80  V  RESULTS DISCUSSION  .  .  81 83  Models and the herbivore role in algal succession  83  Direct and indirect interactions during succession  85  REFERENCES  88  CHAPTER 5: Effects of size of experimental clearings and season of clearing on algal patch recovery  100  INTRODUCTION  100  METHODS  104  Experimental design  104  Data analysis  105  RESULTS  106  Unmanipulated plots  106  Size effects of disturbance  107  Seasonal effects of disturbance  108  Interactions of the disturbance factors  109  Barnacle and limpet densities  110  DISCUSSION  110  Size effects of disturbance  110  The effect of timing of disturbance  113  Responses of barnacles and limpets to the size of disturbance  114  Responses of individual algal species to disturbance  115  REFERENCES  117  CHAPTER 6: Predictive models for community structure following environmental changes  129  THE MECHANISMS AND COMPLEXITY OF SPECIES INTERACTIONS 129  vi  PREDICTIONS OF THE EFFECTS OF ENVIRONMENTAL IMPACTS ON THE COMMUNITY STRUCTURE  133  Case 1: Absence of limpets in the system  134  Case 2: Absence of Mazzaella in the system  134  Case 3: Absence of Fucus in the system  135  Case 4: No Mazzaella when a small/medium (e.g., 25- 100 cm ) gap is 2 formed in summer  136  Case 5: No limpets when a large gap (e.g., 400 cm ) is formed in winter 2  REFERENCES  139  APPENDIX A: ANOVA tables for natural densities of snails from Chapter 3  144  APPENDIX B: Supplementary data and results from other studies in Chapter 4  145  vii LIST OF TABLES Page Table 2.1 Results of ANCOVA on fucoid recruitment in the different Mazzaella pruning plots. The ANCOVA was applied to each sampling date. The covariate is Mazzaeila percent cover in each plot (M-cover). Three treatments are ‘total pruning’, ‘partial pruning’ and ‘no pruning’ (control), see text for details 38 Table 2.2 Comparisons of longevity among the distance groups in each species and between F. distichus and P. limitata at each distance group. Means are number of months  39  Table 2.3 Results of multivariate statistics as a preplanned multiple comparison for testing three hypotheses. A Bonferroni adjusted probability value (p=O.O5 /4= 0.0 125) was used to compare each pair of treatments  40  Table 3.1 Single-diet experiments. Results of three-way ANOVA, a Snail x Algae x Time (2x3x2), with repeated measures on the Time factor  68  Table 3.2 Effects of light and tide on snail behavior. Results of a two-way ANOVA with repeated measures on both light (2) and tide (2) factors. A separate analysis was applied on each sampling date (Days 1, 4, 7) during the multiple-diet experiment 69 Table 4.1 The mean density ±SE (in 20x20 cm plot) of barnacles and limpets in each non-excluded treatment. Data are the mean of 14 sampling dates (April 1992 August 1994; December 1993 data are missing), in each sampling date the density of barnacles and limpets was averaged from the 4 replicate plots  91  Table 4.2 The mean percent cover of the five most abundant ephemeral algae in each treatment. Data are the mean of 13 sampling dates (February 1992 February 1994), in each sampling date the percent cover value of each species was averaged from the 4 replicate plots  92  -  -  Table 5.1 The effect of size of experimental clearings (5x5 and 1 Ox 10 cm) on total algal cover. A separate repeated measures analysis of variance was applied on data from each season. Large plots (20x20 cm) were not included in analysis, as discussed in text 120 Table 5.2 The effect of size and season of experimental clearings on recolonization of the three dominant algae, Fucus, Pelvetiopsis, Mazzaella, and the fugitive algal group. Repeated measures ANOVA was applied to 16 months of data. Large plots (20x20 cm) were not included in analysis as discussed in text. All data were arcsine transformed prior to analysis 121 Table A.1 Summaries of ANOVAs on the natural density of snails at Prasiola Point. A. Densities of snails in monospecific patches of the three dominant algae. B. Densities of snails in general habitats (3 transect lines) 144  viii LIST OF FIGURES Page Figure 1.1 Location of study sites. A  =  Nudibranch Point. B  =  Prasiola Point  9  Figure 2.1 Correlations of M. cornucopiae percent cover and number of fucoid recruits. 41 Figure 2.2 The effect of Mazzaelia turf on fucoids after recruitment stage. Curves are the weighted average of plots  42  Figure 2.3 Relationship between the size class of Mazzaella blade and wet weight (g). Mazzaella blade height was divided into six 5 mm size classes (SC). SC1 = 1-5 mm, SC2 = 6-10 mm, SC3 = 11-15 mm, SC4 = 16-20 mm, SC5 = 21-25 mm, SC6 >26 mm 43 Figure 2.4 The effect of pruning Mazzaella blades on Mazzaella percent cover in the same experimental plots (lOxlO cm)  44  Figure 2.5 The effect of Mazzaella pruning on fucoid recruitment. A. Number of fucoid recruits actually occurring in the plots. B. Number of fucoid recruits adjusted by Mazzaella percent cover (the covariate) in each plot at each sampling date. These values were used in the ANCOVA. Data are means +SE of 9 replicates. PT for Pelvetiopsis in the total pruning plots, PP for Pelvetiopsis in the partial pruning plots, PC for Pelvetiopsis in the control plots; FT, FP and FC for Fucus in the respective plots 45 Figure 2.6 Changes in Mazaella biomass (g) under each treatment after 2 months. CTL for control, CLE for the clear artificial plant, DRK for the dark artificial plant, FUC for Fucus, and PEL for Pelvetiopsis. Data are means ÷SE (n=5 for CLE, FUC and PEL, n=6 for DRK, n=7 for CTh)  46  Figure 2.7 The effect of fucoid shading on Mazzaella size class. Data are means +SE (n=5 for clear and n=6 for dark artificial plant)  47  Figure 2.8 Diagram of potential factors affecting fucoid mortality. See DISCUSSION for further explanation 48 Figure 2.9 Fucus survivorship curves in relation to the changing effects of competition and physical stress with distance from the edge of a Mazzaella turf. The shaded area represents the probability of Fucus survival 49 Figure 3.1 Natural densities of snails at Prasiola Pt. Data are means +SE of 5 monospecific patches of each alga and of 18 randomly selected plots from transect lines in the general habitat (3 transects x 6 points). GEN for the general habitat, MAZ for M. cornucopiae, FUC for F. distichus, PEL for P. limitata. 70 ...  Figure 3.2 Single-diet experiments. 3.0±0.02 g of thallus was used per cage. Positive values indicate weight gain. Negative values indicate weight loss. Data are means +SE of five treatments and three controls. MAZ, M. cornucopiae, FUC, F. distichus, PEL, P. liinitata 71 Figure 3.3 Results of the multiple-diet experiment. The comparisons of weight loss (g) between treatments and controls and among algal species. 1±0.02 g of each algal  ix species were placed making —3.0 g (total) of food per cage. Data are means +SE of fourteen treatments and seven controls. MAZ, M. cornucopiae, FIJC, F. distichus, PEL, P. limitata 72 Figure 3.4 Behavioral choice (attractiveness) of Littorina sp. towards the three algal species. Densities of snails attached to the thalli of each algal species 15 mm after the multiple-diet experiment started. Data are means +SE of sixteen replicates. See Fig. 3.2 for caption comments  73  Figure 3.5 Tidal and light effects on snail movement towards food during the multiplediet experiment. The data shown are the number of snails attached to M. cornucopiae for every tidal and light condition. Tide/light combinations are presented in the same order as simulated in the lab. Data are means +SE of sixteen replicates 74 Figure 4.1 Effects of barnacle and limpet removal on each algal group. Data are means ± SE of four replicates. +B+L indicates both barnacle and limpet present, -B+L barnacle absent, limpet present, +B-L barnacle present, limpet absent, and -B-L both barnacle and limpet absent 93 Figure 4.2 Effects of barnacles and limpets on algal species interactions in each treatment. EPH for ephemeral algae, MAZ for Mazzaella cornucopiae, FUC for Fucus distichus, PEL for Pelvetiopsis li,nitata. Data are means ± SE of four replicates 95 Figure 4.3 The abundance of barnacles and limpets in the experimental plots (20x20 cm). Data are the mean densities ± SE per plot (n=4)  97  Figure 4.4 A summary of barnacle and limpet effects on algal succession. Dominant algae were shown in each experimental condition with their relative abundance.  98  Figure 4.5 The summary of direct and indirect interactions shown between barnacles, limpets, fucoids and ephemeral algae in this study  99  Figure 5.1 Seasonal changes of algal percent cover and the number of fucoid recruits in unmanipulated plots (20x20 cm) at Prasiola Point for two years. Data are the mean ±SE of four replicates. EPH for ephemeral algae, MAZ for Mazzaella cornucopiae, FUC for Fucus distichus, PEL for Pelvetiopsis liinitata 123 Figure 5.2 Recolonization of total algae in three different sizes of clearing. Large clearings = 20x20 cm, medium clearings = lOxlO cm, small clearings = 5x5 cm. Plots cleared in the summer (August 1991) were only compared for the size effect on patch recovery. Data are the mean ±SE of fifteen replicates for small and medium plots and of four replicates for large plots 124 Figure 5.3 Patch recovery of each dominant algae, Mazzaelia, Fucus, Pelvetiopsis, and ephemeral algae in three different sizes of clearing. Data shown are from the summer-cleared (August 1991) plots. Data are the mean ±SE of fifteen replicates for small and medium plots and of four replicates for large plots 125 Figure 5.4 The effect of season of clearing on algal recovery of each species in small plots (5x5 cm). The time for each seasonal clearing is August 1991 (for summer), October 1991 (for fall), February 1992 (for winter) and April 1992 (for spring). Data are the mean ±SE of fifteen replicates 126  x Figure 5.5 The effect of season of clearing on algal recovery of each species in medium plots (lOxlO cm). See Fig. 5.4 for captions and the number of replicates 127 Figure 5.6 Barnacle and limpet densities in the three sizes of patch at three sampling dates (4, 10, 16 months after clearing). Data are the mean ±SE of fifteen replicates for small and medium plots and of four replicates for large plots. When results of ANOVA on three sizes were significant, results of Tukey’s HSD tests were labeled (e.g., LM indicates that density in small patch is significantly greater than that of either large or medium patch). N/S for p>0.05. 128 Figure 6.1 The summary of interactions among the major component organisms shown in this study. PEL and FUC in the upper diagram indicate Pelvetiopsis and Fucus at the recruitment stage, while those in the lower panels indicate the fucoids after recruitment stage. Small letters are the abbreviations for types of interaction: (co) for competition, (fa) for facilitation, (gr) for grazing, (ov) for overgrowth, (pr) for preemption, (pt) for protection, (re) for refuge, (sh) for shading  141  Figure 6.2 A predictive model for community structure, in the absence of Mazzaella, when a smalllmedium gap (e.g., 5x5 to lOxlO cm) is created in summer. The size of the circles represents the relative abundance of component species  142  Figure 6.3 A predictive model for the community structure, in the absence of limpets, when a large gap (e.g., 20x20 cm) is created in winter. The size of the circles represents the relative abundance of component species 143 Figure B.1 Farrell?s model for the effect of consumers on the rate of succession (i?’ consumers increase the rate of succession, 0- no effect, consumers decrease the rate of succession) cited from Farrell (1991) 145 -  , -  Figure B.2 Dynamics of ephemeral algae in limpet-excluded plots (-L), with (+B) and without (-B) barnacles 146 Figure B.3 Direct and indirect interactions shown in other studies  147  xi ACKNOWLEDGMENTS I very much thank my teacher, friend and supervisor, Dr. Robert E. DeWreede, for his constructive suggestions, encouragement and support throughout the past five years. This study is partially funded by his grant from the Natural Science and Engineering Research Council of Canada (grant # 5-89872). My thanks go also to my committee members, Drs. Roy Turkington and Paul G. Harrison, for their critical suggestions and editorial comments on this thesis. I am grateful to Carol Ann Borden for the flexibility that she showed when I was juggling teaching assistant responsibilities with the field trip to Bamfield in every two months. My field works could not have been finished without the help of the following people: Ok Hyun Ahn, Kim Ailcock, Put Ang, Arnold Cheung, Angela Crampton, John Danko, Kristen Drewes, Glenda Eberle, Nick Grabovac, AnnMarie Huang, Andrea Park, Brent Philips, Ricardo Scrosati, Frank Shaughnessy, Laura Wong. I much appreciate their time and effort in the field, particularly in the often rainy, windy and freezing cold weather. I thank my colleague in the Dr. DeWreede’s lab, Put Mg, Frank Shaughnessy, Brent Philips, Ricardo Scrosati, Kristen Drewes, Russell Markel, Tania Thenu and Andrea Sussman, for useful discussion, help and friendship. Staff at the Bamfield Marine Station kindly allowed me to use some of their facilities. For technical help, I am indebted to Drs. Han Ju Eom and Robert Schutz, Department of Human Kinetics at UBC, for spending time helping me to analyze my data. My thanks also go to Drs. Elizabeth Boulding and Thomas Carefoot for their help on identifying snails, limpets and barnacles. I am most grateful to my wife, Meeok Kim, who has shown her continuous patience and support in many ways and for many years. Her precious and painstaking support facilitated this study and exempted me from many house-keeping responsibilities. I would like to thank my parents, Hong Ki and Sun Hee Kim, for their understanding, encouragement, patience and support during my study at UBC as well as during the entire period of my graduate study in North America. I also appreciate my mother-in-law for her earnest prayer for me and my family during the study. Finally, I thank to my lovely Mm Jee and Dong Hyun for their understanding of this student daddy who could not spend much time to play with them the last several months.  1 CHAPTER 1  THESIS INTRODUCTION  General Introduction In 1966, Elton suggested an explicit aim for community ecology: to discover and measure the main dynamic relations between all organisms living on an area over some period of time. His view on community ecology was supported by other ecologists (e.g., Dayton 1971, Connell 1972, Paine 1980) as an appropriate protocol guiding community ecologists. Later, Dayton (1971) amplified Elton’s view by adding disturbance as a mechanism forming open space which is a common limiting resource in marine benthic habitats. Dayton carried out a pioneer study of the marine rocky intertidal community which incorporated dynamic interactions among the component populations, trophic structure, and the effects of major community disturbances. After Dayton’s work, many marine community ecologists have continued to focus their research on factors that influence community structure and dynamics. The effect of major biotic and abiotic factors on community structure and dynamics have been extensively reviewed; these include studies on competition (Connell 1983, Schoener 1983, Denley and Dayton 1985, Olson and Lubchenco 1990, Paine 1990), herbivory or predation (Lubchenco and Gaines 1981, Hawkins and Hartnoll 1983, Estes and Steinberg 1988), and disturbance (Connell and Keough 1985, Sousa 1985). These studies have contributed to the establishment and development of theories and models for community organization. Because the process of organization and dynamics of any community is not a product of a single mechanism but of a combination of various biotic and abiotic factors which often interact with each other, the most convincing demonstration of the factors influencing community structure requires a comprehensive approach, one which examines all major interactions and  2 mechanisms that occur in the community under investigation. Few such studies have been done except for some which focussed on community succession and its mechanisms (Menge 1976, Lubchenco and Menge 1978, Sousa 1979, Turner 1983, Farrell 1991). My research was designed to implement such a comprehensive study. The target community which I chose to study is a high intertidal algal community on a rocky shore. This provides some unique benefits for direct measurement of various biotic interactions. Firstly, because many of the common species are sessile or slow moving and not hidden in the substratum, their abundance and other characteristics can be measured readily. Secondly, most populations are amenable to experimental manipulation in the field, so that replicate treatments and undisturbed controls can be set up at the same time. Thirdly, the upper zones are easily accessible and one can collect data on a very small spatial scale with high accuracy, which is often impossible in most mid and low intertidal and subtidal zones. Factors influencing intertidal algal communities can be divided into two groups: biological factors (e.g., competition and herbivory) and physical factors (e.g., disturbance and physical stress). The importance of biological factors in controlling the abundance and distribution of algal species is generally accepted to decrease with increasing tidal height on rocky shores and physical factors are more important in structuring upper intertidal communities (Castenholz 1961, Connell 1972, 1975, Chapman 1973, Menge 1978, Underwood 1980). However, this generalization has been questioned (Dayton 1975, Underwood and Denley 1984) and needs further investigation with more experimental tests especially for its application to upper intertidal zones. Some ecologists have developed models to explain the relative importance of factors producing structure and dynamics by focusing on the interplay between physical factors and biological factors (Menge and Sutherland 1976, 1987, Connell and Slatyer 1977). Specifically, Menge and Sutherland’s model addresses limits to species distributions correlated with gradients in environmental harshness. Such correlations  3 may arise because abiotic factors impose direct physiological limits on species, or gradients in environmental stress may indirectly influence the abundance of prey (animals or plants) by altering the impact of their consumers (carnivores or herbivores). However, such models do not reflect the likelihood that harsh abiotic factors alter the patterns of species interactions and consequently influence species distribution. Palumbi (1985) showed an explicit example of such an interaction in which two species were competitors in areas of low desiccation, but the interaction was a commensal one under more desiccating conditions. So in one case the interaction would be called inhibition, while in the other it would be termed facilitation. Bertness and Callaway (1994) recently suggested that a positive interaction, facilitation, may be a predictably important force in certain environments, and proposed that it should be incorporated into future community paradigms. Therefore, the above models still need more experimental data from variable habitats to either reject or support them. The main objectives of this thesis are: 1) to investigate interactions and their mechanisms of action in a target community, 2) to provide empirical evidence from the study of a high intertidal habitat so that some current views of the mechanisms and models of community structure can be tested, and 3) based on the results, to predict community-level responses to environmental impacts which may lead to reorganization of the existing community structure.  4 The study site and community  The study areas are the rocky upper intertidal communities at Prasiola Point and Nudibranch Point located in Barkley Sound on the west coast of Vancouver Island, British Columbia, Canada (48°49’ N, 125°10’ W; Fig. 1.1). The two sites are about 0.4 km apart and are both exposed to intermediate waves from the north-west and stronger waves from the west; both sites have a similar flora and fauna. The experimental plots were located on rocky substratum with a gentle slope; the zone ranged from 3.4 m to 5.3 m above Lowest Normal Tide (LNT: Canadian Chart Datum). The algal community was dominated by three perennial macrophytes, Mazzaella cornucopiae (Post. & Rupr.) Hommersand et al. (Gigartinales, Rhodophyta; previously known as the genus Iridaea), Pelvetiopsis limitata Gard. and Fucus distichus L. (both Fucales, Phaeophyta). These species grow both in well-mixed and mono-specific stands depending on wave-exposure. Other stands of M. cornucopiae occurred in extremely wave-exposed sites characterized by the alga Lessoniopsis littoralis (Tilden) Reinke and the invertebrate Pollicipes polymerus Sowerby, whereas fucoid stands were also found in relatively more wave sheltered sites. However, sites for all experimental plots and sampling were set up in areas where mixed stands of the three macroalgae predominated. Most M. cornucopiae was 2-3.5 cm in length and grew as a dense turf on the rocky substratum. The two morphologically similar fucoids are erect and usually 4-7 cm in length at maturity. There are also some ephemeral algae in the community, such as Cladophora columbiana Coil., Enteromorpha spp., Mastocarpus papillatus (Agardh) Kutzin, Porphyra spp., Scytosiphon dotyi Wynne, benthic diatoms, Endocladia muricata (Post. & Rupr.) 3. Ag., Analipusjaponicus (Harvey) Wynne, Urospora sp. and crustose algae (‘Petroceli& phase of Mastocarpus, Hiidenbrandia sp. and Ralfsia sp.). Barnacles are the most abundant invertebrates in the habitat. The dominant species was Balanus glandula Darwin, whereas B. cariosus Pallas and Chthamalus dalli  5 Pilsbury occurred less frequently. The most abundant herbivores were snails (mostly Littorina sp., see Boulding and VanAlstyne 1993 for further description) which occurred in a seasonal pattern with a peak abundance in summer (Chapter 3). Littorina scutulata Gould and L. plena Gould often occurred in my study sites but their frequency was very low (<3% of total snails). Other common herbivores were limpets, primarily Lottia digitalis Rathke and a few L. pelta Rathke. There was an occasional occurrence of mussels, primarily Mytilus trosselis L.  Note: In this thesis, I discuss the ecology of a common high intertidal species of algae, Fucus distichus L. A recent treatment has revised the name to F. gardneri Silva, but many questions remain on the appropriate nomenclature. I have chosen to retain the first name as this is one by which it is commonly discussed in the ecological literature.  6 Organization of the thesis  Non-manipulative field monitoring, manipulative field experiments and laboratory experiments were conducted to address the stated objectives. This thesis consists of six chapters, an introductory chapter, four research chapters, and the concluding chapter. In Chapter 2, I deal with patterns of interactions among the dominant macroalgae and discuss mechanisms of both algal competition (between morphologically distinct groups) and positive interactions between them under harsh physical conditions. Coexistence of the three dominant algae is discussed in the light of these algal interactions. Chapter 3 examines the effect of herbivorous snails on the three dominant algae. The abundance and distribution of the snails were monitored in the field, and their feeding preference on the three algal species was tested in the laboratory. Chapter 4 addresses the role of barnacles and limpets in this algal community. Results of three years of monitoring evaluate their role in algal recolonization and succession. Chapter 5 assesses the response of algae to physical disturbance. Two aspects of disturbance (i.e., size and season) are tested for their effects on the algal patch recovery patterns. Chapter 6 synthesizes events in this intertidal system using information from the previous chapters. Some predictions on the effect of altering presence or density of community biota are discussed, incorporating the insights gained from this research.  7 REFERENCES Bertness, M. D., Callaway, R. (1994). Positive interactions in communities. Tree 9: 191193. Boulding, E. G., Van Aistyne, K. L. (1993). Mechanisms of differential survival and growth of two species of Littorina on wave-exposed and on protected shores. J. Exp. Mar. Biol. Ecol. 169: 139-166. Castenholz, R. W. (1961). The effect of grazing on marine littoral diatom populations. Ecology 42: 783-794. Chapman, A. R. 0. (1973). A critique of prevailing attitudes on the control of seaweed zonation on the sea shore. Bot. Mar. 16: 80-82. Connell, J. H. (1972). Community interactions on marine rocky intertidal shores. Ann. Rev. Ecol. Syst. 3: 169-192. Connell, J. H. (1975). Some mechanisms producing structure in natural communities: a model and evidence from field experiments. In: Cody, M. L., Daimond, J. M. (eds.) Ecology and evolution of communities. Belknap, Cambridge. pp. 460-490. Connell, J. H. (1983). On the prevalence and relative importance of interspecific competition: evidence from field experiments. Am. Nat. 122: 661-696. Connell, J. H., Slatyer, R. 0. (1977). Mechanisms of succession in natural communities and their role in community stability and organization. Am. Nat. 111: 1119-1144. Connell, J. H., Keough, M. J. (1985). Disturbance and patch dynamics of subtidal marine animals on hard substrata. In: Pickett, S. T. A., White, P. S. (eds.) The ecology of natural disturbance and patch dynamics. Academic Press. New York. pp. 125-15 1. Dayton, P. K. (1971). Competition, disturbance and community organization: the provision and subsequent utilization of space in a rocky intertidal community. Ecol. Monogr. 41: 35 1-389. Dayton, P. K. (1975). Experimental evaluation of ecological dominance in a rocky intertidal algal community. Ecol. Monogr. 45: 137-159. Denley, E. J., Dayton, P. K. (1985). Competition among macroalgae. In: Littler, M. M., Littler, D. S. (eds.) Ecological field methods: macroalgae. Handbook of phycological methods. Cambridge University Press. New York. pp. 511-530. Elton, C. S. (1966). The pattern of animal communities. Wiley, New York. pp. 432. Estes, J. A., Steinberg, P. D. (1988). Predation, herbivory, and kelp evolution. Paleobiology 14: 19-36. Farrell, T. M. (1991). Models and mechanisms of succession: an example from a rocky intertidal community. Ecol. Monogr. 61: 95-113. Hawkins, S. J., Hartnoll, R. 0. (1983). Grazing of intertidal algae by marine invertebrates. Oceanogr. Mar. Biol. Ann. Rev. 21: 195-282.  8 Lubchenco, J., Menge, B. A. (1978). Community development and persistence in a low rocky intertidal zone. Ecol. Monogr. 48: 67-94. Lubchenco, J., Gaines, S. D. (1981). A unified approach to marine plant-herbivore interactions. I. Populations and communities. Ann. Rev. Ecol. Syst. 12: 405-437. Menge, B. A. (1976). Organization of the New England rocky intertidal community: role of predation, competition, and environmental heterogeniety. Ecol. Monogr. 46: 355-393. Menge, B. (1978). Predation intensity in a rocky intertidal community. Oecologia 34: 116. Menge, B. A., Sutherland, J. P. (1976). Species diversity gradients: synthesis of the role of predation, competition, and temporal heterogeneity. Am. Nat. 110: 35 1-369. Menge, B. A., Sutherland, J. P. (1987). Community regulation: variation in disturbance, competition, and predation in relation to environmental stress and recruitment. Am. Nat. 130: 730-757. Olson, A. M., Lubchenco, J. (1990). Competition in seaweeds: linking plant traits to competitive outcomes. J. Phycol. 26: 1-6. Paine, R. T. (1980). Food web: linkage, interaction strength and community infrastructure. J. Anim. Ecol. 49: 667-685. Paine, R. T. (1990). Benthic macroalgal competition: complications and consequences. J. Phycol. 26: 12-17. Palumbi, S. R. (1985). Spatial variation in an alga-sponge interaction and the evolution of ecological interactions. Am. Nat. 126: 267-274. Schoener, T. W. (1983). Field experiments on interspecific competition. Am. Nat. 122: 240-285. Sousa, W. P. (1979). Experimental investigations of disturbance and ecological succession in a rocky intertidal algal community. Ecol. Monogr. 49: 227-254. Sousa, W. P. (1985). Disturbance and patch dynamics on rocky intertidal shores. In: Pickett, S. T. A., White, P. S. (eds.) The ecology of natural disturbance and patch dynamics. Academic Press. Orlando. pp. 101-124. Turner, T. (1983). Complexity of early and middle successional stages in a rocky intertidal surfgrass community. Oecologia 60: 56-65. Underwood, A. J. (1980). The effect of grazing by gastropods and physical factors on the upper limit of distribution of intertidal macroalgae. Oecologia 46: 201-213. Underwood, A. J,, Denley, E. J. (1984). Paradigms, explanations, and generalizations in models for the structure of intertidal communities on rocky shores. In: Strong, D. R., Simberloff, D., Abele, L. G., Thistle, A. B. (eds.) Ecological communities: Conceptual issues and the evidence. Princeton University Press. New Jersey. pp. 151-180.  9  I  I  Alaska W7’.  ?j(4  550  -  British Columbia Pacific Ocean  10’W 0 125 I.  500  v.I.  -  —  B -  Barkley Sound  1 .:.:BaifieId WA.  1  A  ‘tb.)  OR.  1400  1350  130°  I  I  I  Fig. 1.1 Location of study sites. A  =  Nudibranch Point. B  1 25 I  =  Prasiola Point.  CA.  —  10 CHAPTER 2  PATTERNS OF INTERACTIONS AMONG NEIGHBOR SPECIES IN A HIGH INTERTIDAL ALGAL COMMUNITY  INTRODUCTION  Species interactions are fundamental in understanding community structure and dynamics. Marine ecologists working on the direct and indirect interactions between marine benthic organisms have reported various such interactions. The usefulness of experimental tests of hypotheses about the roles of species interactions in determining the distribution and abundance of marine organisms has been reviewed (Connell 1974, 1983, Paine 1977, 1990, Schoener 1983, Denley and Dayton 1985, Olson and Lubchenco 1990). Despite the criticism of the nature and design of some of the experiments (Dayton and Oliver 1980, Underwood and Denley 1984), such experimentation is a powerful approach for the development of realistic models of the structure and dynamics of natural communities (Connell 1974, 1983, Dayton and Oliver 1980, Denley and Dayton 1985). Typically ecologists have focused on negative interactions (e.g., competition) while positive interactions (e.g., facilitation or protection) have received less attention or are ignored in some current models of community organization (Connell and Slatyer 1977, Menge and Sutherland 1987, Tilman 1994; but see Bertness and Callaway 1994). This trend is not unique to marine benthic algal ecology. Thus, current reviews on competition of marine benthic macroalgae indicate the necessity of more experimental data on the variability of mechanisms of competition with changing life-history stages of competitors, variable environmental conditions, and on positive interactions, to improve our understanding of the role of species interactions in structuring algal assemblages (Paine 1990, Olson and Lubchenco 1990).  11 The mechanisms of interaction between competitors have traditionally been recognized as consisting of two basic kinds, exploitative competition and interference competition, although these can be further subdivided into as many as six kinds (Schoener 1983). Exploitative competition (e.g., consumption, preemption) is an indirect interaction between competitors, and the interaction is always mediated by the resource in short supply (Keddy 1989). Three important resources for macroalgae are light, space and nutrients (Denley and Dayton 1985). In contrast, interference competition (e.g., whiplash, overgrowth) occurs when one individual directly affects another. The interaction causes direct mortality even when the resource might not be in short supply. However, there are some situations in which both types of competition may occur simultaneously between the same pair of competitors (reviewed by Denley and Dayton 1985). The outcome of competitive interactions may depend on such plant traits as morphological type and life history stage, and also on the mechanism of competition (Olson and Lubchenco 1990). For example, if the interaction is preemption of space, then larger size, spreading habit, and the ability to perennate may afford competitive dominance. If interference is involved, rapid lateral growth, ability to raise the growing edge off the substratum and production of toxins are traits that may affect the outcome of an interaction (Olson and Lubchenco 1990). However, the above relationship between traits and competitive dominance may not be straightforward, and sometimes the outcome of competition may be strongly affected by other biological interactions (e.g., herbivory) or physical interactions (Connell 1975, Lubchenco and Gaines 1981, Hawkins and Hartnoll 1983, Paine 1990). Many authors (Connell 1975, Underwood and Denley 1984, Denley and Dayton 1985, Olson and Lubchenco 1990) have proposed difficulties or cautions in conducting experiments on algal competition. First is the problem of detecting competition in natural habitats. Results of field manipulation experiments on the outcome of competition are often misleading because other biotic or abiotic factors may interfere with or simulate competition  12 (often called “apparent competition” by Holt 1977 and Connell 1990). The most common example of the latter in algal competition is caused by herbivores (Lubchenco and Gaines 1981, Hawkins and Hartnoll 1983, Paine 1990). Second, it is often difficult to identify a single mechanism of competition. For example, exploitative competition (e.g., shading) can be accompanied by inteiference (e.g., allelopathic effect). Third, measuring the outcome of interspecific competition may not be sufficient for understanding coexistence of competing species where intraspecific competition is strong (Connell 1983). Therefore, the knowledge of, or a simultaneous experiment for, the intensity of intraspecific competition is recommended (Connell 1983, Denley and Dayton 1985, Paine 1990). Fourth, the outcome of competition may be changed depending on the developmental stages of competing plants (e.g., germling vs. adult or adult vs. adult) (Olson and Lubchenco 1990). Similarly, the effect of one species on settlement and recruitment of another may not be the same as the effect on the subsequent survival and growth of the species. Fifthly, when rank order of competitive superiority is involved, reciprocal tests on both members in a pair are necessary to deal with the problem that asymmetrical competition resulting in the elimination of inferior species is not always the case. Reversals of competitive rank may be common in natural communities (Connell 1983, Schoener 1983). To obtain such evidence requires at least two experiments (on each member of a pair), which is rarely done (reviewed by Connell 1983, Denley and Dayton 1985). Ideally, experimenters should consider as many of the problems listed above as possible to produce a clear and unambiguous demonstration of interspecific competition. There is no work in algal competition that addresses all the problems listed above; it may in fact be impossible to conduct such a field experiment. It is important, however, that field studies on competition be rigorous and comprehensive, combining manipulative experiments of potentially important factors in communities with a suitable knowledge of the biology of the algae under investigation. In this study, among the five problems listed above most of them (except the third) were taken into account for designing a series of field experiments.  13 Although positive interactions have been investigated less frequently, in marine benthic organisms, than competition, its possibility and relevance have been discussed theoretically by many ecologists (Connell and Slatyer 1977, Vandermeer 1980, Connell 1983, Dethier and Duggins 1984, Sousa 1984, Bertness and Callaway 1994). Some positive interactions occurred as a consequence of direct beneficial influences of one species to the other (e.g., Dayton 1975, Brawley and Johnson 1993), others as a consequence of negative interactions acting indirectly through other species, possibly called indirect commensalism, in a same trophic level (e.g., Duggins 1981, Kastendiek 1982) or in a different trophic level (e.g., Dethier and Duggins 1984, Hay 1986). Particularly in a physically stressful habitat (e.g., upper intertidal zones), positive interactions (e.g., neighbor habitat-amelioration; Bertness and Callaway 1994) rather than competition could be an important factor which allows coexistence of various species in the community. There is no study examining the variation of positive interactions along with microhabitat gradients (e.g., neighbor distance in the case of neighbor habitat-amelioration) in marine intertidal zones. The concept of neighbor distance has been studied particularly in terrestrial plant communities (Fawcett 1964, Mack and Harper 1977, Turkington and Harper 1979, Lindquist et al. 1994). In marine habitats, the importance of biological interactions in small patches has been invoked in a few studies (Connell 1972, Paine 1990). Connell (1983) stated that for organisms that compete for space, and in particular clonal plants and attached colonial animals, competition occurs mainly between neighbors. In this context, neighbor distance was measured as a parameter to test if the outcome of algal interactions was affected by the distance between two interacting species. This study investigates how algal interaction patterns change due to the developmental stages of the competitors and on the neighbor distance between morphologically distinct groups. Because of the morphological similarity of the two fucoids, Fucus distichus and Pelvetiopsis limitata, the study focused on interactions between Mazzaella cornucopiae and the two fucoids. Both manipulative and non-  14 manipulative field experiments were designed to address the following questions and hypotheses. First, does the turf-forming red alga, M. cornucopiae, affect the zygote settlement or germination of the fucoids? Second, does M. cornucopiae affect the postrecruitment survivorship of the fucoids? Third, do the taller erect fucoids affect the growth of the shorter turf-forming M. cornucopiae by shading, whiplash or scour, or by chemical content (e.g., phenolic compounds) by which the fucoids allelopathically suppress M. cornucopiae? Shading and scouring are commonly reported mechanisms of competition in both subtidal and intertidal habitats (reviewed by Denley and Dayton 1985, Olson and Lubchenco 1990, Paine 1990). A few studies have reported that chemical content of certain algae (including some fucoids) affected other algae by allelopathic influences (McLachlan and Craigie 1966, Carison and Carlson 1984). Although Hawkins and Hartnoll (1983) and Paine (1990) suggested the primary importance of the consumer’s role in algal interactions, in this chapter I generally avoid mentioning other biological factors (e.g., herbivory), because the next two chapters (3 and 4) deal with this topic. I wifi present a synthetic view on the patterns of algal interaction including other biological and physical factors in Chapter 6.  METHODS  Experiment 1 :  The relationship between Mazzaella cover and fucoid recruitment  In August of 1991, sixteen permanent plots (20x20 cm) were randomly established on mixed stands of the three dominant algal species at Prasiola Point. Concrete nails were hammered to mark at least 3 corners of each plot. The nails were sometimes supported by placing cement around them, so that most nails lasted for the experimental period. These permanent plots were monitored at two-month intervals using a 20x20 cm quadrat divided  15 by monofilament line into 400 squares, each 1 cm . Percent cover and exact location of 2 Mazzaeiia cornucopiae turf was estimated by mapping the squares that had >50% cover of blades; in this way, the squares with <50% were dropped so estimation will be balanced for overestimation. The position of fucoids was also mapped using this technique and the size class and reproductive condition was also noted. At each sampling date it was possible to recognize newly recruited individual fucoids and thus the method enabled me to follow individual fucoid thalli from birth to death. The sixteen plots initially, and continuously, received four treatments Mazzae11a-removed, Fucus-removed, Pelvetiopsis-removed and ,  control). Because there were no significant treatment effects between the control and either Fucus-removed or Pelvetiopsis-removed plots (J. H. Kim, data not shown in the thesis), Fucus recruits used for analysis were collected from the control and Peivetiopsis-removed plots, and Pelvetiopsis recruits used were from the control and Fucus-removed plots. Monitoring was done until October of 1993. Statistical analyses were done using SYSTAT, version 5.2.1 for Macintosh, (Wilkinson et al. 1992). The Pearson Product Moment Correlation between number of recruits (log-transformed) and percent cover of Mazaelia was calculated for both fucoids. The significance of the correlation coefficients was assessed using Fisher and Yates’ tables (1963).  Experiment 2 : Effects of pruning Mazzaella turf on fucoid recruitment  Correlations found in Experiment 1 do not necessarily indicate a cause and effect relationship, so I conducted a manipulative field experiment to test the hypothesis that removing Mazzaella turf enhances recruitment of Fucus and Pelvetiopsis. The experimental units consisted of three levels of Mazzaella pruning; 1) “total pruning” cutting whole -  Mazzae11a blades leaving only their holdfasts, 2) “partial pruning” cutting the upper part of -  the blades and leaving 1 cm of the bottom, and 3) “control” no pruning. The different -  16 levels of Mazaella pruning were done to test if the dense turf physically blocked fucoid propagule arrival. The different heights of Mazzaeila blades were used to simulate different levels of ‘blocking’ recruits and ‘shading’ growing recruits. At Prasiola Point, 8 permanent transect lines (about 1 m each) were marked by hammering concrete nails at both ends of each line. Using 10 cm intervals on the transect line, I numbered all potential plots for treatments on both sides of the transect line. Treatments were assigned by randomly choosing three plots (lOx 10 cm) per line. The position of individual fucoid plants was mapped using a lOx 10 cm quadrat with subdividing lines (comprising 100 lxi cm subplots); this enabled me to obtain both the number of fucoid recruits and percent cover of M. cornucopiae in each plot on every sampling date. The experiment was initiated in June, 1993. Data were collected in October, 1993, and February and April, 1994. I found a correlation between Mazzaella percent cover and fucoid recruitment (Fig. 2.1), so I first examined the relationship between the pruning treatment and Mazzaella cover as independent variables. The first step in this analysis was a test for homogeneity of slopes, which can be determined by examining the interaction term of a regression, using the pruning treatment and percent cover as independent variables and the parameter values as dependent variables. In all 6 cases (3 sampling dates and 2 species), the probability value for the interaction term was not significant (p>.0,05) so I then conducted ANCOVA’s with the pruning treatment as an independent variable and Mazzaella cover as a covariate. Analyses were done using the BMDP2V statistical package.  Experiment 3 : Effects of Mazzaella turf on post-recruitment survivorship of fucoids  While Experiment I was designed to test the effect of Mazzaella cornucopiae on the recruitment stage of both fucoids, this experiment tests its effects on post-recruitment survivorship of the fucoids. Non-manipulative field monitoring for the sixteen permanent  17 plots (the same plots used in Exp. 1) at Prasiola Point were done from August, 1991 to October, 1993. Using the quadrat with subdividing lines described previously, the position of M. cornucopiae turf and individual fucoid plants in each plot were tape-recorded in the field and then mapped on a sheet in the laboratory. The life span of individual fucoids was obtained by following individual thalli from recruitment to disappearance. Individuals recruited or found after August, 1992, were not included in the analysis because their complete life span could not be followed. Using the map, the distance between each fucoid thallus and the nearest M. cornucopiae was measured. However, because the edge of the Mazzaella turf changed over time due to vegetative growth or decline, the distance for each fucoid from turf was not always the same in bimonthly samples. Therefore, I calculated the mean distance for each fucoid for its life span from distance values measured bimonthly and used these values for analysis. The life span and the mean distance for each fucoid were plotted to investigate their relationship. For data analyses, I divided the distance between the Mazzaeila turf and fucoid recruits into three groups based on the pattern of Fucus life span along with the neighbor distance (Fig. 2.2). Three distance groups include ‘CONTACT’ for those growing within the turf (or 0 cm from turf), ‘CLOSE’ for those growing between 0.3 and 0.7 cm from turf, and ‘FAR’ for those growing >2 cm from turf One-way ANOVA was used to test the difference among the distance groups for each fucoid species. Tukey’s HSD test was also used when ANOVAs were significant. Student’s t-test was used to compare the life span (months) of Fucus distichus and Pelvetiopsis limitata in the same distance group.  Experiment 4 : Effects of fucoids on Mazzaella corn ucopiae  A manipulative field experiment was initiated at Nudibranch Point in June, 1994. The experimental design included five treatments using both live fucoid thalli and artificial plants which were attached to a mesh strip and set up on Mazzaella turf. The five  18 treatments and their characteristics are as follows: 1) live Fucus thalli which tests for canopy (shading), scouring (disturbance) and phenolic (chemical) effects, 2) live Peivetiopsis thalli testing for the same characteristics as live Fucus, 3) ‘clear’ artificial plants which tests only for scouring, but not for canopy nor phenolic effects, 4) ‘dark’ artificial plants which test for canopy and scouring, but not for phenolics, and 5) a control with neither live nor artificial thalli. Adult thalli (about 7 cm in length) of F. distichus and P. li;nitata were haphazardly collected (with holdfasts intact) from within the normal vertical range at Nudibranch Point, and were transferred to running seawater tanks at the Bamfield Marine Station. Then, thalli with similar weight (20.5±1.3 g blotted wet weight) were selected, although some trimming was sometimes needed to equalize thallus weight for each experimental unit. Holdfasts and lower parts (—1 cm) of each thallus were placed into twists of 9.4 mm polypropylene rope (5 cm in length) which had both ends melted to prevent untwisting. Vinyl covered wire (1.1 mm in diam.) was tightly wound around the rope to secure the thalli. For the artificial plants, a transparent, flexible plastic sheet (0.55 mm in thickness) was cut to make one type of treatment, ‘clear’, and an opaque plastic sheet (0.65 mm in thickness) was cut to make the other ‘dark’. Artificial plants (13.5 cm long, 5.7 cm wide for ‘blade’ area, 1.0 cm wide for ‘stipe’) were made by attaching two extra layers of sheet (one of 5.0 cm and the other 8.0 cm long) to the lower part of the artificial plants and, to attach these extra layers, the margin of each extra layer was sewed with thin monofilament line. Therefore, each artificial plant consisted of a three-layered stipe, a two-layered lower blade and one-layered upper blade. This construction provided sufficient strength under forceful wave action and mimicked the motion of live plants in the back and forth water flow observed at high tides. Like real fucoid plants, the lower part (about 2.5 cm) of ‘stipe’ area of the artificial plant was inserted to the rope and tightened. Vinyl covered mesh (3.5 mm diam.), cross-linked every 2.5 cm x 5.1 cm, was used to hold the treatment plants to the substratum. This mesh is strong and yet malleable enough to bend with the  19 rock substratum (Shaughnessy 1994). Mesh strips were cut so that each was 90 cm x 4 cm allowing five treatments (including a control for an empty spot) thaffi haphazardly placed 20 cm apart. The ropes with fucoid thalli and artificial plants inserted were attached to the mesh strip with an electrician’s cable tie. Seven mesh strips (7 replicates) were horizontally attached to the rock substratum at Nudibranch Point where mixed stands of the three species occurred. The orientation of existing live fucoids, which was mostly caused by out-going waves and the slope of rock substratum, was used as an indicator for orienting mesh strips, because all the treatment thalli should necessarily be sitting on the Mazzaelia beds in the same direction at every low tide to achieve the treatment effects, especially for canopy and phenolic effects. Attachment sites for mesh strips were chosen where there  was an untouched Mazzaella bed (by existing fucoids, other algae or barnacles) underneath each treatment thallus. The entire mesh strip was kept in contact with the rock substratum by using either concrete nails or bolts secured at each end and at the middle of the strip. Some artifacts due to using artificial plants were considered as noted in Connell (1974) and Underwood (1980). To test for effects on understory Mazzaelia due to the shade caused by fucoids, replicate light readings (mean ± SE, n=5) were taken under the fucoid canopy, ‘clear’ and ‘dark’ artificial plants and natural light intensity (‘control’) using ’ of light m s a Licor (LI- 185A) quantum photometer. Under no canopy, 2600±0.0 iE 2 ’ under F. distichus, 13.7±4.4 pE m s reached the substratum, compared to 12.9±1.8 iE 2 1 under ‘clear’ artificial plants. To obtain s 2 1 under P. limitata and 2650±6.0 E m s 2 m the ‘dark’ artificial plant that best mimicked the shade cast by fucoid canopies, I haphazardly made eight holes (2.5 mm diam.) on each blade using a hole punch so that 1 of light reached the substratum. Replicate temperature readings were s 2 14.3±4.0 pE m also made under each treatment plant. In the sunny mid-day of June 23, 1994, temperatures under F. distichus, P. limitata and ‘dark’ plants were 18.5±0.7 °C, 19.2±0.9 0, 20±0.4 °C, respectively. High temperatures under the ‘clear’ plants (due to a C greenhouse effect) was lowered by making twelve holes (3.8 mm diam.) on each blade so  20 that a similar temperature was obtained compared to that of the control (23.5±0.5 °C, 23.1±0.3 °C, respectively). Size class (5 mm difference in each class; size class 1 for <5mm, size class 2 for 510mm, etc.) densities of M. cornucopiae blades in 2x2 cm quadrats under each treatment plant were measured in the field, and these were then converted to biomass (g) using the regression coefficient (Fig. 2.3) between size class and wet weight. Data were collected at the beginning (June, 1994) and the end (August, 1994) of the experiment. Although the sites for quadrat placement were not marked, sampling was done at the center under each treatment plant so some portion of the 4 cm 2 areas for the sampling probably overlapped. Therefore, a 5x2 (Treatment x Time) ANOVA with repeated measures on Time factor and a preplanned multiple comparison with Bonferroni adjusted probability (p=O.OS /4= 0.0125) were used for analysis and hypothesis tests. Since it is a repeated measures design, probability values for a multivariate statistic (Wilk’s lambda), rather than for univariate statistics, were used for the multiple comparisons. Hypothesis I (shading effect) was tested by comparing Mazzaella biomass between ‘clea? and ‘dark’ artificial plants. Hypothesis II (scouring effect) and Hypothesis III (phenolic effect) were tested by comparing ‘clear’ vs. control, and ‘dark’ vs. Fucus and Peivetiopsis, respectively. If an effect was significant, a further analysis was done to examine if the effect influenced the size class structure of M. cornucopiae. For each size class, an independent t-test was applied to compare the changes in blade density (density after exp. before exp.) between -  two factors. Since there were six size classes, the t-test was performed six times with a Bonferroni-adjusted probability value (p=O.05 /6  =  0.0083). The homogeneity of variance  was tested using Bartlett’s test (Wilkinson et al. 1992) and Cochran’s test (Winer 1971). The results of both tests were not significant (p>0.05) indicating an equal variance. The residuals were plotted for inspection for normality. A few artificial plants and fucoid thalli were partly broken or lost during the two-month experimental period, and these were  21 dropped from the analyses. As a result, a reduction (e.g., to 5) in replicates occurred in certain treatments.  RESULTS  The relationship between Mazzaella cover and fucoid recruitment  Lower recruitment of Fucus distichus occurred in the piots where greater M. cornucopiae cover existed (Pearson Product-Moment Correlation, r=-O.423) (Fig. 2.1). Pelvetiopsis limitata exhibited the same pattern (r=-O.523) (Fig. 2.1). Both correlations were significant (Fucus, F(1.132)=28.73, p<O.0001; Pelvetiopsis, F(1,135)=50.84, p<O.0001).  Effects of pruning Mazzaella turf on fucoid recruitment  Pruning Mazzaella blades resulted in a significant reduction of percent cover of this alga in most plots, particularly in the ‘total pruning’ plots. The difference in percent cover between the total pruning plots and the partial pruning or control plots was about 24% at the first sampling date. This difference became minor at the later sampling dates as Mazzaella blades regrew (Fig. 2.4). This indicated that the pruning treatment was only effective at the early sampling date, October 1993, which was four months after initiation. The results of ANCOVA, with Mazzaella-pruning treatment as an independent variable and Mazzaella percent cover as a covariate, showed that the pruning-treatment term was not significant for every sampling date, while the Mazzaella percent cover term was significant (p.<O.05) in every case (Table 2.1). This suggests that significantly low Mazzaella cover in the ‘total pruning’ plots (compared to the control plots) was responsible for greater numbers of fucoid recruits in those plots (Fig. 2.5; A); this effect was not due to  22 pruning because Mazaeila cover and fucoid recruitment were negatively correlated (Fig. 2.1). In other words, the number of fucoid recruits adjusted by the effect of Mazzaelia cover (Fig. 2.5; B) was not significantly (p>0.05) changed by the pruning treatments at any sampling date (Table 2.1). Therefore, removing Mazzaella turf (canopy) which was considered ‘blocking’ recruits or ‘shading’ growing recruits from the results of Experiment 1 did not increase the fucoid recruitment in this manipulative experiment.  Effects of Mazzaella turf on post-recruitment survivorship of fucoids  The Mazzaella turf influenced the two fucoid species differently in their survivorship after the recruitment stage. The mean life span of Fucus distichus recruited within the turf was relatively short (2.2 months). However, as the mean distance (see METHODS) between individual F. distichus and turf became greater, Fucus life span showed a sharp increase and reached a peak (5.4 months) at a distance of about 0.5 cm from the edge of the turf (Fig. 2.2). The mean life span then decreased gradually to a distance of approximately 2 cm, and at >2 cm the changes in life span were relatively small. Patterns in life span of Pelvetiopsis limitata were different from those of F. distichus. The longest mean life span of Pelvetiopsis limitata (5.0 months) was found in the individuals within Mazzaelki turf. However, the life span along with the neighbor distance changed irregularly (Fig. 2.2). The results of one-way ANOVA and a post hoc test comparing the life span of the three distance groups indicated that in F. distichus the life span of the ‘close’ group (0.30.7 cm) was significantly higher than that of either ‘contact’ (within turf or 0 cm) or ‘far’ (>2.0 cm) group (Table 2.2). The life spans in the ‘contact’ and ‘far’ groups was not significantly different. However, in P. limitata the life span of all three groups were similar (5.0, 4.6, 4.3 months for ‘contact’, ‘close’, ‘far’, respectively). When comparing the two  23 fucoid species for their life span with the same distance group, there was no significant difference between the two species in any distance group (independent t-test at p=O.05). However, the mean life span in the ‘contact’ group was notably different (2.2 months for Fucus and 5.0 months for Pelvetiopsis; p=O.004 at df=20 if pooled variances were used), although this difference was not statistically significant (the use of separate variances was suggested because of the severely unequal sample sizes, n=16 and 6, respectively; Wilkinson et al. 1992).  Effects of fucoids on Mazzaella cornucopiae  The effects of live fucoid thalli and of the two types of artificial fucoid on Mazzaella biomass showed that there were significant differences in biomass of M. cornucopiae among the treatments (p=O.OO1, one-way ANOVA with repeated measures). Biomass (blade wet weight) of M. cornucopiae generally decreased during the experimental period (June to August, 1994). The mean biomass for the control decreased by 0.38g (16.8%) per 2x2 cm plot (Fig. 2.6). The Mazzaella biomass under the ‘clea? artificial plant showed the least reduction (0. 14g, 9.0%) and that under the ‘dark’ artificial plant had the largest reduction (0.88g. 50.7%) (Fig. 2.6). Of the 3 hypotheses tested, only the canopy effect (shading) was significant. Mazzaella biomass change (after-experiment minus before-experiment) in ‘dark’ was significantly greater than in ‘clear’; that is, biomass of Mazzaella was affected by reduction of light (Table 2,3). Reduction of light also caused a color change of Mazzaella thalli, from light greenish red to dark red. The scouring effect, tested by comparing the control with ‘clear’, showed no significant difference in the biomass change between these two factors, hence the second hypothesis was rejected. The hypothesis of allelopathy (tested by comparing the ‘dark’ with either ‘Fucus’ or ‘Pelvetiopsis’) was rejected for both fucoids  24 (p=O. 189 and p=O.374, respectively), indicating that Mazzaelki biomass was not affected by allelopathy. The significant canopy effect of fucoids on Mazzaella biomass was further analyzed for its potential effect on the size class of the red alga. Mazzaella blade densities decreased in all size classes under the canopy of ‘dark’ artificial blades, whereas the blade densities under the ‘clear’ artificial blades had a variable change depending on the size classes (Fig. 2.7). Independent t-tests for the blade density changes (‘after’ minus ‘before’) between ‘clear’ and ‘dark’ on each size class indicated that only blades in size class 2 were significantly different (p=O.008) in density between the two treatments.  DISCUSSION  The complexity of algal interactions among the three dominants  The results presented here provide an example of bi-directional interactions in which competitive dominance is reversed depending on the developmental stages or life history stages of competitors. Space preemption by Mazzaella corlwcopiae lowered the number of recruits of both Fucus distichus and Pelvetiopsis limitata within its turf. However, once the fucoids successfully recruited, they formed a canopy and reduced Mazzaella biomass by shading. The preemption of suitable space by a turf-forming species and the subsequent reduction of settlement of other morphological forms has been commonly observed by many authors (Lubchenco 1978, Hruby and Norton 1979, Ambrose and Nelson 1982, Chapman 1984, Kennelly 1987). From this literature, the mechanisms by which the Mazzaella turf prevented recruitment of fucoids can be presumed to be as follows, One possibility is that the physical occupation of space by turf reduces the area available for settlement by fucoid propagules. Another is that the turf may outcompete juvenile fucoids  25 for light or nutrients. However, experimental removal of turf did not significantly increase recruitment of either Fucus or Peivetiopsis. This was an unexpected result even though there were slightly more fucoid recruits in the “total-pruning” plots than in control plots. This result might imply that there must be another factor(s) involved in survivorship of juvenile fucoids in the turf-removed plots. These factors may be the relatively stronger desiccation and the increased wave impact caused by removing the turf, discussed in the next section. Morphological differences in adult plants between Mazzaella and the two fucoids brought these algae to another arena of competition when fucoids successfully settled nearby or within the red algal turf. Olson and Lubchenco (1990) documented that the arena of competition changes as the developmental stages of competitors change; as a result, the outcome of competition may depend on the developmental stages that are competing. In this study, as fucoids grew (>3-4 cm), they formed a canopy and reduced the biomass of Mazzaella underneath their thalli because of light interception. A similar case has been reported by Lubchenco (1983). She found that Enteromorpha outcompeted Fucus germlings early in succession by its faster growth rate in the absence of grazers. However, Fucus became established and overtopped Enteroinorpha when grazers were effective, and in the long term Fucus excluded Enteromorpha from the rock surface by preempting space. In the upper intertidal communities at Nudibranch Pt. and Prasiola Pt., successful fall recruitment of both fucoids to the substratum within Mazzaella beds (see Chapter 5) could be possible because Littorina sp. prefered to eat Mazzaelia and the intensity of this preferential feeding could be increased in summer months due to increasing snail densities (see Chapter 3). The summer decline of Mazzaella populations opens up free space for fucoids which have their recruitment peak in October. These series of coincident events change the direction of competition and its outcome. Fucoids did not affect adult Mazzaella either by scouring (or whiplash) or by their chemical contents such as phenolic compounds. However, allelopathic effects of brown  26 algae were noted to be important many years ago when workers studied the antibacterial and antialgal properties of phenolic compounds in dictyotalean and fucalean algae (McLachlan and Craigie 1966, Fletcher, 1975). Schiel and Foster (1986) have pointed out its importance in determining the dynamics of some communities, and some studies have shown that allelopathic effects of brown algae (probably due to phenolic compounds) inhibited the abundance of other algae in intertidal (Fletcher 1975) and subtidal habitats (Kennelly 1987). Fucus distichus and Pelvetiopsis limitata were known to contain phenolic compounds of 4.35% and 4.93% dry mass, respectively (Steinberg 1985). However, the biomass of Mazzaella underneath real plants was slightly less reduced than that underneath ‘dark’ artifitial plants; this indicated that further reduction of Mazzaella biomass caused by the chemicals did not occur at all. Although there was a possibility that fucoids whiplashed and affected the settlement of barnacles and some ephemeral algae in small gaps (5x5 cm) cleared in the proximity of the fucoids (Chapter 5), fucoids did not affect Mazzaeila growing within their boundary of scouring. The result of careful field observation on Mazzaelia around mature fucoid thalli supported this result. The absence of a scouring effect by the fucoids is probably because the relative size of the fucoids is not large enough (Grant 1977, Farrell 1989) to affect adult Mazzaella turf. The reduction of light (about 0.51% of unshaded condition) by fucoids affected the size class density of Mazzaella, with a particularly significant decrease in the size class 2 (0.5-1 cm in blade length). It is not certain why this particular size class was significantly reduced. Shading effects on understory species in algal communities have been extensively studied in subtidal kelp communities (Dayton et. al. 1984, Reed and Foster 1984, Santeiices and Ojeda 1984, Kennelly 1989, Dean et. al. 1989, Reed 1990) and intertidal algal communities (Dayton 1971, 1975, Schonbeck and Norton 1980, Ang and DeWreede 1992). In intertidal habitats, shading has sometimes been reported as a positive factor, especially where or when desiccation is strong (Dayton 1975, Ang 1991). However, in this study light may have been important for Mazzaella growth, even when the population  27 is declining between June and August 1994, such that shading reduced more biomass than the amount of natural reduction in the control sites. This result ensures that the deleterious effect on Mazzaelia by fucoid shading should also be effective at other times of the year; thus the effect may be more intense when Mazzaella actively grows in February to April (to be discussed in Chapter 5). In this study, competitive superiority was reversed depending on the developmental stages of fucoids. Mazzaelia outcompeted fucoids at the recruitment stage of the fucoids by preempting space. However, after the recruitment of fucoids, both fucoids inhibited the growth of Mazzaella by shading. When a reversal of competitive superiority occurs competitive elimination is less likely (Connell 1983). In his review of field experiments on competition, Connell (1983) reported that only a few studies were sufficiently comprehensive to provide evidence concerning such reversals. There were a few examples on non-hierarchical competitive networks among marine subtidal animals (reviewed by Connell and Keough 1985). Similarly among macroalgae, the direction of competitive overgrowth was reversed in the presence of grazers (Paine 1984, Steneck 1985). Unlike these two studies on macroalgae, the change of the direction of competition between Mazzaella and the fucoids was more likely due to the difference in morphology and life history stages of these algae rather than to the presence of herbivorous snails. Preferential feeding on Mazzaella by snails did facilitate the reversal of the competitive direction but it is also possible that such reversal could occur as plants grew in the absence of the grazer. This will be discussed more in chapter 3.  The mechanism of positive interaction and neighbor distance  The relatively high survivorship of Fucus growing at the edge of Mazzaella turf is evidence of a direct facilitation by the neighboring species against potentially limiting physical stresses. A possible explanation for the different longevity patterns of Fucus and  28 Pelvetiopsis with respect to distance from Mazzaella is based on the hypothetical relationships between competitive interactions and physical stresses (Fig. 2.8). Individual plants of Fucus and Pelvetiopsis growing within the Mazzaella turf may encounter less desiccation because of the moisture-holding ability of the red algal tuif (Hay 1981, Padilla 1984). As the fucoids grow up, they may have limited space for holdfast expansion because of Mauaeila’s preemption of space. This increases the susceptibility of fucoids to wave impact. This lethal factor is probably more effective on Fucus which has relatively wider thalli than Pelvetiopsis. However, biomechanical differences between the two fucoids with respect to wave force have not been reported. It has been reported that the competitive outcomes in macroalgae were sometimes influenced by local (microhabitat) physical conditions (e.g., desiccation, wave action) (Padilla 1984). Thus, the short lifespan of Fucus within Mazzaella turf may be a result of competition coupled with physical stress. The microhabitat conditions at the turf edge can be ameliorated by Mazzaella’s buffer against wave impact and also by sufficient moisture from turf, in addition no competition for space apparently occurs. However, at a distance >2 cm, both desiccation and wave impact can be lethal for Fucus because it is “too far” to get the buffer and moisture from Mazzaella (its blade length is about 2-3 cm). I suggest a model for Fucus survivorship in relation to the distance from the primary space-holder, Mazzaella, and with respect to the two limiting factors, competition and physical stress (Fig. 2.9). Competition is probably effective only if the two species are in contact. The intensity of competition then drops suddenly as the neighbor distance increases. At the same time, physical stresses gradually increase as the two species are further apart. The highest survivorship of Fucus occurs at the point (or distance) where the combined effect of two factors is minimal. The effect of Maz.zaella on post-recruitment survival of Fucus as shown by this study suggests three important factors that structure this algal assemblage. First, direct positive interactions may be common, predictable, and pervasive forces in natural  29 communities and in physically harsh environments in particular. The mechanism of the interaction found in this study is an example of neighbor habitat-amelioration (Bertness and Callaway 1994). This can be differentiated from the results of Paine (1980) and Dethier and Duggins (1984) which showed empirical evidence of positive indirect interactions and feedback mechanisms in food webs. In some harsh intertidal habitats, there is limited evidence that neighbor species improved deleterious habitat conditions (e.g., severe heat and desiccation pressure) and facilitated recruitment of barnacles (Bertness 1989) and mussels (Bertness and Grosholz 1985). Dayton (1975) reported an example of direct positive interaction between macroalgae and showed that larger overstorey algae (e.g., Hedophyllum) protected smaller understorey species from desiccation or radiation damage. Second, the concept of neighbor distance, which was first introduced in terrestrial plant interactions (Fawcett 1964, Turkington and Harper 1979), should be adapted to marine benthic algal interactions with a modified mechanistic concept. Since no belowground competition exists among benthic algae, the outcome of algal interactions is solely dependent upon the morphological form and size of thalli (Hay 1981, Paine 1984); this is particularly true if the limiting resource is space or the competition is by interference. This study confirms the idea that for organisms that compete for space, and in particular clonal ones such as plants and some sessile animals, competition occurs mainly between neighbor species (Connell 1983). In addition, neighbor distance in this upper intertidal habitat determined which factor(s) are responsible for the fate of an individual plant. Thalli of Fucus distichus encountered different biotic and abiotic factors (competition to protection to physical stress) as neighbor distance increased. Consequently, the longer survival of a Fucus thallus in this high intertidal site which features an intermediate to strong wave action requires a certain optimum distance (e.g., 0.5 cm±0.2) from its neighbor. Therefore, the maintenance of a Fucus population is partially dependent on the presence of Mazzaella cornucopiae, particularly when Fucus increases its population size toward the site with  30 more extreme conditions (e.g., higher tidal height, more wave-exposed) from its own natural habitats with milder conditions. Third, the importance of interactions at a small spatial scale in this community should be noted. Paine (1990) pointed out that interaction mechanisms on a small (<2 m ) 2 scale are important in understanding the nature of algal assemblages. The range of spatial scales at which one species affected another was surprisingly small (<2 cm). This should be considered in designing field experiments for detecting species interactions of small organisms. Brawley and Johnson (1993) reported that microclimate was an important predictor of the likelihood and spatial pattern of survival of settled larvae, reproductive propagules, and other microscopic stages in the life histories of organisms growing in intertidal and other water-stressed (desiccating) environments. It is certain that relevant spatial scale is mainly determined by the size of organisms interacting (this applies only to plants) and the severity of physical stresses. Bertness and Callaway (1994) proposed that positive interactions during succession and recruitment, as well as among established adults, are unusually common in harsh physical environments for the simple reason that primary space-holders frequently buffer neighbors from potentially limiting stresses. In this perspective, the Mazzaella -Fucus interaction in a stressful environment with strong wave force and desiccation deserves increased empirical attention and should be incorporated into models of community organization in harsh habitats like the upper intertidal zone.  Coexistence of the major species  Biological perspectives  Is coexistence of the three dominants due to the variation in algal interactions among them? This research has shown that patterns of interactions among neighbor species are complicated but well-balanced. The balance in competitive direction (or dominance) between Mazzaella and fucoids as well as the balance in the positive and negative  31 interactions between Mazzaella and Fucus supported the persistence of each member in the community. Looking at their distinctive functional morphology, the shorter turf-forming Mazzaella probably has a competitive disadvantage over upright and relatively fast-growing fucoids because the turf-former may have a reduced photosynthetic activity limited to the upper portions of the thallus due to self-shading and also may suffer nutrition depletion within the turf matrix (Hay 1981). In contrast, Mazzaella must be the species best adapted to high desiccation and strong wave action among the three dominants because it has waterholding capacity and self-cushioning thalli in a dense turf (Hay 1981, Padilla 1984). Fucus distichus is the only species among the three dominants which also appears in the lower zone (upper-mid intertidal), where it grows up to 3-4 times larger than individuals in my study sites. Although this alga has a physiological adaptation (i.e., high photosynthetic rate in air) to tidal emergence (Johnson et al. 1974, Quadir et al. 1979), it still may have some morphological limitation (e.g., relatively wider thallus than Pelvetiopsis) to survive in the force of breaking waves. Consequently, the advantage from the positive interaction with Mazzaella, the relatively high recruitment (compared to Peivetiopsis; J. H. Kim unpubl. data) and the facilitation of settlement by barnacles (Chapter 4) are the important parameters which allow Fucus to maintain its co-dominance in this community. Natural habitats for Pelvetiopsis limitata appear to be confined to the upper intertidal, unlike F. distichus (Sousa 1984, Farrell 1989, 1991). According to a recent review on fucoid algae by Chapman (1995), there was no information available for the physiological and reproductive ecology of P. limitata. However, in this study P. limitata seemed to be a relatively better-adapted species to the upper intertidal zones in Prasiola and Nudibranch Points than F. distichus, because survival and colonization of this alga was not affected by some biological (i.e., the presence of Mazzaella) and physical (i.e., size of disturbance; to be discussed in Chapter 5) factors. From a biological perspective, coexistence of these three macroalgae in this community is likely because of the biological adaptation of each species to physical stress  32 and because the outcomes of their interactions are well-balanced. Thus, competitive elimination is unlikely because competition among the species is symmetric (Connell 1983). However, it still remains an open question whether the biological interactions are relatively more or less important in structuring a community under harsh physical conditions such as those experienced in this algal community. The data presented in this chapter provide experimental evidence that detectable biological interactions occur in this community, and such interactions, together with other factors described in subsequent chapters of this thesis, are important determinants of community dynamics.  33 REFERENCES Ambrose, R. F., Nelson, B. V. (1982). Inhibition of giant kelp recruitment by an introduced brown alga. Bot. Mar. 25: 265-267. Ang, P. 0. (1991). Natural dynamics of a Fucus distichus (Phaeophyceae, Fucales) population: reproduction and recruitment. Mar. Ecol. Prog. Ser. 78: 7 1-85. Ang, P. 0., DeWreede, R. E. (1992). Density-dependence in a population of Fucus distichus. Mar. Ecol. Prog. Ser. 90: 169-181. Bertness, M. D. (1989). Intraspecific competition and facilitation in a northern acorn barnacle population. Ecology 70: 257-268. Bertness, M. D., Grosholz, T. (1985). Population dynamics of the ribbed mussel, Geukensia demissa: the cost and benefits of an aggregated distribution. Oecologia 67: 192-197. Bertness, M. D., Callaway, R. (1994). Positive interactions in communities. Tree 9: 191193. Brawley, S. H., Johnson, L. E. (1993). Predicting desiccation stress in microscopic organisms: the use of agarose beads to determine evaporation within and between intertidal microhabitats. J. Phycol. 29: 528-535. Carlson, D. J., Carlson, M. L. (1984). Reassessment of exudation by fucoid macroalgae. Limnol. Oceanogr. 29: 1077-1087. Chapman, A. R. 0. (1984). Reproduction, recruitment and mortality in two species of Laminaria in southwest Nova Scotia. J. Exp. Mar. Biol. Ecol. 78: 99-109. Chapman, A. R. 0. (1995). Functional ecology of fucoid algae: twenty-three years of progress. Phycologia 34: 1-32. Connell, J. H. (1972). Community interactions on marine rocky intertidal shores. Ann, Rev. Ecol. Syst. 3: 169-192. Connell, J. H. (1974). Field experiments in marine ecology. In: Mariscal, R. (ed.) Experimental marine biology. Academic Press. New York. pp. 2 1-54. Connell, J. H. (1975). Some mechanisms producing structure in natural communities: a model and evidence from field experiments. In: Cody, M. L., Diamond, J. M. (eds.) Ecology and evolution of communities. Bellcnap, Cambridge. pp. 460-490. Connell, J. H. (1983). On the prevalence and relative importance of interspecific competition: evidence from field experiments. Am. Nat. 122: 661-696. Connell, J. H. (1990). Apparent versus “reaU’ competition in plants. In: Grace, J. B., Tilman, D. (eds.) Perspectives on plant competition. Academic Press, Inc. New York. pp. 9-26. Connell, J. H., Slatyer, R. 0. (1977). Mechanisms of succession in natural communities and their role in community stability and organization. Am. Nat. 111: 1119-1144.  34 Connell, J. H., Keough, M. J. (1985). Disturbance and patch dynamics of subtidal marine animals on hard substrata. In: Pickett, S. T. A., White, P. S. (eds.) The ecology of natural disturbance and patch dynamics. Academic Press. New York. pp. 125-151. 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. Dayton, P. K. (1975). Experimental evaluation of ecological dominance in a rocky intertidal algal community. Ecol. Monogr. 45: 137-159. Dayton, P. K., Oliver, J. S. (1980). An evaluation of experimental analyses of population and community patterns in benthic marine environments. In: Tenore, K. R., Coull, B. C. (eds.) Marine benthic dynamics. University of South Carolina Press. pp. 93120. Dayton, P. K., Currie, V., Gerrodette, T., Keller, B. D., Rosenthal, R., Ven Tresca, D. (1984). Patch dynamics and stability of some California kelp communities, Ecol. Monogr. 54: 253-289. Dean, T. A., Thies, K., Lagos, S. L. (1989). Survival of juvenile giant kelp: the effects of demographic factors, competitors and grazers. Ecology 70: 483-495. Denley, E. J., Dayton, P. K. (1985). Competition among macro algae. In: Littler, M. M., Littler, D. S. (eds.) Ecological field methods: macroalgae. Handbook of phycological methods. Cambridge University Press. New York. pp. p. 511-530. Dethier, M. N., Duggins, D. 0. (1984). An ??indirect commensalism” between marine herbivores and the importance of competitive hierarchies. Am. Nat. 124: 205-2 19. Duggins, D. 0. (1981). Interspecific facilitation in a guild of benthic marine herbivores. Oecologia 48: 157-163. Farrell, T. M. (1989). Succession in a rocky intertidal community: the importance of disturbance size and position within a disturbed patch. J. Exp. Mar. Biol. Ecol. 128: 57-73. Farrell, T. M. (1991). Models and mechanisms of succession: an example from a rocky intertidal community. Ecol. Monogr. 61: 95-113. Fawcett, R. G. (1964). Effects of certain conditions on yield of crop plants. Nature 204: 858-860. Fisher, R. A., Yates, F. (1963). Statistical tables for biological, agricultural and medical research (6th ed.). Oliver & Boyd. Edinburgh. pp. 138. Fletcher, R. L. (1975). Heteroantagonism observed in mixed algal cultures. Nature (London) 253: 534-535. Grant, W. S. (1977). High intertidal community organization on a rocky headland in Maine, U.S.A. Mar. Biol. 44: 15-25. Hawkins, S. J., Hartnoll, R. 0. (1983). Grazing of intertidal algae by marine invertebrates. Oceanogr. Mar. Biol. Annu. Rev. 21: 195-282.  35 Hay, M. E. (1981). The functional morphology of turf-forming seaweds: Persistence in stressful marine habitats. Ecology 62: 739-750. Hay, M. E. (1986). Associational plant defenses and the maintenance of species diversity: turning competitors into accomplices. Am. Nat. 128: 617-641. Holt, R. D. (1977). Predation, apparent competition, and the structure of prey communities. Theor. Pop. Biol. 12: 197-229. Hruby, T., Norton, T. A. (1979). Algal colonization on rocky shores in the Firth of Clyde. J. Ecol. 67: 65-77. Johnson, W, S., Gigon, A., Gulman, S. L., Mooney, H. A. (1974). Comparative photosynthetic capacities of intertidal algae under exposed and submerged conditions. Ecology 55: 450-453. Kastendiek, J. (1982). Competitor-mediated coexistence: interactions among three species of benthic macroalgae. J. Exp. Mar. Biol. Ecol. 62: 201-210. Keddy, P. A. (1989). Competition. Chapman and Hall. New York. pp. 202. Kennelly, S. 3. (1987). Inhibition of kelp recruitment by turfing algae and consequences for an Australian kelp community. 3. Exp. Mar. Biol. Ecol. 112: 49-60. Kennelly, S. J. (1989). Effects of kelp canopies on understorey species due to shade and scour. Mar. Ecol. Prog. Ser. 50: 2 15-224. Lindquist, J. L., Rhode, D., Puettmann, K. 3., Maxwell, B. D. (1994). The influence of plant population spatial arrangement on individual plant yield. Ecol. Appi. 4: 518524. Lubchenco, 3. (1978). Plant species diversity in a marine intertidal community: importance of herbivore food preference and algal competitive abilities. Am. Nat. 112: 23-39. Lubchenco, J. (1983). Littorina and Fucus: effects of herbivores, substratum heterogeneity and plant escapes during succession. Ecology 64: 1116-1123. Lubchenco, 3., Gaines, S. D. (1981). A unified approach to marine plant-herbivore interactions. I. Populations and communities. Ann. Rev. Ecol. Syst. 12: 405-437. Mack, R. N., Harper, I. L. (1977). Interference in dune annuals: spatial pattern and neighborhood effects. J. Ecol. 65: 345-363. McLachlan, 3., Craigie, 3. S. (1966). Antialgal activity of some simple phenols. J. Phycol. 2: 133-135. Menge, B. A., Sutherland, J. P. (1987). Community regulation: variation in disturbance, competition, and predation in relation to environmental stress and recruitment. Am. Nat. 130: 730-757. Olson, A. M., Lubchenco, 3. (1990). Competition in seaweeds: linking plant traits to competitive outcomes. 3. Phycol. 26: 1-6.  36 Padilla, D. K. (1984). The importance of form: differences in competitive ability, resistance to consumers and environmental stress in an assemblage of coralline algae. J. Exp. Mar. Biol. Ecol. 79: 105-127. Paine, R. T. (1977). Controlled manipulations in the marine intertidal zone and their contributions to ecological theory. Spec. Pubi. Acad. Nat. Sci. 12: 245-270. Paine, R. T. (1980). Food webs: linkage, interaction strength and community infrastructurre. J. Anim. Ecol. 49: 667-685. Paine, R. T. (1984). Ecological determinism in the competition for space. Ecology 65: 1339-1348. Paine, R. T. (1990). Benthic macroalgal competition: complications and consequences. J. Phycol. 26: 12-17. Quadir, A., Harrison, P. J., DeWreede, R. E. (1979). The effects of emergence and submergence on the photosynthesis and respiration of marine macrophytes. Phycologia 18: 83-88. Reed, D. C. (1990). An experimental evaluation of density dependence in a subtidal algal population. Ecology 71: 2286-2296. Reed, D. C., Foster, M. S. (1984). The effects of canopy shading on algal recruitment and growth in a giant kelp forest. Ecology 65: 937-948. Santelices, B., Ojeda, F. P. (1984). Effects of canopy removal on the understory algal community structure of coastal forests of Macrocystis pyrifera from southern South America. Mar. Ecol. Prog. Ser. 14: 165-173. Schiel, D. R., Foster, M. S. (1986). The structure of subtidal algal stands in temperate waters. Oceanogr. Mar. Biol. Annu. Rev. 24: 265-307. Schoener, T. W. (1983). Field experiments on interspecific competition. Am. Nat. 122: 240-2 85. Schonbeck, M. W., Norton, T. A. (1980). Factors controlling the lower limits of fucoid algae on the shore. J. Exp. Mar. Biol. Ecol. 43: 13 1-150. Shaughnessy, F. J. (1994). Population differentiation of two sympatric species of red algae, Mazzaella splendens and Mazzaella linearis, in Baridey Sound, British Columbia, Canada. Ph. D. Thesis. University of British Columbia. Sousa, W. P. (1984). The role of disturbance in natural communities. Ann. Rev. Ecol. Syst. 15: 353-391. Sousa, W. P. (1984). Intertidal mosaics: Patch size, propagule availability and spatially variable patterns of succession. Ecology 65: 19 18-1935. Steinberg, P. D. (1985). Feeding preferences of Tegukifuneblaris and chemical defenses of marine brown algae. Ecol. Monogr. 55: 333-349.  37 Steneck, R. S. (1985). Adaptations of crustose coralline algae to herbivory: patterns in space and time. In: Toomey, D. F., Nitecki, M. H. (eds.) Paleoalgology: Contemporary Research and Applications. Springer-Verlag, Berlin. pp. 352-366. Tilman, D. (1994). Competition and biodiversity in spatially structured habitats. Ecology 75: 2-16. Turkington, R., Harper, J. L. (1979). The growth, distribution and neighbour relationships of Trifolium repens in a permanent pasture. I. Ordination, pattern and contact. J. Ecol. 67: 201-218. Underwood, A. J. (1980). The effects of grazing by gastropods and physical factors on the upper limit of distribution of intertidal macroalgae. Oecologia 46: 201-2 13. Underwood, A. J., Denley, E. J. (1984). Paradigms, explanations, and generalizations in models for the structure of intertidal communities on rocky shores. In: Strong, D. R., Simberloff, D., Abele, L. G., Thistle, A. B. (eds.) Ecological communities: Conceptual issues and the evidence. Princeton University Press. New Jersey. pp. 151-180. Vandermeer, J. (1980). Indirect mutualisms: variations on a theme by Stephen Levine. Am. Nat. 116: 441-448. Wilkinson, L., Hill, M. A., yang, E. (1992). SYSTAT: Statistics, version 5.2.1 edition. SYSTAT. Evanston, IL, USA. Winer, B. J. (1971). Statistical principles in experimental design. McGraw-Hill. New York.  38  Table 2.1 Results of ANCOVA on fucoid recruitment in the different Mazzaella pruning plots. The ANCOVA was applied to each sampling date. The covariate is Mazzaella percent cover in each plot (M-cover). Three treatments are ‘total pruning’, ‘partial pruning’ and no pruning (control), see text for details.  October 1993 Source of variation  February 1994  April 1994  MS  MS  F  0.03 0.967 9.41 0.005  df  MS  2 1 23  3.48 1.51 0.242 23.12 10.02 0.004 2.31  1.60 0.90 0.419 34.29 19.2 0.000 1.78  0.07 20.06 2.13  2 1 23  2.23 11.67 2.64  0.19 14.12 1.61  0.60 0.51 0.608 21.18 17.87 0.000 1.19  F  12  F  12  12  Fucus Treaunent M-cover Error  Pelvetiopsis Treatment M-cover Error  0.84 0.444 4.42 0.047  0.12 0.889 8.78 0.007  39 Table 2.2 Comparisons of longevity among the distance groups in each species and between F. distichus and P. lirnitata. at each distance group. Means are number of months.  CONTACT (0 cm)  CLOSE (0.3 0.7 cm)  Mean Fucus t-test (p) Pelvetiopsis  2.19  Mean 0.95  0.086 5.00  FAR (>2.0 cm)  -  16  5.42  *  1.50  0.53  52  0.322 6  4.62  0.51  One-way ANOVA  Mean  SE  N  3.57  0.51  56  0.004  44  0.831  0.175 53  4.25  0.56  Tukey’s HSD on Fucus CONTACT  *  CLOSE  CONTACT  1.000  CLOSE  0.008  1.000  FAR  0.405  0.031  FAR  1.000  Separate variances are used due to the severely unequal sample sizes (Wilkinson et al. 1992).  40 Table 2.3 Results of multivariate statistics as a preplanned multiple comparison for testing three hypotheses. A Bonferroni adjusted probability value (p=0.05 / 4 = 0.0 125) was used to compare each pair of treatments.  Wilk’s lambda Canopy effect (Clear vs. Dark)  1i  0.541  9.320  2, 22  0.001  0.974  0.293  2, 22  0.749  (Darkvs.Fucus)  0.859  1.801  2,22  0.189  (Dark vs. Pelvetiopsis)  0.9 14  1.029  2, 22  0.374  Scouring effect (Control vs. Clear) Allelopathic effect  41  FUCUS  -  3  -  r=-O.423 n=1 34 p<O.0001  21 Cl) I D Ct  -  0 w Ct U  0 Ct  4  D  3  w  z (!3 0 -J  r=-O.523 n=137 p<0.0001  2 1 0 —1 0  20  40  60  80  100  MAZZAELLA PERCENT COVER  Fig. 2.1 Correlations of M. cornucopiae percent cover and number of fucoid recruits.  42  25  I  FUCUS 20 15  -  -  -  -  10-  •.  Cl)  -  —n  25”  z  O -J  PELVETIOPSIS 20 15 10••  -  _..  0  I  I  0  2  I  • —  -,-..-...-....  .  •  I  •  I  4  ---....,  •  I  6  8  10  DISTANCE FROM MAZZAELLA TURF (cm)  Fig. 2.2 The effect of Mazzaella turf on fucoids after recruitment stage. Curves are the weighted average of piots.  43  0.09  I -  0.08  0.07-  I  I  W=0.007*(S1.467)  0.06  :  I  :  o.os-  0  I  r2=0. 895 N=162 I  0)  I  I •  :  :  :  -  0.04-  -  0.03  I 0.02:  0.010.00  I  0  1  2  3  I  I  I  I  4  5  6  7  8  MAZZAELLA SIZE CLASS  Fig. 2.3 Relationship between the size class of Mazzaella blade and wet weight (g). Mazzaelta blade height was divided into six 5 mm size classes (SC). SC1 = 1-5 mm, SC2 = 6-10 mm, SC3 = 11-15 mm, SC4 = 16-20 mm, SC5 = 21-25 mm, SC6 >26 mm.  44  70  60 LU C) I— z LU 0  50  LU 0  30  20 OCT93  FEB 94  APR 94  Fig. 2.4 The effect of pruning Mazzaella blade on Mazzaella percent cover in the same experimental plots (lOx 10 cm).  45  5 PT PP PC FT AFP FC  A 4 3  2  1 0  B 4 3  2 1 0 OCT93  FEB94  APR94  Fig. 2.5 The effect of Mazzaella pruning on fucoid recruitment. A. number of fucoid recruits actually occurring in the plots. B. number of fucoid recruits adjusted by Mazzaella percent cover (the covariate) in each plot at each sampling date. These values were used in the ANCOVA. Data are means +SE of 9 replicates. PT for Pelvetiopsis in the total pruning plots, PP for Pelvetiopsis in the partial pruning plots, PC for Pelvetiopsis in the control plots; PT, FP and FC for Fucus in the respective plots.  46  3  C”  E C)  2  ci)  0  Co  Cl)  1  E  0  co 0  After experiment 0 Before experiment CTL  CLE  DRK  FUC  PEL  TREATM ENTS  Fig. 2.6 Changes in Mazzaella biomass (g) under each treatment after 2 months. Cli for control, CLE for the clear artificial plant, DRK for the dark artificial plant, FUC for Fucus, and PEL for Pelvetiopsis. Data are means +SE (n=5 for CLE, FUC and PEL, n=6 for DRK, n=7 for CTL).  47  5.  a) Cl)  c,) C-)  -5.  >,  Dark ØCIear  -io 1  2  3  4  5  6  Size Class  Fig. 2.7 The effect of fucoid shading on Mazzaella size class. Data are means +SE (n=5 for clear and n=6 for dark artificial plant).  WAVE  cm  WAVE  WAVE  Biomechanical difference?  Limited space for holdfast expansion?  WAVE  DESICCA11ON  Fig. 2.8 Diagram of potential factors affecting fucoid mortality. See DISCUSSION for further explanation.  Distance from Mazzaella turf  0  DES I CCATI ON  2.5 2.0 1.5 1.0 0.5  DES I CCATI ON  LETHAL FACTOR  4— NON-LETHAL FACTOR  —  00  49  (I) 0 Cl) Cl) LU  C.)  z  0  I— U)  I I— LU  -J  0 U) >-  0 0  I  0  0.5  1.0  1.5  2.0  3.0  4.0  10  Distance (cm) from Mazzaella  Fig. 2.9 Fucus survivorship curves in relation to the changing effects of competition and physical stress with distance from the edge of a Mazzaella turL The shaded area represents  the probability of Fucus survival.  50 CHAPTER 3  PREFERENTIAL FEEDING BY A LITTORINID: IMPLICATIONS FOR ITS ROLE IN A HIGH INTERTIDAL ALGAL COMMUNITY  INTRODUCTION  The importance of biological factors in controlling the abundance and distribution of algal species is generally thought to decrease with increasing tidal height in rocky shores. Physical factors (e.g., disturbance, desiccation) have been shown to be more important in structuring upper intertidal communities, while algal abundance in the lower intertidal levels appears governed principally by herbivory and competition (Castenholz 1961, Connell 1972, 1975, Chapman 1973, Menge 1978, Underwood 1980). Many of the studies dealing with herbivory and its influence on algal community structure have been confined to the low and mid intertidal zones. Few authors have shown experimental evidence of the impact of herbivory on algal abundance in upper intertidal zones (Dayton 1975, Robles and Cubit 1981, Cubit 1984, Farrell 1991). The relative importance of biological and physical factors requires more experimental field tests to verify its application to upper intertidal zones. In the uppermost intertidal zones of rocky shores dominated by three perennial macrophytes, Mazzaella cornucopiae, Pelvetiopsis limitata and Fucus distichus, dramatic increases of littorinid snails were observed each summer in these habitats. The snails were small (<4 mm in shell length) and seemed to be restricted to upper exposed intertidal areas (Boulding and Van Aistyne 1993, personal observation). Since the foraging behavior of this species has not been described (Norton et al. 1990, Boulding and Van Aistyne 1993), its potential role in the algal community is not known.  51 Littorinid snails are primarily herbivorous and are able to feed on a variety of macro- and micro-algae in numerous habitats using their versatile radulae (see reviews by Newell 1970, Underwood 1979, Hawkins and Hartnoll 1983, Norton et al. 1990). Some littorines are highly selective grazers demonstrating strong preferences for some species and rejecting others after prolonged periods of starvation (Norton et al. 1990). Authors have recognized feeding preference by measuring two components of diet, edibility and attractiveness (Nicotri 1980, Watson and Norton 1985). The former, which is the primary measure of preference, reflects both the speed with which a given food item satisfies the physiological needs of a herbivore and the ease with which that item can be handled and ingested. Attractiveness is largely dependent on the capacity of the herbivore to detect plant chemicals and the non-nutritive characteristics of the plant as a habitat (Nicotri 1980). These two factors, when tested together, have shown a correlation in some studies (Watson and Norton 1985, 1987; used Littorina littorea, L. inariae, L obtusata); in others no correlation was found (Nicotri 1980; used an isopod, Idotea baltica). Adaptation of intertidal littorines to air exposure in relation to their activity and foraging has been described by many authors (Newell et al. 1971, McMahon and RussellHunter 1977, Newell 1979, Underwood 1979, Underwood and Chapman 1985, Voltorina and Sacchi 1990). Newell et al. (1971) demonstrated that L. littorea feeds mainly when moistened by the tide and is quiescent when exposed to air. Similar results have been reported using L. scutulata and L. sitkana from upper intertidal zones on Vancouver Island, Canada, indicating that limited feeding activity occurs during day time emergence whereas active feeding occurs in the early morning emergence, especially on moistened surfaces (Voltorina and Sacchi 1990). In contrast, McMahon (1990) indicated that some eulittoral species of littorines displayed adaptation shown by active movement and foraging during day time emergence. Because of contrasting results, it is difficult to generalize about feeding behavior of the upper intertidal littorines. Underwood and Chapman (1985) have thus pointed out that care should be taken when examining intertidal snail behavior because  52 ambiguous results can be obtained when they occur under different environmental conditions, including tidal height, topography and substratum. The objective of this chapter is to investigate littorine distributions in the field and feeding preference of the snails on the 3 dominant algae in the laboratory. The role of littorine herbivory as a potential biotic factor affecting the structure of the upper intertidal algal community is discussed. Experiments were designed to address the following questions: (1) Are there temporal and spatial variations in the distribution of the littorinids? (2) Which alga is preferentially eaten when presented separately to the littorinids? I use the term ‘single-diet’ for this experiment hereafter. (3) Which alga is preferentially eaten when all algae are simultaneously presented to the littorinids? I use the term ‘multiple-diet’ for this case. Both edibility and attractiveness of the algae were tested. (4) How is feeding behavior and movement of the littorinids affected by simulated tidal conditions (low or high) and light (present or absent)? A field experiment on feeding preference was not attempted because of (1) difficulties in controlling such a small herbivore in the wave-exposed sites, (2) possible cage effects due to the small size of holes in metal mesh needed to retain the snails (Denley and Dayton 1985), and (3) the ineffective non-cage method to exclude littorinids, e.g., the use of copper-based antifouling paints (Cubit 1984).  MATERIALS AND METHODS  Sampling site and organisms  The algae and snails were collected from the upper intertidal zone at Nudibranch Point. Due to limited space and to avoid the impact of manipulated density, natural densities of snails were determined at Prasiola Point, 0.4 km south of Nudibranch Point,  53 where physical and biological conditions (e.g., wave-exposure, slope, algal flora) were similar to the collection site. There is an unresolved taxonomic problem for the Littorina species used in this study. Boulding et al.(1993) distinguished the taxon used in this study from other related species in the northeastern Pacific coast, such as L. sitkana and L. scutulata, based on morphological and allozyme data, and suggested this taxon was an undescribed species. The littorinid in this study could not be identified as a known species for the following reasons. First, there were habitat differences between Littorina sp. and L subrotundata which occurs commonly in estuaries (Boulding et al. 1993), even though Littorina sp. is the species most closely related to or conspecific with L. subrotundata (Reid and Golikov 1991). Secondly, Littorina sp. used in this study have non-planktonic larvae (i.e., direct development) (Boulding et al. 1993), which is different from the planktonic larvae of L. scutulata one of the more common species in the northeastern Pacific (Behrens 1971, ,  Buckland-Nicks et al. 1973, Mastro et al. 1982). Thirdly, L sitkana resides in more sheltered and lower habitats, as well as differing in size and foot color from the species (Buckland-Nicks et al. 1973, Boulding and Van Alstyne 1993). Therefore, I have followed Boulding’s suggestion and used the name Littorina sp. Juvenile L. scutulata or L. plena may have been mistakenly included in the experiments although their numbers were low (<3% of the total snails used), and reflect the in situ species abundance of littorinids at the study site.  Natural densities of snails  Snail densities were monitored at Prasiola Pt., at 6-8 week intervals for 16 months. Three transect lines (each 3 m long) were marked along the same tidal height at the study site. Areas for these transects included variable habitat conditions, such as mixed or monostands of the three macrophytes, bare rock, and variable barnacle cover. Snails from  54 a lOxlO cm quadrat placed at each of 6 randomly selected points (different points for each sampling date) on each transect line were vacuumed for 12 seconds using a rechargeable hand vacuum (Hoover Wet and Dry), counted and returned to the original plot. Another set of collections was made to estimate snail densities in monospecific patches of the three dominant algae. In each of 5 marked (but randomly selected initially) monospecific patches per species (5 replicates per algal species), a lOxlO cm quadrat was randomly placed at each sampling date, and snails were counted using the same collecting protocol. Data were analyzed using two-way analysis of variance to examine both temporal and spatial distribution (among monospecific patches). Another two-way ANOVA was applied to data from the transect lines (natural densities from the general habitat) to assess temporal and spatial (among transect lines) distribution. Snail densities from one sampling date to another were treated independently because they were not obtained from fixed quadrats. Cochran’s test was performed to check homogeneity of variance (Winer 1971). Plots of residuals, where applicable, were examined after each ANOVA to check if error terms were normally distributed. These procedures were applied to all univariate statistics in this study. Data were square-root transformed, where necessary, to meet the assumption of equal variances among treatments. All data analyses were performed using SYSTAT, version 5.2.1 for Macintosh, (Wilkinson et al. 1992).  Single-diet experiments  This experiment was conducted in June 1992. Littorina sp. was collected with the hand vacuum and maintained in running seawater tanks without any added food for 3 days before each alga was added. Cages were made from plastic pot liners (3090 cm ), into 3 each of which four windows were cut. Nylon mesh was attached over each window as well as over the open top of the cages. Healthy, nongrazed, nonreproductive thalli of each  55 macrophyte were presented individually to snails in separate experimental cages. I collected short fucoids (3-5 cm in length), so that a similar surface area for each species was exposed to herbivores. Thalli (blotted wet weight 3.0±0.03 g per cage) of each alga were placed in each of 5 cages containing snails and in each of 3 control cages without snails. Each thallus was tied to the bottom of the cage with a thin monofilament hook to prevent it from floating when cages were submerged. The 24 cages were randomly placed in three aerated, running sea water tanks (0.134 m ). 3 Tide and light conditions were created to approximate summer air-exposed times for algae and snails in the upper intertidal zones of British Columbia. High tide was simulated by attaching 6 pieces of cork to the outside of each cage so that it floated with 3/4 of the cage submerged for 6 hours a day (8:00-11:00 AM and 8:00-11:00 PM). This ensured that part of the cage was out of water and allowed snails to choose to be in or out of the water. Low tide conditions were achieved by taking the cages out of the water for 18 hours each day (11:00 AM-8:00 PM and 11:00 PM-8:00 AM). Light was controlled as follows: 12 h days and 12 h nights, and consequently combinations of tide and light regime, such as day/high, day/low, night/high and night/low, alternatively were achieved. I placed 75 snails, 3-4 mm in length, in each treatment cage reflecting the average natural density during summer months. After 5 days, thalli were removed, blotted, weighed and returned to their original cage. After another 5 days (10 days in total), thalli were weighed again and the experiment terminated. The grazing effect was determined by comparing algal weights between treatment (with snails) and control (without snails) groups. Data analysis employed 2x3x2 ANOVA (Snail x Algae x Time) with repeated measures on Time factor. This enabled me to interpret the data in various ways. The Snail effect, which has 2 levels, present (=treatment) and absent (=control), tested whether consumption by snails actually occurred. The Snail x Algae interaction indicates the presence or absence of feeding preference among algae. The result of Snail x Algae x Time  56 interaction indicates whether the consumption pattern or preference was consistent over the course of the feeding trial. A simple main effect as a post hoc analysis was applied when the Snail x Algae interaction was significant (Kiockars and Sax 1990).  Multiple-diet experiments  The treatment consisted of 16 cages, each containing 120 snails and about 3.0 g of food per cage in units of 1.0±0.02 g blotted wet weight of each alga. Seven cages, containing the same amount of food but no snails, were used as the control. I increased the number of herbivores to obtain more obvious visual evidence for the part of thallus actually eaten than in the single-diet experiment. This density was still within the range of natural densities in the field. Thalli of each species were attached (in the same way as the singlediet experiment) alternately around the perimeter of each cage, so that each algal species was the same distance away from the snails initially located in the center of each cage. The tidal regime and light control were the same as in the single-diet experiments. Only 14 treatment cages were used for the analysis because of mishandling of some thalli at the end of the experiment. The experiment ran for 10 days in August 1992 with a single measurement at day 10. Edibility was measured by comparing algal weight between the treatments and controls. However, since two or more foods were held in the same experimental unit (e.g., container or cage), consumption of one type of food was not independent of consumption of other types. Therefore, ANOVA or t-tests were not appropriate ways to analyze the data because the assumption of independence of response variables for univariate statistical tests (t-tests, ANOVA) was invalid (Peterson and Renaud 1989). In order to avoid problems where ANOVA was incorrectly applied (Vadas 1977, Schiel 1982, Paul et al. 1987), or where no statistic was used (Barker and Chapman 1990) in multiplediet experiments (see Peterson and Renaud 1989 for review), I used a two-sample  57 (treatment and control) Hotelling’s T 2 test with three dependent variables (algal species), a multivariate analog of the Student’s t-test. Since Hoteffing’s T 2 is distributed as F, F is used to determine critical values (Tabachnick and Fidell 1989). When results were significant by Hotelling’s T 2 test, I applied univariate t (or F) statistics to determine which variance(s) contributed most to the significant Hotelling’s T . 2  Snail behaviors and tide/light effects  During the multiple-diet experiment, snails that attached to each algal species within 15 minutes after snail introduction were counted to assess the attractiveness of algae. The result of the snails’ initial movement towards food (attractiveness) was compared with the actual amount of food eaten (edibility) at the end of the experiment to determine a correlation between these two quantities. In each of 14 cages, the 3 algae were ranked by the number of snails attached. The Kruskal-Wallis test was used on the ranked data, followed by the Kolmogorov-Smirnov test if the former test was significant. The multiple-diet experiment was used to assess potential effects of tide and light regimes on feeding behavior of the littorinid. The number of snails on each algal species was subsequently counted four times each day; one for each tide/light combination. These counts proceeded for 9 days. These data were evaluated only for those species where significant consumption(s) occurred. The reason for this is to interpret these data with  regard to the snails’ foraging behavior corresponding to the tide/light regimes. For data analysis, I used two-way ANOVA with repeated measures on both tide (2)  and light (2) factors. Tests were done for samples of 1, 4, 7 days after the beginning of the experiment. These sampling dates, although arbitrarily chosen, represent changes of snail behavior according to tide and light in the course of feeding time.  58 RESULTS  Natural densities of snails  Snail densities in the upper intertidal zones at Prasiola Point varied significantly over sampling dates (ANOVA, F(s,los)=33.336, p<O.OO1; data were square root (N+1) transformed; Table A.1). Snail densities increased during summer months (May to October) and decreased to relatively low densities at other times of the year except in February, 1992 (Fig. 3.1). Densities of snails in monospecific patches of M. cornucopiae were significantly greater than those in fucoid patches (F( )=163.790, 0 j 2 8 p.<O.OOl). This spatial variation among the monospecific patches of three dominant macrophytes was consistent throughout all sampling dates. Mean density of snails in M. cornucopiae patches reached up to 73 per 100 cm 2 in May of 1992. Densities on monospecific patches of each fucoid species were similar. The abundance of littorinids in the general habitat at Prasiola Point showed a similar seasonal pattern to those in monospecific patches (Fig. 3.1). There were, however, spatial variations in snail densities from transect to transect (F(2j32)=7.810, p=O.OOl). In addition, there was a temporal-spatial interaction in these densities (F( )=3.306, 1 , 16 32 p=O.O28), which indicated that differences in snail densities among transect lines varied over time. Descriptive results for the spatial variations are not shown in the Fig. 3.1 due to the graphic complexity.  Single-diet experiment  Snails in experimental cages significantly lowered algal biomass relative to that in control cages without snails (Table 3.1: Snail effect, p=O. 003), indicating that grazing occurred. A significant (p=0.009) Snail x Algae interaction averaged over Time indicated  59 that Littorina sp. showed a feeding preference among the three algal species (Table 3.1). A simple main effect post hoc analysis for this Snail x Algae interaction revealed that Mazzaella cornucopiae was the only species significantly grazed (t(critjcal value at p=O.05) 6 . 2 67  t(observed)= 4 . 3 50  343 t(observed)O.  for M. cornucopiae; t(observed)=O. 450 for F. distichus;  for P. li,nitata). After 5 days, the biomass of M. cornucopiae in the  control cages increased by 9% over the original biomass (Fig. 3.2). However, growth did not surpass consumption by snails in the experimental (treatment) cages; as a result, 2.6% of the original wet weight (—3.0 g) was lost. Control fucoids did not grow. Littorinids consumed algae throughout the experimental period. Weight changes in the controls were relatively constant, whereas the amount lost to feeding in the treatments after 10 days was considerably greater than after 5 days (Time x Snail interaction, p=O.OOl in Table 3.1; Fig. 3.2). The significant Snail x Algae interaction can be attributed to the Time x Snail x Algae interaction, which suggests that the pattern of preferential feeding varied with time (Time x Snail x Algae, p=O.O08). This occurred because consumption of the three species increased disproportionally in the second period (percent weight loss from day 5 to day 10; 11.1% to 19.5%, Mazzaella, 1.5% to 2.5%, Fucus, 0.6% to 2.4%, Pelvetiopsis).  Multiple-diet experiment  2 test showed significant differences in consumption among the three Hotelling’s T species provided in the multiple-diet experiment (120 snails, 10 days) (F( )=6.186, 317 pO.OOS; Fig. 3.3). The univariate F-test allowed ranking preference for the algae as follows: Mazzaella > Pelvetiopsis  >  Fucus (F( )=19.6l,p<O.OOl; 1 , 1 y3.12, 19 9 F(l,  )=2.859, p=O. 107, respectively). As in the single-diet experiment, M. 19 p=O.O93; F(l, cornucopiae was preferred over the two other species. Pelvetiopsis limitata ranked second  60 in the multiple-diet experiment, although this species was the least consumed in the singlediet experiment.  Snail behaviors and tide/light effects  Within 15 mm of starting the multiple-diet experiment, Littorina sp. movement toward a food choice showed that more herbivores chose M. cornucopiae over the two fucoids (Kolmogorov-Smirnov test, p<0.000i for Mazzaella  >  either Fucus or  Pelvetiopsis, p=O.8954. for Fucus and Pelvetiopsis; Fig. 3.4). This attractiveness of algae correlated with the results of thallus loss (edibility) in the multiple-diet experiment in terms of the same most preferred species. However, the rank order of two fucoids in the multiple-diet experiment for edibility (Pelveliopsis  >  Fucus) was reversed for attractiveness  (Fucus > Pelvetiopsis); however, the degree of difference between the two species was small in both cases (Fig. 3.4). Results of the littorine movements according to the tide/light regime are shown in Table 3.2. I used the number of snails attached to M. cornucopiae because significant grazing only occurred on this alga. More snails averaged over ‘day’ and ‘night’ were attached to M. cornucopiae at ‘low tide’ than ‘high tide’ for every sampling date (Fig. 3.5; Table 3.2, Tide effects p=O.Ol9, p=O.Ol5, p=O.OO7; dayl, day4, day7, respectively). On the other hand, light did not affect snail movement on any of these dates. In particular, snails’ movement towards M. cornucopiae was distinctive when night high tide switched to night low tide (the order of tide/light regime for each day shown in Fig. 3.5 corresponds to the actual order in the experiment). The pattern that appeared with the tidal shift from high to low at ‘night’ was not consistent during the ‘day’ time; the same pattern was shown only on day 1, not days 4 and 7 (Tide x Light interactions were significant only on days 4 and 7).  61  DISCUSSION  Repeated measures on the amount of food consumed showed a significant interaction between food preference and time. This indicates that food preference can vary with length of feeding time. In many feeding preference studies, only results from a single measurement have been considered and the variation in grazing pattern in the course of the feeding trial was typically ignored. However, a short feeding time may allow for uncomplicated estimation of consumption due to less autogenic change in the food materials used (Peterson and Renaud 1989). In a natural situation, herbivores face a variety of foods over an extended period of time. Therefore, I suggest that assessment of algal consumption over time is important to reduce the risk of misinterpretation of grazing or preference patterns. This is particularly critical when differences in preference among food choices are not large. The responses of Littorina sp. to changes in tide and light condition were not consistent throughout the experiment (Day 1, 4 and 7). Nevertheless, the greatest number of snails appeared on thalli found in the ‘low tide’ at ‘night’. Particularly there were active movements to the food when ‘high tide’ shifted to ‘low tide’ at ‘night’. However, these data are a measurement of snail behavior or movement, not a measurement for grazing. Thus, there is no direct evidence in the data for a relationship between littorinid movement and foraging. Nonetheless, a correlation between these two quantities can be found in a few other studies (Newell et al. 1971, Underwood and Chapman 1985, Watson and Norton 1987). Since only M. cornucopiae was significantly grazed, the number of snails which appeared on this alga could be partly related to the snail’s foraging activity. if snails were attached to the red alga just for the benefit of moisture during ‘low tide’, more snails would be expected to move toward the thalli in ‘day time’ (relatively greater desiccation) than in ‘night time’. However, no evidence for this was observed throughout the experiment although laboratory conditions could not completely mimic desiccation levels in  62 natural habitats. Results of the present study support McMahons (1990) results that active movement and foraging of eulittoral snails occur during tidal emergence. In addition, our results are in accordance with the feeding behaviors of two other local species, L. scutulata and L. sitkana, in which the greatest consumption of foods occurred during low tide in the early morning (Voltorina and Sacchi 1990). The results in this study consistently suggest that M. cornucopiae should be the most palatable food for Littorina sp. Results of the attractiveness trials were similar. I conclude that Littorina sp. had no consistent preference between the two fucoids. In the single-diet experiment, F. distichus was consumed slightly more than P. limitata and was selected over P. limitata for attractiveness in the multiple-diet experiment. In contrast, in the multiple-diet experiment more P. limitata was consumed than F. distichus. A variety of cues have been suggested as responsible for attracting the littorines to a particular alga. In the multiple-diet experiment, it is unlikely that the phenomenon of snails following previously laid tracts (deposited by one snail and followed by others) affects the results. If the initial foraging were random (Norton et al. 1990), then significant differences in choice are highly unlikely to occur due to “tracking. If movement is in response to chemical or other cues, tracks laid by early arrivals will be followed by later snails; presumably this same behaviour occurs on the shore. Why is the red alga chosen over the fucoids in the laboratory as well as in the field? In the present study, the littorines’ choice may have been affected by phenolic compounds, which are known to deter feeding by herbivorous snails on some marine brown algae including fucoids (Hay and Fenical 1988). Steinberg (1985) reported that F. distichus and P. limitata contained similar amounts of tissue phenolics (4.35±0.28% and 4.93±0.39% dry mass, respectively) and demonstrated that these two fucoids were least preferred by Tegulafunebralis, the black turban snail, among 13 common Phaeophyta from central California. Secondary metabolites of red algae are relatively poorly known as herbivore deterrents except for a few species (e.g., elatol in Laurencia) (see the review by Hay and  63 Fenical 1988). No studies appear to have reported on chemical deterrents in the genus Mazzaelia (previously Iridaea). The absence of chemical deterrents may explain the immediate (e.g., 15 mm) attraction of Littorina sp. to M. cornucopiae at the start of the multiple-diet experiment. Thallus toughness may be related to feeding preferences although no attempt was made to measure this characteristic. The relatively softer texture and thinner thallus of M. cornucopiae may have been one of the reasons why M. cornucopiae was selected over the more solid, thicker fucoids. This suggestion is supported by visual observations of thalli after each experiment. Only blade tops (relatively thinner than the bottom part of the blade) of M. cornucopiae were heavily eaten. A few studies on the relationship between thallus toughness and feeding preferences (Steinberg 1985, Pennings and Paul 1992) suggest that the specific herbivore determines the relative importance of plant toughness and chemical defenses as feeding deterrents. For example, larger species of littorinids, e.g., L. littorea, have been known to feed on fucoid algae (Lubchenco 1983, Barker and Chapman 1990). In addition to being the preferred food, the dense turf of M. cornucopiae may provide a better refuge for Littorina sp., especially in wave-exposed sites (Boulding and Van Alstyne 1993). Small snails may be better protected from dislodgment by moving to the bottom of Mazzaella turf during high tide rather than hiding beneath fucoid thalli which exhibit a sweeping movement. However, while Mazzaella may be a better refuge, this study indicates that Mazzaella was a preferred food. Furthermore, there were many more snails in Mazzaella beds than on Fucus and Pelvetiopsis on the shore at low tide (when all the data were collected) when there is no stress for dislodgment caused by wave action. Therefore, the higher abundance of snails on Mazzaella observed in the field is probably due to a combination of both preferred food and better habitat. The present study suggests that the summer decline of the M. cornucopiae population (Fig. 5.1) is partially caused by littorinid herbivory. Desiccation may also reduce the abundance of this red alga. Olson (1985, 1992) observed that blades of M.  64 cornucopiae occurring on the upper margin of its habitats in Oregon, U.S.A. were usually bleached, and a portion of bleached blade disappeared at the end of summer. She argued that the upper limit of this population was affected by a combination of desiccation and herbivory by limpets. I observed that bleached blades commonly occurred in the upper zone whereas loss of non-bleached thalli occurred throughout the population. It has generally been assumed that the importance of physical factors is higher in the upper intertidal shore and that of biological factors is greater in the lower shore (Castenholz 1961, Connell 1972, Chapman 1973, Menge 1978, Underwood 1980). The validity of this paradigm has been questioned (Dayton 1975, Cubit 1984, Underwood and Denley 1984, Farrell 1991). My data strongly suggest that the most abundant herbivore, Littorina sp., could decrease Mazzaella biomass which may, in turn, affect interactions among the three algal dominants. For example, the dense turf of Mazzaelkx prevents recruitment of both Fucus and Pelvetiopsis in this habitat (Chapter 2). Summer decline of the Mazzaella population and its slow recovery over the winter may open up space for fucoid recruitment. Active recruitment of both Fucus and Pelvetiopsis during fall (i.e., October; Fig. 5.1) may assist these two species in maintaining their place in this community.  65 REFERENCES Barker, K. M., Chapman, A.R.O. (1990). Feeding preferences of periwinkles among four species of Fucus. Mar. Biol. 106: 113-118. Behrens, S. (1971). The distribution and abundance of the intertidal prosobranchs Littorina scutulata (Gould 1849) and L. sitkana (Phillipi 1845). M.Sc. Thesis. Univ. of British Columbia. Vancouver, British Columbia. Boulding, E. G., Van Aistyne, K. L. (1993). Mechanisms of differential survival and growth of two species of Littorina on wave-exposed and on protected shores. J. Exp. Mar. Biol. Ecol. 169: 139-166. Boulding, E. G., Buckland-Nicks, J., Van Alstyne, K. L. (1993). Morphological and allozyme variation in Littorina sitkana and related Littorina species from the Northeastern Pacific. Veliger 36: 43-68. Buckland-Nicks, J., Chia F. S., Behrens, S. (1973). Oviposition and development of two intertidal snails, Littorina sitkana and Littorina scutulata. Can. 3. Zool. 51: 359365. Castenholz, R. W. (1961). The effect of grazing on marine littoral diatom populations. Ecology 42: 783-794. Chapman, A. R. 0. (1973). A critique of prevailing attitudes on the control of seaweed zonation on the sea shore. Bot. Mar. 16: 80-82. Connell, J. H. (1972). Community interactions on marine rocky intidal shores. Ann, Rev. Ecol. Syst. 3: 169-192. Connell, J. H. (1975). Some mechanisms producing structure in natural communities: a model and evidence from field experiments. In: Cody, M. L., Diamond, 3. (eds.) Ecology and ecolution of communities. Belknap, Cambridge, p. 460-490. Cubit, J. D. (1984). Herbivory and the seasonal abundance of algae on a high intertidal rocky shore. Ecology 65: 1904-1917. Dayton, P. K. (1975). Experimental evaluation of ecological dominance in a rocky intertidal algal community. Ecol. Monogr. 45: 137-159. Denley, E. 3., Dayton, P. K. (1985). Competition among macroalgae. In: Littler, M. M., Littler, D. S. (eds.) Handbook of phycological methods. Ecological field methods: Macroalgae. Cambridge University Press, Cambridge, p. 511-530. Farrell, T. M. (1991). Models and mechanisms of succession: an example from a rocky intertidal community. Ecol. Monogr. 61: 95-113. Hawkins, S. J., Hartnoll, R. G. (1983). Grazing of interdal algae by marine invertebrates. Oceanogr. Mar. Biol. Ann. Rev. 21: 195-282. Hay, M. E., Fenical, W. (1988). Marine plant-herbivore interactions: The ecology of chemical defense. Ann. Rev. Ecol. Syst. 19: 111-145.  66 Klockars, A. J., Sax, G. (1990). Multiple comparisons. Sage Publications, Inc. Newbuny Park, California. Lubchenco, 3. (1983). Littorina and Fucus: Effects of herbivores, substratum heterogeneity and plant escapes during succession. Ecology 64: 1116-1123. McMahon, R. F. (1990). Thermal tolerance, evaporative water loss, air-water oxygen consumption and zonation of intertidal prosobranchs: a new synthesis. Hydrobiologia 193: 241-260. McMahon, R. F., Russell-Hunter, W. D. (1977). Temperature relations of aerial and aquatic respiration in six littoral snails in relation to their vertical zonation. Biol. Bull. 152: 182-198. Mastro, E., Chow, V., Hedgecock, D. (1982). Littorina scutulata and Littorinaplena: sibling species status of two prosobranch gastropod species confirmed by electrophoresis. Veliger 24: 239-246. Menge, B. (1978). Predation intensity in a rocky intertidal community. Oecologia 34: 1-16. Newell, R. C. (1970). Biology of intertidal animals. Lagos Press, London. Newell, R. C. (1979). Biology of intertidal animals. Marine Ecological Surveys Ltd., Faversham, Kent. Newell, R. C., Pye, V. I., Ahsanullah, M. (1971). Factors affecting the feeding rate of the winkle Littoriiw littorea. Mar. Biol. 9: 138-144. Nicotri, M. E. (1980). Factors involved in herbivore food preference. 3. Exp. Mar. Biol. Ecol. 42: 13-26. Norton, T. A., Hawkins, S. I., Manley, N. L., Williams, G. A., Watson, D. C. (1990). Scraping a living: a review of littorinid grazing. Hydrobiologia 193: 117-138. Olson, A. M. (1985). Early succession in beds of the red alga, Iridaea cornucopiae Post. & Rupr. (Gigartinaceae): Alternate pathways. M.S. Thesis. Oregon State University. Corvallis, Oregon. Olson, A. M. (1992). Evolutionary and ecological interactions affecting seaweeds. Ph.D. Thesis. Oregon State University. Corvallis, Oregon. Paul, V. J., Littler, M. M., Littler, D. S., Fenical, W. (1987). Evidence for chemical defense in tropical green algae Caulerpa ashrneadii (Caulerpaceae: Chiorophyta): isolation of new bioactive sesquiterpenoids. 3. Chem. Ecol. 13: 1171-1185. Pennings, S.C., Paul, V. 3. (1992). Effect of plant toughness, calcification, and chemistry on herbivory by Dolabella auricularia. Ecology 73: 1606-16 19. Peterson, C. H., Renaud, P. E. (1989). Analysis of feeding preference experiments. Oecologia 80: 82-86. Reid, D. G., Golikov, A. N. (1991). Littorina naticoides, new species, with notes on other smooth-shelled Littorina from the Northeastern Pacific. Nautilus 105: 7-15.  67 Roble, C. D., Cubit, J. (1981). Influence of biotic factors in an upper intertidal community: dipteran larvae grazing on algae. Ecology 62: 1536-1547. Schiel, D. R. (1982). Selective feeding by the echinoid, Evechinus chioroticus, and the removal of plants from subtidal algal stands in Northern New Zealand. Oecologia 54: 379-388. Steinberg, P. D. (1985). Feeding preferences of Tegukifunebralis and chemical defenses of marine brown algae. Ecol. Monogr. 55: 333-349. Tabachnick, B. 0., Fidell, L. S. (1989). Using multivariate statistics. 2nd edition. Harper & Row Publishers, New York. Underwood, A. J. (1979). The ecology of intertidal gastropods. Adv. Mar. Biol. 16: 111210. Underwood, A. J. (1980). The effect of grazing by gastropods and physical factors on the upper limits of distribution of intertidal macroalgae. Oecologia 46: 201-213. Underwood, A. J., Denley, E. J. (1984). Paradigms, explanation and generalizations in models for the structure of intertidal communities on rocky shores. In: Strong, D. R., Simberloff, D., Abele, L. G., Thistle, A. (eds.) Ecological Communities: Conceptual Issues and the Evidence. Princeton University Press, New Jergy, pp. 151-180. Underwood, A. J., Chapman, M. G. (1985). Multifactorial analyses of directions of movement of animals. J. Exp. Mar. Biol. Ecol. 91: 17-43. Vadas, R. L. (1977). Preferential feeding: an optimization strategy in sea urchins. Ecol. Monogr. 47: 337-371. Voltorina, D., Sacchi, C. F. (1990). Field observatios on the feeding habits of Littorina scutulata Gould and L. sitkana Philippi (Gastropoda, Prosobranchia) of southern Vancouver Island (British Columbia, Canada). Hydrobiologia 193: 147-154. Watson, D. C., Norton, T. A. (1985). Dietary preferences of the common periwinkle, Littorina littorea (L). J. Exp. Mar. Biol. Ecol. 88: 193-211. Watson, D. C., Norton, T. A. (1987). The habit of feeding preferences of Littorina obtusata (L.). and L. inariae Sacchi et Rastelli. J. Exp. Mar. Biol. Ecol. 112: 6172. Wilkinson, L., Hill, M. A., yang, E. (1992). SYSTAT: Statistics, Version 5,2.1 Edition, Systat, Inc., Evanston. Winer, B. J. (1971). Statistical principles in experimental design. McGraw-Hill Book Company, New York.  68 Table 3.1 Single-diet experiments. Results of three-way ANOVA, a Snail x Algae x Time (2x3x2), with repeated measures on the Time factor.  Source  if  Between Subjects Snail  .399  1  12.00  .003  Algae  .247  2  7.41  .005  Snail x Algae  .207  2  6.22  .009  Error  .033  18  Time  .032  1  14.68  .001  TimexSnail  .035  1  15.93  .001  Time xAlgae  .007  2  2.95  .078  xAlgae  .014  2  6.30  .008  Error  .002  18  Within Subjects  Time x Snail  69 Table 3.2 Effects of light and tide on snail behaviors. Results of two-way ANOVA with repeated measures on both light (2) and tide (2) factors. A separate analysis was applied on each sampling date (Days 1, 4, 7) during the multiple-diet experiment.  Source  Within Subjects Light  E  DAY 1 1  Error  33.063 42.196  15  Tide  297.563  1  Error  43.229  15  LightxTide Error  30.250  1  21.317  15  Within Subjects  .562  1  Error  21.563  15  Tide  72.250  1  Error  9.650  15  256.000  1  16.467  15  Light  21.391  1  Error  24.724  15  Tide  153.141  1  Error  16.007  15  221.266  1  18.199  15  Error Within Subjects  Lightx Tide Error  .390  6.883  .019  1.419  .252  .026  .874  7.487  .015  15.547  .001  .865  .367  9.567  .007  12.158  .003  DAY 4  Light  Lightx Tide  .784  DAY 7  0  10  20  30  40  50  AU CC 1991 NC  FE AP  MA  JL 1992  CC  DE  MAZ  VJFUC  PEL  •GEN  Fig. 3.1 Natural densities of snails at Prasiola Pt. Data are means +SE of 5 monospecific patches of each alga and of 18 randomly selected piots from transect lines in the general habitat (3 transects x 6 points). GEN for the general habitat, MAZ for M. cornucopiae, FUC for F. distichus, PEL for P. limitata.  U  I C’)  W  Cl)  0  0  60  70  80  0  71  15  DAY5  10 5  i—  w  -10  I] CONTROL TREATMENT  -15 15  DAYIO  10  w  MAZ  PEL  FUC  Fig. 3.2 Single-diet experiments. 3.0±0.02 g of thallus was used per cage. Positive values indicate weight gain. Negative values indicate weight loss. Data are means +SE of five treatments and three controls. MAZ, M. cornucopiae, FUC, F distichus, PEL, P. la.  72  0.20 IZI CONTROL TREATMENT 0.15 Cl) U)  0  -J  0.10 (!3 w 0.05  0.00 MAZ  PEL  FIJC  Fig. 3.3 Results of the multiple-diet experiment. The comparisons of weight loss (g) between treatments and controls and among algal species. 1±0.02 g of each algal species were placed making -3.0 g (total) of food per cage. Data are means +SE of fourteen treatments and seven controls. MAZ, M. cornucopiae, FUC, F. distichus, PEL, P. limitata.  73  30  U)  w  20  I— U)  z  w -J  z  I  U)  0  _i_  MAZ  I  I  PEL  FUC  Fig. 3.4 Behavioral choice (attractiveness) of Littorina sp. towards the three algal species. Densities of snails attached to the thalli of each algal species 15 mm after the multiple-diet experiment started. Data are means +SE of sixteen replicates. See Fig. 3.2 for caption comments.  74  Night/Low Night/High Day/Low Day/High  30  g  20  I— U)  z  w 0  -J  z Cl)  10  0 DAY1  DAY 4  DAY7  Fig. 3.5 Tidal and light effects on snail movement towards food during the multiple-diet  experiment. The data shown are the number of snails attached to M. cornucopiae for every tidal and light condition. Tide/light combinations are presented in the same order as simulated in the lab. Data are means +SE of sixteen replicates.  75 CHAPTER 4  THE EFFECT OF BARNACLES AND LIMPETS ON ALGAL SUCCESSION  INTRODUCTION  When a disturbance opens up a free space and space-related resources in a community, some organisms quickly invade the space and use the liberated resources, while other organisms (species) replace the early colonists and come to predominate later. This process of species introduction and replacement in a community over time is termed succession. The mechanisms involved in succession which determine the actual pathway between the early and the later colonists have long been of interest to ecologists. Early views of succession focused on the competitive interactions between species in different successional stages and the influence of physical environments on colonists (Colinvaux 1973, Horn 1974). Later, Connell and Slatyer (1977) proposed three general models of succession by considering the net effect of early colonists on the establishment of the later successional species: 1) the early colonists enhanced the establishment of the later colonists (facilitation); 2) the early colonists do not affect establishment of the later colonists (tolerance); 3) the early colonists slow the settlement or growth of the later colonists (inhibition). These models have provided valuable guides for researchers although many complexities have been revealed in natural systems; these include seasonality of demographic characteristics of component species (Turner 1983), indirect interactions (Dethier and Duggins 1984, Dungan 1986) and influences of consumers (Lubchenco and Gaines 1981, Hawkins and Hortnoll 1983, Farrell 1991). As initially pointed  Out  by Connell and Slatyer (1977), the influence of consumers  is of critical importance to the course of succession, and several recent studies of  76 succession have involved manipulation of consumers (reviewed by Lubchenco and Gaines 1981, Hawkins and Hartnoll 1983). These studies indicated that consumers often have a strong influence on the rate of succession, but that succession could be either accelerated (Lubchenco and Menge 1978, Sousa 1979, 1984, Day and Osman 1981, Lubchenco 1983), unaffected (Turner 1983, Jernakoff 1985) or slowed (Dayton 1975, Sousa et al. 1981, Peer 1986, Farrell 1991). Farrell (1991) recently formulated a general predictive model (Fig. B. 1) of the effect of consumers on the rate of succession. In his model, the rate of succession is determined by a combination of two factors: Connell and Slatyer’s (1977) classic models of facilitation, tolerance and inhibition, and the successional status of the species whose abundances are reduced by consumers. According to Farrell’s model, succession can be accelerated only if consumers reduce early successional species in the inhibition model (one of the nine cases), and succession will be slowed in six other cases, leaving two cases with no effect of the consumer on the rate of succession. Farrell (1991) also provided examples of the tolerance and facilitation models in a high intertidal community on Oregon coasts dominated by Balanus-Pelvetiopsis, in which the major consumer, the Lottia spp. (acmaeid limpets), slowed the community succession. In this chapter, I test some cases in Farrell’s model. Barnacles and limpets together have been frequently investigated for their role in shaping the intertidal community, since they are both common and they coexist. The interaction dynamics between barnacles and limpets have sometimes been studied separately from other components of the system (Choat 1977, Denley and Underwood 1979, Creese 1982, Underwood et al. 1983). However, more recently, the three-way interactions of barnacles-limpets-algae have received increasing attention (Jernakoff 1983, Dethier and Duggins 1984, Sousa 1984, Dungan 1986, Van Tamelen 1987, Farrell 1988, 1991). Facilitation of algal recruitment by barnacles has typically been reported as resulting from either of two processes. First, substratum alteration caused by barnacles could enhance algal propagule settlement because barnacles provided both a secondary  77 substratum and a shelter from desiccation stress (Norton 1983, Farrell 1991). This is an example of a direct positive interaction of barnacles on algae. Secondly, barnacles, as they grow to a large size, may inhibit the foraging activities of limpets, providing algae with refugia from their herbivores (Hawkins 1981, Hawkins and Hartnoll 1983, Dungan 1986, Van Tamelen 1987, Little et al. 1988). The effect of barnacles on algae in this case is an indirect positive interaction through limpets. Investigations on multiple interactions have often provided much information on the variety of mechanisms of species interactions (such as indirect interactions); such studies are particularly useful for elucidating the mechanisms of succession. In particular, some recent work has attempted to separate indirect effects among the component species from the complex barnacle-limpet-algae interactions (Dungan 1986, Van Tamelen 1987, Farrell 1991). Indirect interactions occur when the effect of one species on another is mediated by a third species. For example, Van Tamelen (1987) found that limpets indirectly facilitated barnacles by grazing algae which had a deleterious effect (e.g., overgrowth) on barnacles. In natural communities, this type of interaction has sometimes been reported both among organisms in the same trophic level (e.g., Dethier and Duggins 1984, Kastendiek 1982) and among those at different trophic levels (e.g., Underwood et al. 1983, Dungan 1986, Van Tamelen 1987, Farrell 1991). Although researchers have given increased attention to the mechanisms of indirect interactions, their influence in conjunction with the mechanisms of succession, especially algal succession, is not well known (but see Lubchenco 1983). In the context of barnacle limpet-algal interactions, both Dungan (1986) and Van Tamelen (1987) provided insufficient information to evaluate the mechanisms of replacement of different algal species in different successional stages. This problem also applies in part to Farrell’s (1991) study, in that he focused on succession of different barnacle species. The aim of this chapter is to assess the mechanisms of algal succession and the direct and indirect interactions producing those mechanisms in the barnacle-limpet-algal  78 community on the upper intertidal shore. A factorial design with barnacles and limpets removed singly and in combination was used to test the following hypotheses (1 and 2) and answer questions (3 and 4): 1) Hypothesis I: limpets, as herbivores, slow the rate of algal succession; 2) Hypothesis II: facilitation of algal settlement by barnacles is a direct positive interaction which is not associated with the presence of the limpet; 3) Are there any indirect interactions between the component organisms? if there are, what are the mechanism(s) of the interaction(s)?; 4) Does the pattern of algal succession in this community fit any of the models proposed by Connell and Slatyer (1977)?  METHODS  Experimental design  The experiment was initiated at Nudibranch Point in July 1991. Sixteen 25x25 cm plots were cleared using a pneumatic hammer/drill (Chicago Pneumatic 9AK Handril) with a chisel bit and a SCUBA tank as an air source. The first stage in the air-pathway was adjusted down to 80 psi for the drill. For clearing plots, all organisms (including holdfasts, barnacles, crustose algae, etc.) on the substratum were completely removed by chiseling off the rock surface. Care was taken to maintain the level of the cleared surfaces at similar heights to the adjacent uncleared substratum to minimize significant modification of microhabitats. The denuded piots were rechecked the next day to remove any organisms that had been overlooked. Each corner of the plot was marked by hammering in concrete nails which were sometimes supported by cement. The experiment consisted of a randomized block design with four blocks. Each block contained a replicate of the following four treatments: (1) barnacles and limpets present (+bar +lim), (2) barnacles removed, limpet present (-bar +lim), (3) barnacle present, limpets removed (+bar -lim), and (4) barnacles and limpets removed (-bar -lim).  79 These four treatments constituted a complete factorial design. The immigration of limpets (Lottia spp. ) from the surrounding areas was greatly reduced by applying 2.5-3 cm wide copper-paint (57.7% metallic copper, Pettit Paint Ltd., Borough of Rockaway, New Jersey) barriers, which limpets did not cross (Cubit 1975, personal observation). Since previous studies employing this paint reported little or no effect from the copper on algal recruitment or growth (Sousa 1979, Lubchenco and Cubit 1980, Slocum 1980, Robles and Cubit 1981, Farrell 1988,1991), this method has been frequently used to exclude limpets in intertidal studies. Any invading limpet was manually removed from the exclusion plots at each sampling date (usually at intervals of 8 weeks). However, it was observed that the major littorinid, Littorina sp., of this habitat (Chapter 3) was free to invade the experimental plots. Copper paint was ineffective as a barrier against some gastropod snails such as Tegulafunebralis, Littorina scutulata, and L. plena (Mastro et al. 1982). Newly recruited barnacles in the exclusion plots were also removed manually at each sampling date by scraping or squashing individuals with a small flat screw driver (4 mm in width). Caution was taken not to affect other existing organisms such as benthic diatoms and holdfasts of macroalgae during the removal of barnacles. The abundance of recolonizing algae was measured bimonthly within the central 20x20 cm area of each 25x25 cm plot to minimize potential marginal effects. A quadrat (20x20 cm) with lxi cm subplots marked off within the quadrat frame (using monofilament lines) was used to measure percent cover. Percent cover of each species was obtained by counting the number of the subplots in which thalli occupied >50% of its area. The accuracy and repeatability of this technique has been tested by Dethier et al. (1993) and they reported that this method was generally more accurate than the random-point-quadrat method. Algal cover was enumerated only for those thalli with their holdfasts within the plots. Densities of barnacles and limpets were subsampled from randomly selected 2x2 cm subplots and then estimated for their density per 20x20 cm plot.  80 Data Analysis  I used an analysis of variance (ANOVA) to analyze for treatment effects on each of four selected sampling dates (October 1992, April 1993, December 1993, and August 1994), followed by Tukey’ HSD tests if ANOVA results were significant. These sampling dates were selected to represent the treatment effects on each algal species in all seasons. As well, by performing only four tests, the probability of a type I error was not greatly increased. The Bonferroni adjustment on the P value (p=O.Ol 25 at alpha =0.05) was used for each of the four tests. If significant heterogeneity of variances existed among the ANOVA cells (tested by using the Fmax test and Cochran’s tests, Winer 1971), percent cover data were arcsin-transformed and density data (for barnacles and limpets) were logtransformed prior to analysis. On some sampling dates for certain algal species (as noted in the RESULTS), ANOVA could not be performed because the transformation did not reduce the heteroscedasticity to a nonsignificant (alpha=0.05) level. For these cases, I used non-parametric statistics, a Kruskal-Wallis test and Mann-Whitney U test (Wilkinson et al. 1992) on untransformed data. A test for normality was not performed because of an insufficient sample size for the test (n=4). For ease of interpretation, untransformed data were plotted in all figures. All data analyses were performed using SYSTAT, version 5.2.1 for Macintosh (Wilkinson et al. 1992).  RESULTS  The effects of factorial removal of barnacles and limpets on algal recovery are shown in Fig. 4.1 (for the treatment effects on each algal species) and Fig. 4.2 (for the algal species interactions under each treatment). The recolonization and abundance of ephemeral algae was significantly different between the limpet-excluded and limpet present plots. The herbivore heavily grazed on ephemeral algae and the grazing effect was  81 significant on the two fall and winter sampling dates (ANOVA, p=O.000 for October 1992,  p=O.OO3 for December 1993) but not in the summer (August 1994, p=O. 184) and spring (April 1993, p=O.O22) dates (Fig. 4.1; A). The effect of limpets on ephemeral algae subsequently influenced algal succession. In the absence of limpets, the abundance of ephemeral algae dramatically increased within two months in the second fall (October 1992) in both limpet- and barnacle/limpet-excluded plots (Fig. 4.2; C and D), and this high percent cover of ephemerals was maintained until April 1993 (Fig. B.2). During this period there were low percent covers of fucoids, which started to colonize (summer, 1993) when the ephemerals declined apparently due to seasonal events. No evidence for such an interaction between fucoids and ephemeral algae was found in the two limpet-present treatments (Fig. 4.2; A and B). In December 1993, the ephemeral algae (mainly diatoms) became abundant again. However, the dominance of ephemeral algae was ended by Fucus distichus the following spring (April 1994 for the limpet-excluded plots and June 1994 for the barnacle/limpet-excluded plots) (Fig. 4.2; C and D). In the absence of limpets, it took almost 3 years for Fucus distichus, the late successional species, to dominate the space. This pattern can be compared to the two limpet-present treatments (Fig. 4.2; A and B), where fucoid domination was apparent in a year. In the control plots (+bar +lim), both Fucus and Pelvetiopsis colonized at an early stage (e.g., October 1992) and outcompeted other species (e.g., ephemerals, Mazzaella) for the entire experimental peiiod (Fig. 4.2; A). In the barnacle-excluded plots (-bar +lim), only Pelvetiopsis limitata was common and all other algae remained at a very low percent cover for the entire experimental period (Fig. 4.2; B). There was no obvious barnacle effect on algal abundance in the absence of limpets (Fig. 4.2; C and D). However, in the presence of limpets, barnacles facilitated the colonization of Fucus distichus; this alga had only a very low percent cover under similar conditions but without barnacles (Fig. 4.1; C; Mann-Whitney U tests with +B+L and -B+L, p=O.Oll for October 1992, p=O.O2O for April 1993, p=O.Ol’7 for December 1993,  82  pO.O2l for August 1994).  In the presence of limpets, Pelvetiopsis limitata was more  abundant in the barnacle-excluded plots than in the piots with barnacles (Fig. 1; D; ANOVA, p<O.O 125 on August 1994, otherwise p>O.O 125). The effect of barnacles on limpets and the effect of limpets on barnacles are shown in Fig. 4.3 and Table 4.1. Barnacle densities were greater in the limpet excluded plots (-i-bar -Jim) than in the limpetpresent plots (+bar +lim) at the later successional stages (Independent t-tests, t(df6)=3.0 15, p=O.O24 for August 1993, t(df=6)=-2.988, p=O.024. for October 1993). However, limpets were not significantly affected by the presence or absence of barnacles even though their number was slightly higher in the presence of barnacles (Table 4.1 and Fig. 4.3). The abundance of Mazzaella cornucopiae remained at less than 2% cover for the experimental period, thus there was no treatment effect for this red alga (Fig. 4.1; B). Among the ephemeral algae, diatoms were initially the most abundant algae in the limpetexcluded plots, but were not seen when limpets were present (Table 4.2). A pattern similar to that for diatoms was observed for Urospora spp., indicating that limpets effectively grazed on microscopic and delicate filamentous algae.  DISCUSSION  Models and the herbivore’s role in algal succession  In this community, limpets (mostly Lottia digitalis) selectively limited the abundance of ephemeral algae, and can be considered as a keystone species which determines the sequence of algal succession. Dominant algae in the different treatments and different successional stages are summarized in Fig 4.4. In the absence of limpets, succession began with abundant settlement of ephemeral algae, and ended with Fucus distichus. However, in the presence of limpets, both Fucus and Pelvetiopsis appeared at the early stages and the dominance of fucoids continued to the later stages. Limpets  83 grazing on ephemeral algae in this study speeded up the early successional sequence by releasing fucoids from inhibition by ephemeral species. Therefore, both Fucus and Pelvetiopsis predominated in the community from the early stages, and consequently the succession by-passed the early colonist (ephemeral algae) stage. Apparently, there was no species replacement in this process except for the slight dominance of Pelvetiopsis as it replaced Fucus later in time (Fig. 4.2; A). Hypothesis I was rejected and succession of this community was accelerated by the herbivores. However, in Farrell s (1991) Balanus t Pelvetiopsis community (Facilitation Model) in Oregon, the limpets slowed the rate of succession by negatively affecting a small barnacle (Chthamalus dalli) which Tacilitated’ the larger and the later successional barnacle (Balanus glandula). In the high intertidal community at Nudibranch Point, early colonists (ephemeral algae) inhibited the settlement of later colonists (fucoids), typical of the Inhibition Model proposed by Connell and Slatyer (1977). This inhibition caused a considerable delay for species replacement between ephemeral algae (e.g., diatoms, Porphyra spp.) and fucoids (especially Fucus), but the delay disappeared in the presence of the herbivores (limpets). Ephemeral algae often cause inhibition of algal succession in rocky intertidal communities (Sousa 1979, 1984, Robles and Cubit 1981, Lubchenco 1983, Breitburg 1984). Sousa (1984) reported a similar result among a similar array of algal species in which the rapid increase of Ulva, diatoms and Urospora in the early successional stage (e.g., for 12 months after disturbance) inhibited the settlement of some later colonists such as Fucus, Mastocarpus and Pelvetiopsis. This occurred on a high shore of northern California when limpets were excluded. Lubchenco (1983) also reported that in the mid-intertidal of a New England rocky shore, in the absence of Littorina littorea (a herbivorous snail), ephemeral algae such as Ulva, Enteromorpha and Porphyra inhibited the appearance of the later successional species, e.g., Fucus. The Inhibition model represents the dominant type of succession in many marine habitats (Connell and Slatyer 1977, Sousa 1979, Breitburg é  84 1984); experimental evidence for the Tolerance Model has been rarely reported (but see Farrell 1991). However, facilitation was originally thought to occur as a result of early successional species altering the physical environment in ways favorable to later successional species (Connell and Slatyer 1977). For this reason Connell and Slatyer suggested that facilitation might be more frequent in harsh physical environments. In high intertidal habitats, this paradigm was supported by Farrell (1991) in a Balanus-Pelvetiopsis community. He found that the early successional barnacle, Chthamalus dalli, facilitated the later successional barnacle, Balanus glandula. There was no evidence for facilitation by ephemeral algae for fucoids during succession in my study. However, it is worth noting that inhibition by ephemeral algae was seasonal and the intensity of inhibition seemed much reduced in summer, because of both the life history characteristics of ephemeral algae and strong desiccation (some ephemeral species (e.g., Porphyra spp.) persisted in the lower part of my study area). Nevertheless, this study adds a second example for the inhibition model of succession occurring in the high intertidal zone, together with Sousa’s work. The results of my study show a different process occurring in a habitat similar to that studied by Farrell. This difference probably occurred because Chthamalus was rare in my study site and ephemeral algae were rare in Farrell’s site, even though tidal heights in both studies are similar. In the high intertidal community at Nudibranch Point, algal succession follows an inhibition model, which may be rare in the physically stressful high intertidal (Connell and Slatyer 1977). The different results in the rate and the sequence of succession in the presence or absence of limpets indicates that the consumption of algae by limpets has a significant effect in this habitat. Thus, this evidence is another example of significant biological interactions occurring in the high intertidal habitat, others of which have been shown elsewhere in this thesis.  85 Direct and indirect interactions during succession  The results of removing barnacles and limpets brought to light complex interactions among the component organisms related to the successional process. The major direct and indirect interactions affecting algal succession are shown in Fig. 4.5. Limpet’s grazing on ephemeral algae constituted a direct negative interaction. In addition, barnacles enhanced Fucus settlement; this is a direct positive interaction (facilitation). However, the barnacle effect in general varied with the presence or absence of limpets. In the absence of limpets, there was no significant barnacle effect on algal abundance. When limpets were present, however, barnacles significantly facilitated the settlement and growth of F. distichus. In contrast, the abundance of F. distichus was low in the barnacle-excluded plots and P.  limitata was much more abundant. My results rejected the second hypothesis (that barnacle facilitation of algal settlement is a direct positive interaction which is not associated with the presence of limpets). If limpet grazing was more active in the barnacle-excluded plots, the high abundance of Pelvetiopsis implied that this alga was probably the less preferred prey of the other species, especially compared to Fucus. This speculation is based on indirect evidence (Fig. 4.5). The reason for the different responses of the two fucoids to barnacles, or indirectly to limpets, was not explored in my study. The positive effect of limpets on fucoids is an indirect facilitation mediated by ephemeral algae (Fig. 4.5). This indirect interaction may occur because the limpets prefer to eat ephemeral algae rather than fucoids, but I did not test this preference directly. However, some evidence that the small acmaeid limpet (Lottia digitalis) selectively grazed on delicate filamentous, membranous algae and benthic diatoms (over larger perennial macroalgae or even larger ephemerals such as Porphyra spp.) has been documented (Castenholz 1961, Nicotri 1977, Cubit 1984, Farrell 1988). In my study, the L digitalis was also more effective on diatoms and Urospora spp. than other ephemerals (Table 4.2).  86 A similar mechanism of indirect interaction has been reported by Lubchenco (1983), but for a different herbivore, Littorina littorea. However, patterns of direct and indirect interactions among barnacles-herbivoresalgae reported in other studies do not show a consistent pattern. The results of some selected studies (e.g., Dethier and Duggins 1984, Dungan 1986, Van Tamelen 1987) are summarized in Fig. B.3. Dungan (1986) found that with similar organisms as in my study, barnacles indirectly facilitated algae by preempting space required by limpets which grazed on the algae. Van Tamelen’s (1987) results were different from those of Dungan (1986) in that algae (in his term, both micro- and macro-algae) had a deleterious effect on barnacles by reducing their growth rate and increasing mortality; however, limpets provided an indirect benefit to barnacles by reducing algae. Mechanisms of indirect interactions can be much more complicated by including different consumers (Vandermeer 1980, Dethier and Duggins 1984) and/or different prey (Lubchenco 1983, Dethier and Duggins 1984, this study). Dethier and Duggins (1984) demonstrated an example of complex interactions with two consumers (chiton and limpet) and two prey (diatoms and macroalgae). They found that a chiton (Katharina tunicata) indirectly enhanced limpet abundance by reducing macroalgae which competitively inhibited benthic diatoms. The consequent increase of diatoms (caused indirectly by the chiton) provided both food and habitat for the specialist consumer limpets. For the interaction between barnacles and limpets, this study suggested that limpets apparently had a deleterious effect on barnacles, but barnacles had no effect on limpets. Lower densities of barnacles shown in the +bar +lim plots and in the +bar -lim plots (October 1992 April 1993 in Fig. 4.3) could be the result of limpets bulldozing juvenile -  barnacles (Dayton 1971, Underwood et al. 1983) and due to the seasonal increase of ephemeral algae which might overgrow and reduce barnacles (Denley and Underwood 1979, Underwood et al. 1983, Farrell 1988, 1991, personal observation), respectively.  87 Differences in the role of each component organism in barnacle-limpet-algal assemblages found in previous studies and my study are primarily due to the heterogeneity of habitats and communities. For example, predominant algal species were different in each study cited: Ralfsia crusts in Dungan’s, Ulva and Enteromorpha in Van Tamelen’s, Hedophylluin and diatoms in Dethier and Duggin&. Tidal heights were also different, suggesting that physical conditions are also different. Foster (1990) has reminded us not to generalize results from one region to other regions that are less well described and where experiments have not been repeated. Therefore, conclusions about direct and indirect interactions among even similar component species may be limited in their applicability to that local habitat; this suggests the necessity of experimental replication over extended geographical areas. Nevertheless, the evidence for indirect interactions in this system provides ecologically relevant information, in that the herbivores differently affected algae in different successional stages and thus facilitated the succession rate by reducing the presence of the early successional group. This mechanism resulted in a “by-passing” of the early colonist stage. The role of barnacles and limpets in this Mazzaeila-Fucus-Pelvetiopsis community was viewed in the context of algal succession. It was unfortunate that the abundance of Mazzaella was too low to evaluate its interaction with the two invertebrates. The reason for the low abundance of Mazzaella in this experiment will be explained in Chapter 5. Interaction dynamics between barnacles and limpets and algae shown here will be applied to predict the patterns of community dynamics in Chapter 6.  88 REFERENCES Breitburg, D. L. (1984). Residual effects of grazing: inhibition of competitor recruitment by encrusting coralline algae. Ecology 65: 1136-1143. Castenholz, R. W. (1961). The effect of grazing on marine littoral diatom populations. Ecology 42: 783-794. Choat, I. H. (1977). The influence of sessile organisms on the population biology of three species of acmaeid limpets. J. Exp. Mar. Biol. Ecol. 26: 1-26. Colinvaux, P. A. (1973). Introduction to ecology. Wiley, New York. 621 pp. Connell, J. H., Slatyer, R. 0. (1977). Mechanisms of succession in natural communities and their role in community stability and organization. Am. Nat. 111: 1119-1144. Creese, R. G. (1982). Distribution and abundence of acmaeid limpet Patelloida latistrigata, and its interaction with barnacles. Oecologia 52: 85-96. Cubit, 1. (1975). Interactions of seasonally changing physical factors and grazing affecting high intertidal communities on a rocky shore. Ph.D. Thesis. University of Oregon. Eugene, Oregon Cubit, J. D. (1984). Herbivory and the seasonal abundance of algae on a high rocky intertidal shore. Ecology 65: 1904-1917. Day, R. W., Osman, R, W. (1981). Predation by Petiria miniata (Asteroidea) on bryozoans: prey diversity may depend on the mechanism of succession. Oecologia 51: 300-309. 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. Dayton, P. K. (1975). Experimental evaluation of ecological dominance in a rocky intertidal algal community. Ecol. Monogr. 45: 137-159. Denley, E. J., Underwood, A. J. (1979). Experiments on factors influencing settlement, survival and growth of two species of barnacles in New South Wales. J. Exp. Mar. Biol. Ecol. 36: 269-293. Dethier, M. N., Duggins, D. 0. (1984). An “indirect commensalism” between marine herbivores and the importance of competitive hierarchies. Am. Nat. 124: 205-219. Dethier, M. N., Graham, E. S., Cohen, S., Tear, L. M. (1993). Visual versus randompoint percent cover estimation: ‘objective’ is not always better. Mar. Ecol. Prog. Ser. 96: 93-100. Dungan, M. L. (1986). Three-way interactions: barnacles, limpets and algae in a Sonoran Desert rocky intertidal zone. Am. Nat. 127: 292-3 16. Farrell, T. M. (1988). Community stability: effects of limpet removal and reintroduction in a rocky intertidal community. Oecologia 75: 190-197.  89 Farrell, T. M. (1991). Models and mechanisms of succession: an example from a rocky intertidal community. Ecol. Monogr. 61: 95-113. Foster, M. S. (1990). Organization of macroalgal assemblages in the Northeast Pacific: the assumptions of homogeneity and the illusion of generality. Hydrobiologia 192: 2133. Hawkins, S. J. (1981). The influence of season and barnacles on the algal colonization of Patella vulgata exclusion areas. J. Mar. Biol. Ass. U.K. 61: 1-15. Hawkins, S. J., Hartnoll, R. G. (1983). Grazing of intertidal algae by marine invertebrates. Oceanogr. Mar. Biol. Annu. Rev. 21: 195-282. Horn, H. S. (1974). The ecology of secondary succession. Ann. Rev. Ecol. Syst. 5: 2537. Jernakoff, P. (1983). Factors affecting the recruitment of algae in a midshore region dominated by barnacles. J. Exp. Mar. Biol. Ecol. 67: 17-3 1. Jernakoff, P. (1985). The effect of overgrowth by algae on the survival of the intertidal barnacle Tesseropora rosea Krauss. J. Exp. Mar. Biol. Ecol. 94: 89-97. Kastendiek, J. (1982). Competitor-mediated coexistence: interactions among three species of benthic macroalgae. J. Exp. Mar. Biol. Ecol. 62: 201-210. Little, C., Williams, G. A., Morritt, D., Perrins, J. M., Stirling, P. (1988). Foraging behavior of Patella vulgata L. in an fish sea-lough. J. Exp. Mar. Biol. Ecol. 120: 1-21. Lubchenco, J. (1983). Littorina and Fucus: effects of herbivores, substratum heterogeneity and plant escapes during succession. Ecology 64: 1116-1123. Lubchenco, J., Menge, B. A. (1978). Community development and persistence in a low rocky intertidal zone. Ecol. Monogr. 48: 67-94. Lubchenco, J., Cubit, J. (1980). Heteromorphic life histories of certain marine algae as adaptations to variations in herbivory. Ecology 61: 676-687. Lubchenco, I., Gaines, S. D. (1981). A unified approach to marine plant-herbivore interactions. I. Populations and communities. Ann. Rev. Ecol. Syst. 12: 405-437. Mastro, E., Chow, V., Hedgecock, D. (1982). Littorina scutulata and Littorina plena: sibling species status of two prosobranch gastropod species confirmed by electrophoresis. Veliger 24: 239-246. Nicotri, M. E. (1977). Grazing effects of four marine intertidal herbivores on the microflora. Ecology 58: 1020-1032. Norton, T. A. (1983). The resistence to dislodgement of Sargassum inuticuin germlings under defined hydrodynamic conditions. J. Mar. Biol. Assoc. U.K. 63: 181-194. Peer, R. L. (1986). The effects of microcrustaceans on succession and diversity of an algal microcosm community. Oecologia 68: 308-314.  90 Robles, C. D., Cubit, J. (1981). Influence of biotic factors in an upper intertidal community: dipteran larvae grazing on algae. Ecology 62: 1536-1547. Slocum, C. J. (1980). Differential susceptibiitty to grazers in two phases of an intertidal alga: advantages of heteromorphic generations. J. Exp. Mar. Biol. Ecol. 46: 99110. Sousa, W. P. (1979). Experimental investigations of disturbance and ecological succession in a rocky intertidal algal community. Ecol. Monogr. 49: 227-254. Sousa, W. P. (1984). Intertidal mosaics: patch size, propagule availability, and spatially variable patterns of succession. Ecology 65: 1918-1935. Sousa, W. P., Schroeter, S. C., Gaines, S. D. (1981). Latitudinal variation in intertidal algal community structure: the influence of grazing vegetative propagation. Oecologia 48: 297-307. Turner, T. (1983). Complexity of early and middle successional stages in a rocky intertidal surfgrass community. Oecologia 60: 56-65. Underwood, A. J., Denley, E. J., Moran, M. 1. (1983). Experimental analyses of the structure and dynamics of mid-shore rocky intertidal communities in New South Wales. Oecologia 56: 202-2 19. Van Tamelen, P. G. (1987). Early successional mechanisms in the rocky intertidal: the role of direct and indirect interactions. J. Exp. Mar. Biol. Ecol. 112: 39-48. Vandermeer, J. (1980). Indirect mutualisms: variations on a theme by Stephen Levine. Am. Nat. 116: 441-448. Wilkinson, L., Hill, M. A., yang, E. (1992). SYSTAT: Statistics, version 5.2.1 edition. SYSTAT. Evanston, IL, USA. Winer, B. J. (1971). Statistical principles in experimental design. McGraw-Hill. New York.  91 Table 4.1 The mean density ±SE (in 20x20 cm plot) of barnacles and limpets in each non-excluded treatment. Data are the mean of 14 sampling dates (April 1992 August 1994; December 1993 data are missing), in each sampling date the density of barnacles and limpets was averaged from the 4 replicate plots. -  BARNACLES  Mean SE  LIMPETS  +B+L  ÷B-L  558.2  993.6  26.8  21.2  72.7  198.9  3.2  2.6  +B+L  Endocladia muricata Mastocarpus papillatus  1.0±0.03  0.5±0.21  0.4±0.07  0.1±0.00  Scytosiphon doyi  Enteromorpha spp.  Endockidia muricata  Crustose algae  Crustose algae  Scytosiphon dotyi  Porphyra spp.  BAR  6.1±3.46  -  Porphyra spp.  LIM SPECIES  +  % ±SE  BAR  SPECIES  +  -  +  0.1±0.03  0.2±0.02  0.2±0.06  0.3±0.04  1.4±0.75  % ±SE  LIM  BAR  Mastocarpus papillatus  Enteromorpha spp.  Urospora spp.  Porphyra spp.  Diatoms  SPECIES  + -  0.7±0.35  3.5±1.49  3.7±2.09  7.1±4.82  17.8±5.99  % ±SE  LIM BAR  -  Callithainnion spp.  Scytosiphon dotyi  Urospora spp.  Porphyra spp.  Diatoms  SPECIES  -  0.5±0.22  2.3±0.89  3.1±1.65  7.2±4.93  25.0±7.49  % ±SE  LIM  sampling dates (February 1992 February 1994), in each sampling date the percent cover value of each species was averaged from the 4 replicate plots.  Table 4.2 The mean percent cover of the five most abundant ephemeral algae in each treatment. Data are the mean of 13  93  A. Ephemeral algae 100 +B+L -B+L ----0----. +B-L ----ix----- B-L —D---—  c—--  80. 60. 40• (I)  I  0 -J  T  20•  I  0  E  0 0  0. I  0  c’J  z a: w > 0 C) z w C-) a: w 0  5..  I  I  I  I  B. Mazzaella  4. 3. 2.  0  $  b  b  Fig. 4.1 Effects of barnacle and limpet removal on each algal group. Data are means ± SE of four replicates. +B+L indicates both barnacle and limpet present, -B+L barnacle absent, limpet present, +B-L barnacle present, limpet absent, and -B-L both barnacle and limpet absent.  :1  •1  0  I\) 0 0  0) 0 0  Cl) Cl)  CD  p  0 I  I  -H  0  -  F) 0  II II  [hII  I  I  0 0  I-u:II  F—”JI\I  PERCENT COVER IN 20x20 cm PLOTS 0  0)  C.)  1  p  95  A.+BAR+LIM 60-  —a---- EPH MAZ  ----o---- FUC  40-  PEL  ----h---.-  T  0 T/L  20(I)  I  T  0 -J 0  E C)  0  0 I  I  I  I  I  I  I  I  I  I  I  I  I  I  I  I  I  I  I  0  c’J  z  B.  -  BAR  +  LIM  80  w  > 0 0 I—  60•  z  w  0  40  w 1  20.  I -r  0.  %  <cb 5 .5 bcb  Fig. 4.2 Effects of barnacles and limpets on algal species interactions in each treatment. EPH for ephemeral algae, MAZ for Mazzaella cornucopiae, FUC for Fucus distichus, PEL for Pelvetiopsis lirnitata. Data are means ± SE of four replicates.  96 100.  C.+BAR-LIM  80.  IT  60.  TJ  40(I)  F  0 -J  20•  E  0.  0  0 0  I  0  c’j  z  I  I  I  I  I  I  I  w  0  0 F  I  I  I  I  I  I  I  I  I  I  D.-BAR-LIM 100-  I  It >  I  I  80-  z w  60.  w  40-  TI  T  0  “1  0  200cb  Fig. 4.2 continued.  b  bqbb <  97  BARNACLES 30002500-  —D—  +B+L  I  T  ----o----• +B-L 20001500-  1000c’J  E C-)  0 0  z Cl)  w  I  5000-  i  i  i  i  i  i  T  LIMPETS  I— (I)  z w U  çOi’  Fig. 4.3 The abundance of barnacles and limpets in the experimental plots (20 x 20 cm). Data are the mean densities ± SE per plot (n=4).  98  Yearl &2 EARLY  .b.  ii  Ephemeral algae Diatoms Porphyra  Ephemeral algae Diatoms Porphyra  Fucus (more)  Pelvetiopsis (lots)  Pelvetiopsis (less)  Fucus (few)  Year 3 LATE Fucus distichus  Fucus distichus  Fucus (more)  Pelvetiopsis (lots)  LU  0—I  +  +  Pelvetiopsis (less)  Fucus (few)  +  BARNACLES BARNACLES (Balanus glandula)  Fig. 4.4 A summary of barnacle and limpet effects on algal succession. Dominant algae were shown in each experimental condition with their relative abundance.  99  A  LEG END Solid line : strong interaction Dashed line : weak interaction Arrow head : negative interaction Round head: positive interaction Clear point : indirect interaction Dark point : direct interaction  Fig. 4.5 The summary of direct and indirect interactions shown between barnacles, limpets, fucoids and ephemeral algae in this study.  100 CHAPTER 5  EFFECTS OF SIZE OF EXPERIMENTAL CLEARINGS AND SEASON OF CLEARING ON ALGAL PATCH RECOVERY  INTRODUCTION  The species composition and abundance of organisms in many rocky intertidal and subtidal communities is highly influenced by disturbance (Sousa 1984b, Connell and Keough 1985, Sousa 1985). Disturbance is defined here as any discrete event that decreases the amount of living biomass in an area and opens up space for the establishment of new individuals or colonies. Investigators have extensively studied various aspects of disturbance on succession and community structure in marine habitats including frequency (Connell 1978, Sousa 1979a,b, Miller 1982), size (Osman 1977, Paine and Levin 1981, Keough 1984, Farrell 1989, Benedetti-Cecchi and Cinelli 1994), intensity (Sousa 1980, see Connell and Keough 1985 for references), location (Palumbi and Jackson 1982, Sousa 1984a, Dayton et al. 1992), and time or seasonality (Foster 1975, Hawkins 1981, Turner 1983a, Jara and Moreno 1984, Breitburg 1985, Kennelly 1987, Dayton et al. 1992, Benedetti-Cecchi and Cinelli 1994). The formation of gaps is of considerable significance to sessile organisms which require open space. Organisms secure space by overgrowing or by spreading vegetatively into open space, or by growing from dispersed propagules. Various attributes of a newly created gap (i.e., its size, shape, location and time of creation) can affect subsequent patch colonization and ensuing biological interactions (Sousa 1985). Species with particular life histories or reproductive strategies may colonize some areas but not others. For example, the centers of very large gaps are most likely to be colonized by species producing propagules which travel relatively great distances. Such dispersal capability is less important in smaller gaps, since most recolonizing propagules will be produced by  101 individuals established adjacent to the gap (Davis and Cantlon 1969, Osman 1977, Sousa 1979b, Begon et al. 1990). This generalization has been examined experimentally in marine habitats (Keough 1984, Sousa 1984a, Farrell 1989). The rate of colonization may vary with patch size because biological environments of patches may also vary with size. Sousa (1985) describes that the most obvious influence of patch size derives not from the size of a patch per se but from the manner in which the ratio of patch perimeter to area changes with patch size. For example, smaller patches have a greater ratio of edge to area than larger patches. Therefore, the number of propagules received per unit area in smaller patches is greater than that of larger patches. His argument clearly supports the idea that smaller patches are colonized more rapidly than larger ones, and that this is particularly true when the distance of dispersal is small. However, smaller patches may be influenced by factors which negatively influence germination of propagules or post-recruitment survivorship of newly-settled individuals and consequently slow the colonization process. These factors may include shading (Reed and Foster 1984, Kennelly 1989), scouring (Velimirov and Griffiths 1979, Kennelly 1989)  and increased herbivory (Sousa 1 984a). In intertidal mussel beds, small clearings harbored higher densities of grazers, particularly limpets, than do large patches because small patches provided a greater ratio of edge area for limpets which tended to aggregate at edges. (Paine and Levin 1981, Sousa 1984a). A similar result has been reported from a high intertidal zone (Farrell 1989). Consequently, small patches may have a slow colonization due to increasing herbivory (Sousa 1985). Barnacles (e.g., Balanus) were influenced by different sizes of disturbance in some studies. Farrell (1989) found that barnacles were more abundant in large gaps probably due to less deleterious effects imposed by algae. However, it is difficult to compare the effect of patch size on the rate of recovery because different studies used different sizes of experimental clearing. In addition, two recent studies dealing with this problem and using similar sizes of clearing produced conflicting results. Farrell (1989) found that small gaps (4x4 cm) were colonized most rapidly,  102 followed by medium (8x8 cm) and larger gaps (16x16 cm) in the upper intertidal zones of the Oregon coast, U.S.A. Benedetti-Cecchi and Cinelli (1994) reported that the algal colonization rate was highest in larger gaps (20x20 cm), followed by medium (lOxlO cm) and small (5x5 cm) gaps in a low intertidal habitat of Italy. Deterministic views of community succession using one dimensional models of disturbance (i.e., size) make it more complicated and difficult to predict the patterns of community dynamics than when two or three dimensional models of disturbance are used (i.e., frequency, seasonality, etc.) (Miller 1982, Malanson 1984). It is highly unlikely that natural disturbance occurs uniformly (e.g., same size or same intensity). Thus its natural occurrence may involve combinations of factors which could often be correlated (Sousa 1980, Paine and Levin 1981, Miller 1982, Malanson 1984). For example, Paine and Levin (1981) reported that the largest clearings in mussel beds were created by winter storm waves and such clearings were less frequent than small clearings produced by waves of lesser magnitude in any season. In this case, the interaction of size and season characterizes this disturbance. It is clearly important to determine the key characteristics of disturbances which are critical in understanding the dynamics of the community investigated (for instance, see Sousa 1980, Breitburg 1985, Benedetti-Cecchi and Cinelli 1994). Both individual factors of disturbance and their interaction are important for understanding patch recovery and succession (Miller 1982, Malanson 1984). The time or season at which a gap is created by disturbance is known to determine early successional algal assemblages because of seasonal availability of propagules of certain species (Foster 1975, Paine 1977, Emerson and Zedler 1978, Hawkins 1981, Breitburg 1985, Kennelly 1987). Such initial variability may influence subsequent colonization, so that the gaps created in different seasons become dominated by different organisms in later successional stages (Turner 1983b, Jara and Moreno 1984). In intertidal habitats, Benedetti-Cecchi and Cinelli (1994) reported examples of significant seasonal effects on recovery of patches, although these patterns varied with algal species and site.  103 However, other studies have negated the above theory, and indicated that the effect of season of disturbance on producing an initial colonization pattern became less evident as succession proceeded and the patches were eventually dominated by the same perennial species (Sousa 1 979a, Hawkins 1981). More information from habitats at different tidal heights and with different algal assemblages are required for generalization. The objective of this chapter is to demonstrate the effects of size and season of experimentally cleared gaps on algal recolonization and within-patch dynamics in a high intertidal, Mazzae!la-Fucus-Pelvetiopsis dominated community. The study was designed to address the following questions. 1) Is the rate of algal recovery faster in a smaller patch (gap)? The answer for this question provides a comparison for the effect of gap size on recovery rate with two recent studies (Farrell 1989, Benedetti-Cecchi and Cinelli 1994) by using similar sizes of clearings. 2) Do different clearing seasons affect the dominance of a particular species in this community? This question is asked to test the two contrasting views on the effect of timing of disturbance on patch structure described earlier. 3) What species come in when a gap of certain size is created in a certain time of the year? Predictions about community structure following various combinations of the two aspects of disturbance may be possible to some degree when the responses of each component species are known using the interaction term of size and season of disturbance. 4) Are limpets more abundant in small patches than large patches? Are barnacles more abundant in large patches than small ones? The answers to these questions test some of the hypotheses described earlier.  104 METHODS  Experimental design  The study was conducted at Prasiola Point. The experimental units consisted of square piots of three different sizes (5x5, lOxlO and 20x20 cm). To test for seasonal effects, small (5x5 cm) and medium (lOxlO cm) piots were cleared four times a year summer (August 1991), fall (October 1991), winter (February 1992) and spring (April 1992); large plots (20x20 cm) were cleared only in July 1991. At the beginning of the experiment, a permanent transect line was placed along the shore within the zone 3.4-4.2 m above Lowest Normal Tide (LNT: Canadian Chart Datum). The position of the line was marked by hammering concrete nails or drilling bolts at every meter. The transect was placed only on mixed stands of the three species of interest, so that the line was sometimes disconnected due to heterogeneity in algal assemblages on the rock substratum. Using 10 cm intervals along the transect line, I numbered all potential plots for clearing small and medium plots on both sides of the transect line. Four large plots (20x20 cm) were cleared away from the transect line but within the same tidal zone. At each clearing season, fifteen plots of each small and medium size were chosen using a random number table. When selecting plots for clearing, I intentionally skipped some random numbers to avoid clearing a plot adjacent to a previously cleared one; thus, at least one uncleared plot existed between cleared ones. I assumed that this method enabled me to treat data as though they were unaffected by those from adjacent clearings. Four unmanipulated permanent 20x20 cm plots were also marked in order to monitor natural seasonal fluctuations of algal abundance. Plots were cleared by hand chiseling (for the small and medium plots) or by using a compressed-air powered drill with a chisel bit (for the large plots). This allowed me to remove all organisms (including holdfasts, barnacles, crustose algae, etc.) in the plots.  105 The denuded plots were rechecked the next day to remove any organisms that remained. In this way, completely new substrata were attained in every clearing. Recolonization by macroalgae was monitored bimonthly for two years. Quadrats of the same size as the experimental plots, with lxi cm subplots marked off within the quadrat frame (25, 100 and 400 subplots for small, medium and large quadrats, respectively), were used for measuring the abundance of organisms. Percent cover of each species was estimated by counting the number of subplots in which a species occupied >50% of its area. This technique was consistently used for estimating algal percent cover in my study. Algal cover consisted of only those thalli with their holdfasts in the plot. Crustose algae could not be identified in the field; therefore, I treated them as a single entity. Densities of barnacles and limpets were monitored at every sampling date. Barnacles became too numerous to count in the medium and large plots so subunits were censused. Five and twenty randomly chosen 2x2 cm areas were censused in each medium and large plot, respectively. Data from 4, 10 and 16 months after initiation, of summercleared plots, are presented to compare the distribution of the invertebrates in three different patch sizes.  Data analysis  Percent cover of algal species from the same plot was measured repeatedly on successive sampling dates and these values were thus correlated over time (dependent variables). Therefore, data were analyzed using a repeated measures analysis of variance (RM-ANOVA). A three-way (Size x Season x Time) ANOVA with repeated measures on the Time factor was separately applied for each of the three dominant algae and a single additional group consisting of ephemeral algae. Data from large plots (20x20 cm) were not included in the ANOVA because they were cleared in only one season. Therefore, the  106 effects of this latter treatment were deduced by inspection of graphical data. The repeated measures of percent cover for each species did not exhibit homogeneity of variance for all pairs of measurements within subjects (“compound symmetry”; Gurevitch and Chester 1986, Myers and Well 1991). This precludes use of a standard univariate RM-ANOVA to analyze the within-subject hypotheses due to inflation of type I error rates (Myers and Well 1991). In these cases, I presented both the probability values of Huynh-Feldt adjusted degrees of freedom as well as the results of multivariate analysis of variance (MANOVA) where the assumption of sphericity was not required (Mead 1988, Myers and Well 1991). Cochran’s C test and Fmax test (Winer 1971) were used to check homogeneity of variance. In most cases, except for total algal cover data, this assumption was violated and therefore appropriate transformations were used (Underwood 1981). Percent cover of each algal species was arcsine-transformed prior to analysis. Barnacle and limpet densities were respectively log- and (n+l) square root-transformed. Transformation did not reduce the heterogeneity to nonsignificant levels in a few cases; for example, some data from the initial sample dates. Although these data were dropped from the repeated measures ANOVA’s, this procedure did not affect statistical results. Plots of residuals against estimated values were visually examined after each ANOVA to check if error terms were normally distributed. The only exceptions were the barnacle and limpet data from large plots, due to small (n=4) sample size. All data were analyzed using SYSTAT, version 5.2.1 for Macintosh (Wilkinson et al. 1992).  RESULTS  Unman ipulated plots  Mazzaella comucopiae and Fucus distichus were the most conspicuous species in unmanipulated plots during the study (mean cover, 30.2% and 25.2%, respectively; Fig.  107 5.1). Pelvetiopsis limitata (11.8%) was relatively abundant in the first year, but was lower (e.g., April and June 1993) in the second year. Ephemeral algae covered only 4.4% of the plots during this period. The abundance of most algae in unmanipulated plots showed a similar seasonal pattern; it was highest in February to June and was lowest in August to December. Seasonal changes in algal abundance were greater in both fucoids and ephemeral species than in M. cornucopiae (Fig. 5.1). Peaks in recruitment of both fucoids were observed in October of 1991, 1992 and 1993 except for P. limitata in October 1992. The timing of maximal recruitment was in contrast to the peaks in percent covers of both species, the latter occurring in April.  Size effects of disturbance  Total algal cover following summer clearing of each of the three plot sizes is shown in Fig. 5.2. The lOxlO cm plots had a significantly greater percent cover of total algae than the 5x5 cm plots. This pattern occurred also in other plots cleared in different seasons, except for those cleared in the fall (Table 5.1). The pattern was also consistent (no significant interaction between Time and Size) throughout the experimental period except for winter-cleared plots (Table 5.1). Colonization in the 20x20 cm plots was the slowest among the three sizes cleared in summer. Total algal cover in large plots remained less than 5% until 14 months after clearing whereas 8 months after clearing the total algal cover reached 26.4% in small plots and 35.5% in medium plots (Fig. 5.2). Percent cover of both Fucus distichus and ephemeral algae was greatest in medium plots (Size effects, p=O.OO1 and p=O.OO3, respectively; Table 5.2, Fig. 5.3). However, for Mazzaella cornucopiae and Pelvetiopsis limitata, differences in percent cover between small and medium plots were not great (p=O.592 and pO.2O4, respectively). Although no statistic was applied, it was obvious that percent covers of Mazzaella in small and medium plots were much greater than that in large plots (Fig. 5.3). This alga never successfully  108 settled in large plots. Colonization by F. distichus occurred more rapidly in medium plots than in small and large plots (which it colonized at similar rates) although this pattern was not recognizable during the first 6 months (Fig. 5.3). However, ephemeral algae colonized rapidly from early in the succession, especially in the medium plots. Ephemeral species in medium and small plots consisted of a similar assemblage. In medium plots, crustose algae, Cladophora colunibiana, Scytosiphon dotyi and Endocladia muricata were most abundant (in decreasing abundance). Small plots were settled by crustose algae, S. dotyi,  C. columbiana and Polysiphonia sp. In contrast, the ephemeral group in large plots was dominated by Porphyra sp., followed by E. muricata and S. dotyi. I occasionally found a few small clumps of M. cornucopia, especially in spring, but these usually disappeared by the next sampling dates. Pelvetiopsis limitata was the species least affected by the size of the disturbance; it showed similar cover in all three plot sizes over time.  Seasonal effects of disturbance  In medium and small plots, there was no difference in algal abundance among plots cleared in different seasons, except for P. limitata (pO.O2, Table 5.2; analysis included data through 16 months). The mean percent cover of P. lim.itata during the experiment from summer clearing averaged over the two sizes was 2.54%, which was significantly greater than those from the winter (0.67%) and the fall clearings (0.45%) (Tukey’s HSD, p=O.049 and p=O.O , respectively). Fucus had a high percent cover in summer cleared 22 medium plots (Fig. 5.5). In small plots, Mazzaeila showed relatively faster colonization in summer and fall cleared plots than in spring and winter cleared plots (Fig. 5.4).  109 Interactions of the disturbance factors  For percent cover of ephemeral algae there was a significant interaction between size of plot and season of clearing (p=O.OO9; Table 5.2). The greater abundance of ephemeral algae in medium plots was found only in spring and summer-cleared plots, whereas a similar abundance of these algae occurred in the medium and small plots cleared in fall and winter (Figs. 5.4 and 5.5). This difference, however, was not consistent during the successional period (Time x Size x Season interaction, p=O.OO1); the largest difference occurred only in October 1992 (Figs. 5.4 and 5.5). Responses of algae to the different sizes and seasons of disturbance were quite variable. For example, relatively high cover and rapid colonization of Mazzaella were found in summer and fall clearings of the small plots. In contrast, the rate of colonization of this alga was slower in medium fall-cleared plots (Figs. 5.4 and 5.5). Taking time into account, the interpretation of the effect of size of clearing on Fucus colonization gave additional variation. The pattern of Fucus abundance (medium plots> small plots) was not consistent during the experiment (Time x Size in MANOVA, p=O.OOl; the probability value in the MANOVA was used when the nonsphericity within subjects was severe); for example, the size effect was significant only in summer plots 10 months after clearing (Fig. 5.3). In contrast, the seasonal effect on Pelvetiopsis abundance (summer> winter, fall clearings) was similar between medium and small plots (Size x Season interaction, p=O.l 12) and also was consistent over successional time (Time x Season interaction, p=O.337; Table 5.2, Figs. 5.4 and 5.5).  110  Barnacle and limpet densities  There were significant differences in barnacle density (ANOVA, F( )=4.088, 3 , 2 1 53.708, 19.36 1; p<O.05, p<O.0001, p.<O.0001; in the three patch sizes 4, 10, 16 months after clearing, respectively; Fig. 5.6). Tukey’s HSD tests indicated that barnacle densities in the small patches were significantly lower than in either the medium or large patches on all tested dates (Fig. 5,6). Differences between medium and large patches were not significant for the first two sample dates. However, mean density in medium patches was higher than in large patches at 16 months after initiation. Limpet densities were not significantly different between patch sizes for any sampling date (p>0. 1, in all cases; ANOVA was used on each sample date). Although the relative abundance of limpets at month 4 was higher in small patches (5.9 per 100 cm ) 2 than in medium patches (2.7 per 100 cm ), the mean densities averaged over all three 2 sample dates were similar for all patch sizes (4.8, 4.7, 4.1 per 100 cm 2 in small, medium, large, respectively).  DISCUSSION  Size effects of disturbance  In this study the rate of patch colonization was influenced by the size of disturbance. The medium (lOx 10 cm) patches were colonized the fastest, followed by the small (5x5 cm) and large (20x20 cm) patches. The present results were different from those of the previous studies either by Farrell (1989) or by Benedetti-Cecchi and Cinelli (1994), and indicate that negative factors that inhibited patch colonization of both small and large plots were effective in my study.  111 In small piots, adult plants surrounding the small patches tend to shade out and/or whiplash the cleared space, decreasing the attachment and growth of propagules of their own or other species (Dayton 1975, Sousa 1979a, Velimirov and Griffths 1979, Reed and Foster 1984, Kennelly 1989). These competitive interactions between the species recently settled in a patch and the organisms surrounding that patch have been invoked as mechanisms limiting colonization of small gaps (Connell and Keough 1985, Sousa 1985). Fucus and Pelvetiopsis were the most likely species to shade or whiplash the cleared space in small plots with respect to the size of fucoids (5-7 cm in length) and plots. It is possible for fucoids growing adjacent to the patch to cover completely the 5x5 cm plots at low tides; this was commonly observed in the field. Whiplashing fucoids at high tides might deleteriously affect the settlement of barnacles and their own recruitment in this habitat, My results clearly negated the hypothesis that smaller patches are colonized more rapidly, proposed by Sousa (1985), and indicated that the above deterrent factor(s) appear to be more important in patch recovery of the 5x5 cm plots in this community than a potential benefit of the greater edge ratio. Sousa (1984a) reported that slow colonization in small patches was caused by increased herbivore density (i.e., limpets). I could not find a strong evidence for intensive herbivory in small plots in my data. Limpet abundance was similar among the three patch sizes, with a slightly higher density occurring in small patches only during early successional stages. Algal recovery in the large patches was not completely inhibited, but only delayed in comparison with colonization of small and medium patches. Whatever the mechanism(s) accounting for the delay in recolonization in large plots, its (their) effect was strong in the first year, but generally decreased as succession proceeded. The slow (or delayed) colonization in the large patches may be explained by the following factors; 1) relatively stronger heat stress and desiccation in the center area of large patches may cause higher mortality of settling algae, 2) short dispersal distance of component species (e.g., fucoids) may be responsible for limitation of colonization, or 3) simply, the lower ratio of edge area  112 is responsible for the slow colonization. Based on the sizes of component species in this community, it is clear that much of the area in the 20x20 cm plots is directly exposed to sun light. Consequently, zygotes and embryos of fucoids may undergo higher mortality due to severe cellular dehydration (Davison et al. 1993). Although there are no numeric data available on the dispersal distance of Fucus and Pelvetiopsis, this distance is thought to be ‘very short’ (Sousa 1984a, Hoffman 1987, Farrell 1989, Menge et al. 1993). The intensity of the factors limiting colonization in my two extreme sizes of patches may decrease as patch size changes towards an intermediate one. Medium patches may simply receive the advantages of the reduction of the negative factors of the two size extremes. Similarly, the intermediate disturbance hypothesis (Horn 1975, Connell 1978) suggests that an intermediate frequency of disturbance may allow the coexistence of species with different life history strategies, which in turn allows for the highest species diversity. Furthermore, Miller (1982) claimed that size of disturbance provides an extension of the intermediate disturbance hypothesis. He explained that patches of intermediate size may equally allow colonization of both the ‘competitive species’, which cover a disturbed area by dispersing or growing primarily from around the patch perimeter, and the ‘colonizing species’, which colonize rapidly through high growth and/or dispersal rate. In this community, the three dominant algae may not functionally fit to the above categories. However, medium patches in this study were generally recolonized by a relatively wellmixed combination of all groups including Mazzaella (rare in large patches) and Fucus (low in small patches). However, it may still be difficult to generalize the effect of patch size on the rate of colonization because the rate may vary with habitat and algal assemblage. For example, in Farrell’s study (1989), although colonization in 3 sizes of clearing and tidal height were similar, he found that the effect of surrounding plants, which was a negative one in small patches of my study, was a positive one in its effect on the settlement of other or their own species. Furthermore, Mazzaella cornucopiae was absent in his study areas, but Endocladia muricata, which is rare in my site, was abundant and colonized in the large  113 patches. In Benedetti-Cecchi and Cinelli’s study (1994), the tidal height and algal assemblage were different from my study, and they found that most algal species (e.g., Corallina, Polysiphonia) colonized by the settlement of propagules, so vegetative propagation was unimportant. My result is the first report which experimentally supports Miller’s hypothesis from intertidal habitats and calls for replicate studies in other habitats.  The effect of timing of disturbance  The differences in algal abundance among patches of similar age but produced at different seasons were minor for most species. However, there was a trend that different algal species responded differently to the seasonal effect of disturbance. First of all, Fucus and Pelvetiopsis, which are characterized as being slow-growing, perennial, and dispersed by propagules (e.g., zygotes), tended to dominate patches created just before’ (e.g., 2 month) their recruitment peaks, compared to patches made at other times. As a result, summer clearings (August 1991) created 2 months before the recruitment peak of fucoids (Oct. 1991; Fig. 5.1) gave rise to greater cover of Pelvetiopsis in both small and medium plots than plots cleared in any other season. Fucus showed a similar pattern but only in the medium patches. New substrata created just before’ the peak of reproduction may collect greater numbers of fucoid propagules with no interference from previously-settled species. For these species, the time of gap creation is important because the initial recruitment may determine the later succession, as shown in other studies (Turner 1983b, Jara and Moreno 1984). Jara and Moreno (1984) reported that plots cleared in spring were dominated by ephemeral algae but, if cleared in fall, plots were filled by barnacles; these results were due to seasonal availability of spore and inhibition between settled organisms and later species. On the other hand, for species such as Mazzaelki which propagate mainly vegetatively, the timing of clearing generally did not influence dominance of this alga in the plots. However, in medium patches the abundance of Mazzaella in winter-cleared patches  114 (made 2 months before the abundance peak) was slightly greater than in the other patches cleared in different seasons. In this study, seasonal effects of disturbance were highly species-specific; season-of-clearing had more effect on species dispersing by propagules and less on species reproducing by vegetative ingrowth. Also, this study was supportive of the generalization that the abundance of a species is most enhanced by disturbance if it is created when the propagules of the species are available for settlement (Denley and Underwood 1979, Sousa 1979a, Hawkins 1981, Breitburg 1985, Kennelly 1987).  Responses of barnacles and limp ets to the size of disturbance  The abundance of barnacles was influenced by patch size, and this may be due to the deleterious effects of the algal canopy. Several authors have found that algal cover reduces barnacle settlement, or decreases their survival, by scouring young barnacles (Hawkins 1983, Woodin and Jackson 1979) or overgrowing their apertures (Farrell 1991, Dayton 1974). In contrast, it has been reported that barnacles are more abundant under algal canopies due to a decrease in desiccation-caused mortality (Dayton 1971, Farrell 1989). In this study, the negative effects of macroalgae on barnacles appeared to be more important in controlling barnacle densities in the small patches. Both fucoids (4-7 cm) seemed ‘large’ enough to ‘whiplash’ young barnacles in the 5x5 cm plots. Mazzaella cornucopiae sometimes appeared to kill barnacles as it overgrew them; this might be another factor responsible for the low density of barnacles in small patches. Limpet abundance was not influenced by clearing size; mean densities during the succession were similar in all patch sizes. Therefore, the relationship between limpet grazing and algal colonization as related to patch size in this study did not support the results obtained from the other studies. Some studies have found that limpet densities decrease with increasing patch size in mussel beds (Suchanek 1978, Sousa 1984a) and in an upper intertidal zone (Farrell 1989), because limpets tend to aggregate at patch edges.  115 Therefore, small piots with a greater unit boundary area might collect a relatively greater density of limpets (Farrell 1989). Some authors claimed that increased limpet density in small patches were responsible for slower colonization of this patch (see Sousa 1985 for references). In this study, limpet densities were highest in small patches, but only initially; thus the density distribution of limpets among patch sizes showed high variability through time.  Responses of individual algal species to disturbance  The progression of species settlement and replacement in a disturbed patch can be predicted to some degree when disturbance is analyzed by a multi-dimensional model. Several models have predicted the species diversity of patches by incorporating some characteristics of disturbance together rather than considering each one separately (Miller 1982, Malanson 1984). This study has shown patterns of succession following disturbance on the basis of a two dimensional view, the size of disturbance and the time at which the disturbance occurred. What species come in when a gap of certain size is created in a certain season? The answer to this question can now be predicted with some certainty for this community. The following four cases are the possible predictions about community structure made by using the results of this chapter. 1) if a small gap is created in fall, the gap will be dominated only by Mazzaella. 2) if a small gap is formed in summer, the gap will be dominated by both Mazzaella and Pelvetiopsis. However, it is expected that the canopy-forming fucoid may finally exclude the short, turf-forming red alga by shading (Chapter 2). 3) if a medium gap is made in summer, Mazzaella and ephemeral algae will dominate space early in succession (e.g., <6 months), but later Fucus will gradually colonize and become a co dominant. The patch may be a mixture of the 3 algal species, and their relative proportion  will fluctuate seasonally. A similar pattern of patch dynamics is expected in a medium gap  116 created in other seasons. 4) If a large gap is created in summer, the patch will recover relatively slowly, but fucoids are expected to dominate the gap with only a seasonal appearance of ephemeral algae. Mazzaella may not appear for at least first 3 years after disturbance (Chapter 4). To predict the natural dynamics of this Mazzaella-Fucus-Pelvetiopsis dominated community, at least two other factors must be known; frequency of disturbance size and the mechanisms of interspecific interactions of the dominants (shown in Chapter 2). if large patch sizes occur less frequently, and smaller sizes are more frequent in natural habitats (Sousa 1979b, Paine and Levin 1981, Farrell 1989), the population of M. cornucopiae will remain as a co-dominant in this community. Although the size frequency of disturbance  was not measured in this study, the three sizes of my experimental plots were similar to those found to occur naturally (e.g., Oregon coasts, Farrell 1989). In a single-speciesdominated community, the sequence of succession within a patch can be deterministic and then lead to the monopolization of space within the patch by one competitively dominant and/or long-lived species. However, for multi-species dominated communities information is required on the competitive abilities of all the component species in order to better understand the within-patch dynamics. In Chapter 6, I present a more precise prediction about community structure with the insight of interspecific interactions of component organisms which I achieved from the study in Chapter 2.  117 REFERENCES Begon, M., Harper, J. L., Townsend, C. R. (1990). Ecology: individuals, populations and communities (2nd ed.). Blackwell Scientific Publications. London. 945 pp. Benedetti-Cinelli, L., Cinelli, F. (1994). Recovery of patches in an assemblage of geniculate coraline algae: variability at different successional stages. Mar. Ecol. Prog. Ser. 110: 9-18. Breitburg, D. L. (1985). Development of a subtidal epibenthic community: factors affecting species composition and the mechanisms of succession. Oecologia 65: 173-184. Connell, J. H. (1978). Diversity in tropical rain forests and coral reefs. Science 199: 13021310. Connell, J. H., Keough, M. J. (1985). Disturbance and patch dynamics of subtidal marine animals on hard substrata. In: Pickett, S. T. A., White, P. S. (eds.) The ecology of natural disturbance and patch dynamics. Academic Press. Orlando. pp. 125-151. Davis, R. M., Cantlon, J. E. (1969). Effect of size area open to colonization on species composition in early old-field succession. Bull. Torr. Bot. Club 96: 660-673. Davison, I. R., Johnson, L. E., Brawley, S. H. (1993). Sublethal stress in the intertidal zone: tidal emersion inhibits photosynthesis and retards development in embryos of the brown alga Pelvetiafastigiata. Oecologia 96: 483-492. Dayton, P. K. (1971). Competition, disturbance and community organization: the provision and subsequent utilization of space in a rocky intertidal community. Ecol. Monogr. 41: 35 1-389. Dayton, P. K. (1974). Dispersion, dispersal and persistence of the annual intertidal alga, Postelsiapalmaeformis Ruprecht. Ecology 54: 433-438. Dayton, P. K. (1975). Experimental evaluation of ecological dominance in a rocky intertidal algal community. Ecol. Monogr. 45: 137-159. Dayton, P. K., Tegner, M. J., Parnell, P. E., Edwards, P. B. (1992). Temporal and spatial patterns of disturbance and recovery in a kelp forest community. Ecol. Monogr. 62: 421-445. Denley, E. J., Underwood, A. J. (1979). Experiments on factors influencing settlement, survivorship and growth of two species of barnacles in New South Wales. J. Exp. Mar. Biol. Ecol. 36: 269-293. Emerson, S. E., Zedler, J. B. (1978). Recolonization of intertidal algae: an experimental study. Mar. Biol. 44: 3 15-324. Farrell, T. M. (1989). Succession in a rocky intertidal community: the importance of disturbance size and position within a disturbed patch. J. Exp. Mar. Biol. Ecol. 128: 57-78. Farrell, T. M. (1991). Models and mechanisms of succession: an example from a rocky intertidal community. Ecol. Monogr. 61: 95-113.  118 Foster, M. S. (1975). Algal succession in a Macrocystis pyrifera forest. Mar. Biol. 32: 313-329. Gurevitch, J., Chester, J. S. T. (1986). Analysis of repeated measures experiments. Ecology 67: 25 1-255. Hawkins, S. J. (1981). The influence of season and barnacles on the algal colonization of Patella vulgata exclusion areas. J. Mar. Biol. Ass. U.K. 61: 1-15. Hawkins, S. J. (1983). Interactions of Patella and macroalgae with settling Semibakinus balanoides (L.). J. Exp. Mar. Biol. Ecol. 71: 55-72. Hoffmann, A. J. (1987). The arrival of seaweed propagules at the shore: a review. Bot. Mar, 30: 151-165. Horn, H. S. (1975). Markovian processes of forest succession. In: Cody, M. L., Diamond, J. M. (eds.) Ecology and evolution of communities. Harvard University Press. Cambridge, Mass. pp. 196-211. Jara, H. F., Moreno, C. A. (1984). Herbivory and structure in a midlittoral rocky community: A case in southern Chile. Ecology 65: 28-38. Kennelly, S. J. (1987). Physical disturbances in an Australian kelp community. I. Temporal effects. Mar. Ecol, Prog. Ser. 40: 145-153. Kennelly, S. J. (1989). Effects of kelp canopies on understorey species due to shade and scour. Mar. Ecol. Prog. Ser. 50: 123-130. Keough, M. J. (1984). Effects of patch size on the abundance of sessile marine invertebrates. Ecology 65: 423-437. Malanson, G. P. (1984). Intensity as a third factor of disturbance regime and its effect on species diversity. Oikos 43: 411-413. Mead, R. (1988). The design of experiments. Cambridge University Press. Cambridge. Menge, B. A,, Farrell, T. M., Olson, A. M., Van Tamelen, P., Turner, T. (1993). Algal recruitment and the maintenance of a plant mosaic in the low intertidal region on the Oregon coast. J. Exp. Mar. Biol. Ecol. 170: 91-116. Miller, T. E. (1982). Community diversity and interactions between the size and frequency of disturbance. Am. Nat. 120: 533-536. Myers, J. L., Well, A. D. (1991). Research design and statistical analysis. HarperCollins publishers Inc. New York. 713. Osman, R. W. (1977). The establishment and development of a marine ephifaunal community. Ecol. Monogr. 47: 37-63. Paine, R. T. (1977). Controlled manipulations in the marine intertidal zone and their contributions to ecological theory. Spec. Publ., Acad. Nat. Sci. Philadelphia No. 12: 245-270.  119 Paine, R. T,, Levin, S. A. (1981). Intertidal landscapes: disturbance and dynamics of pattern. Ecol. Monogr. 51: 145-178. Palumbi, S. R., Jackson, I. B. C. (1982). Ecology of cryptic coral reef communities. 2. Recovery from small disturbance events by encrusting bryozoa: the influence of “host” species and lesion size. J. Exp. Mar. Biol. Ecol. 64: 103-115. Reed, D. C., Foster, M. S. (1984). The effects of canopy shading on algal recruitment and growth in a giant kelp forest. Ecology 65: 937-948. Sousa, W. P. (1979a). Experimental investigations of disturbance and ecological succession in a rocky intertidal algal community. Ecol. Monogr. 49: 227-254. Sousa, W. P. (1979b). Disturbance in marine intertidal boulder fields: the non-equilibrium maintenance of species diversity. Ecology 60: 1225-1239. Sousa, W. P. (1980). The responses of a community to disturbance: the importance of successional age and species’ life histories. Oecologia (Berlin) 45: 72-81. Sousa, W. P. (1984a). Intertidal mosaics: patch size, propagule availability, and spatially variable patterns of succession. Ecology 65: 19 18-1935. Sousa, W. P. (1984b). The role of disturbance innatural communities. Ann. Rev. Ecol. Syst. 15: 353-391. Sousa, W. P. (1985). Disturbance and patch dynamics on rocky intertidal shores. In: Pickett, S. T. A., White, P. S. (eds,) The ecology of natural disturbance and patch dynamics. Academic Press. Orlando. pp. 101-124. Suchanek, T. H. (1978). The ecology of Mytilus edulus L. in exposed rocky intertidal communities. J. Exp. Mar. Biol. Ecol. 31: 105-120. Turner, T. (1983a). Complexity of early and middle successional stages in a rocky intertidal surfgrass community. Oecologia 60: 56-65. Turner, T. (1983b). Facilitation as a successional mechanism in a rocky intertidal community. Am. Nat. 121: 729-738. Underwood, A. J. (1981). Techniques of analysis of variance in experimental marine bilogy and ecology. Oceanogr. Mar. Biol. Ann. Rev. 19: 513-605. Velimirov, B., Griffiths, C. L. (1979). Wave-induced kelp movement and its importance for community structure. Bot. Mar. 22: 169-172. Wilkinson, L., Hill, M. A., yang, E. (1992). SYSTAT: Statistics, version 5.2.1 edition. SYSTAT. Evanston, IL, USA. Winer, B. J. (1971). Statistical principles in experimental design (2nd ed.). McGraw-Hill. New York. Woodin, S. A., Jackson, J. B. C. (1979). Interphyletic competition among marine benthos. Am. Zool. 19: 1029-1043.  28  Error  7 196  Time x Size  Error  WL-i-+ F WL++ F  MS F MS F MS  MS F MS  Statistic  0.191 12.08 0.599 1.91  4112.2 29.18 246.7 1.75 140.9  5872.2 4.58 * 1281.0  Spring +  0.166 15.80 0.643 1.75  3565.1 14.24 211.1 0.84 250.3  7125.6 4.98 * 1431.8  Summer  0.385 5.03** 0.624 1.90  2685.7 13.24 405.4 1.20 202.8  2782.4 1.81 1537.2  fall  0.31 6.98* 0.57 2.37  3116.7 18.41 560.2 3.31 * 169.3  7191.0 6.32 * 1138.2  Winter  Univariate within subjects F and probability values were adjusted using the Huynh-Feldt estimator (epsilon = 0.5 134, 0.47 16, 0.6971, 0.5752 for spring, summer, fall, winter, respectively). Systat (Wilkinson et al. 1992) provides automatically adjusted F values and the probabilities in the case of nonsphericity. +14 replicates per size of clearing were used for analysis due to missing data. +-i-Wilks’ lambda. *p.<0.05; **p.<O.Ol; ***p<0001 (ANOVA F).  TimexSize  Time  Effect  Multivariate repeated measures analysis  7  Time  Within subject  1  i  Size  Between subject  Source of variation  Table 5.1 The effect of size of experimental clearings (5x5 and lOx 10 cm) on total algal cover. A separate repeated measures analysis of variance was applied on data from each season. Large plots (20x20 cm) were not included in analysis, as discussed in text.  Time Time x Size Timex Season Timex Size x Season  Effect  repeated-measures  0.150 0.007 0.042 0.023 0.016  0.037 0.181 0.207 0.126  MS  0.653 0.969 0.578 0.68  Wilks lambda  analysis  7 7 21 21 784  Within subjects Time Timex Size TimexSeason Timex Size x Season Error (Time)  Multivariate  1 3 3 112  df  Size Season SizexSeason Error  Between subjects  Source of variation  8.031 0.493 3.051 2.088  E  9.345 0.448 2.610 1.433  0.29 1.437 1.644  E  Mazzaella  0.000 0.838 0.000 0.004  P  0.000 0.761 0.003 0.153  0.592 0.236 0.183  6 6 18 18 666  1 3 3 111  Lf  0.560 0.795 0.719 0.783  Wilks’ lambda  0.148 0.019 0.006 0.007 0.006  0.233 0.042 0.032 0.020  M  E  13.911 4.550 2.065 1.504  E  26.976 3.455 1.044 1,333  11.499 2.057 1.557  Fucus  0.000 0.004 0.007 0.087  0.000 0.016 0.405 0.216  0.001 0.110 0.204  Table 5.2 The effect of size and season of experimental clearings on recolonization of the three dominant algae, Fucus, Pelvetiopsis, Mazzaella, and the fugitive algal group.* Repeated measures ANOVA was applied to 16 months of data. Large plots (20x20 cm) were not included in analysis as discussed in text. All data were arcsine transformed prior to analysis.  0.850 0.888 0.843 0.872  Wilks’ lambda  analysis  0.012 0.002 0.002 0.002 0.002  0.013 0.034 0,019 0.009  MS  4.808 3.443 1.608 1.279  E  6.060 1.162 1.135 1.016  1.380 3.572 2.041  E  Pelvetiopsis  0.001 0.01 1 0.089 0.230  E  0.000 0.325 0.337 0.427  0.243 0.016 0.112  7 7 21 21 784  1 3 3 112  df  0.632 0.834 0.506 0.622  Wilk& lambda  0.081 0.017 0.026 0.017 0.006  0.259 0.049 0.117 0.029  MS  8.980 1.703 4.059  F  algae  8.819 3.019 3.893 2.607  E  12.809 2.673 4.060 2.637  Ephemeral  0.000 0.006 0.000 0.000  P  0.000 0.021 0.000 0,001  0.003 0.171 0.009  P  Univariate within subjects F and probability were the adjusted values after the Huynh-Feldt estimator (epsilon = 0.5347, 0.5 127, 0.7675, 0.7 147 for Mazzaella, Fucus, Pelvetiopsis, fugitive algal group, respectively). Systat (Wilkinson et al. 1992) provides automatically adjusted F values and the probabilities in the case of nonsphericity. See Myers and Well (1991) for details in calculating the adjusted degrees of freedom for each species.  *  Time Time x Size Time x Season Time x Size x Season  Effect  repeated-measures  4 4 12 12 448  Within subjects Time Time x Size Time x Season Time x Size x Season Error (Time)  Multivariate  1 3 3 112  df  Size Season Size x Season Error  Between subiects  Source of variation  Table 5.2 continued.  123  60 E o 0  50  0  C4Q  z w  > 0 0 F  w  0 0 Wi 0  0 30 ----O----RJC  ---—&----  E  PEL  I  C.)  0  o20•  (\J  cc W  (I)  I—  D  cc 10  0 W  cc  I 0  J I  I  oc  I  I  b  Fig. 5.1 Seasonal changes of algal percent cover and the number of fucoid recruits in unmanipulated plots (20x20 cm) at Prasiola Point for two years. Data are the mean ±SE of four replicates. EPH for ephemeral algae, MAZ for Mazzaella cornucopiae, FUC for Fucus distichus, PEL for Pelvetiopsis limitata.  124  w > 0 0 I— z w 0 w 0  2  4  6  8  10  12  14  16  18  20  MONTHS AFTER CLEARING  Fig. 5.2 Recolonization of total algae in three different sizes of clearing. Large clearings = 20x20 cm, medium clearings = lOxlO cm, small clearings = 5x5 cm. Plots cleared in the summer (August 1991) were only compared for the size effect on patch recovery. Data are the mean ±SE of fifteen replicates for small and medium plots and of four replicates for large plots.  125  30-  Fucus  Mazzaella  I  20-  10-  Ui >  T  0-  0 0  I  I  I  I  I  I  I  I  I  I  I  I  I  I  z  Ui  0  30-  Ephemeral algae  Pelvetiopsis  Ui 0  20-  I  LARGE MEDIUM SMALL  —C--— --—0—-----0----  10.  T T  0-  i—-I  I  I  I  -I  I  I  2 4 6 8 101214161820  I  I  I  2  4  6  I  I  I  I  I  I  I  8 101214161820  MONTHS AFTER CLEARING  Fig. 5.3 Patch recovery of each dominant algae, Mazzaella, Fucus, Pelvetiopsis, and ephemeral algae in three different sizes of clearing. Data shown are from the summercleared (August 1991) plots. Data are the mean ±SE of fifteen replicates for small and medium piots and of four replicates for large plots.  126  30-  Mazza ella  Fucus —U--—-.0.--------0-------&---  20-  10.  SPRING SUMMER FALL WINTER  I  T  I  w>  0  0-  C-) I—  I  z  I  I  I  I  I  I  I  I  I  I  I  IlillIllIll  II  w  0  a: U]  -  Ephemeral algae  Pelve tiopsis  20-  T  H  10-  I 0-  oc  vW F,  Fig. 5.4 The effect of season of clearing on algal recovery of each species in small plots (5x5 cm). The time for each seasonal clearing is August 1991 (for summer), October 1991 (for fall), February 1992 (for winter) and April 1992 (for spring). Data are the mean ±SE of fifteen replicates.  127  30•  Mazza ella  Fucus  20  I  T 10.  >  T  T  11  0•  0 C.) I  I—  z  w C) 30. w 0  20  I  I  I  I  I  I  III  II  Pelvetiopsis -D----0-------.---.-:  I  I  I  I  I  I  I  I  I  I  Ephemeral algae  I  SPRING SUMMER FALL WINTER  I  10. T  0-  Fig. 5.5 The effect of season of clearing on algal recovery of each species in medium plots (lOxlO cm). See Fig. 5.4 for captions and the number of replicates.  128  A. BARNACLE 20(  150  100 c’J  E  C) C 0  a: w a C’)  w  F-  B. LIMPET 10  (I)  z w ID  8  6  4  2  MONTHS AFTER CLEARING Fig. 5.6 Barnacle and limpet densities in the three sizes of patch at three sampling dates (4, 10, 16 months after clearing). Data are the mean ±SE of fifteen replicates for small and medium piots and of four replicates for large plots. When results of ANOVA on three sizes were significant, results of Tukey’s HSD tests were labeled (e.g., S LM indicates that density in small patch is significantly greater than that of either large or medium patch). N/S for p>O.05.  129 CHAPTER 6  PREDICTIVE MODELS FOR COMMUNITY STRUCTURE FOLLOWING ENVIRONMENTAL CHANGES  The objectives of this chapter are to synthesize the major interactions among all the component species in the community and to predict the structure of the community when the presence or density of individual components is altered.  THE MECHANISMS AND COMPLEXITY OF SPECIES INTERACTIONS  The results presented in this thesis demonstrate that there are detectable biological interactions among all component species in the community and that these interactions can have important consequences for the specific patterns of distribution and abundance of each organism. The major interactions among the component species are summarized in Fig. 6.1. Competition was the predominant interaction occurring between algal species (A, G, H, J, N, 0). Competitive interactions in this community largely comprised two different mechanisms: preemption (A, J, and G) and shading (H). In the absence of limpets ephemeral algae rapidly covered the substratum at the early successional stage and prevented the settlement of the later colonists such as fucoids. However, the seasonal appearance and fluctuating abundance of the ephemeral algae allowed fucoids to gradually invade space; thus the fucoids eventually outcompeted the ephemerals. Therefore, the interaction (A) occurs under two conditions: at the early successional stage and in the absence of limpets. As long as a disturbance occurs and succession proceeds in the gap, the first condition will occur. Reduced limpet density can occur as a result of environmental impacts (such as introduction of certain predators, harvesting by humans).  130 The possible consequence of this situation on community structure is discussed in the next section of this chapter, where key species in this community are considered. The effect of space preemption by Mazzaella on fucoid recruitment (G) was measured in a mature community, unlike the interaction (A). Low recruitment of the two fucoids within the Mazzaella beds indicated that a dense turf of this red alga blocked fucoid recruits and/or inhibited the growing recruits (0), although removal of Mazzaella turf did not significantly increase fucoid recruitment. This indicates that the outcome of an interaction may not be consistent in the natural habitats and that such an outcome is often correlated with other factors which sometimes are not easily detectable. The interaction (J) between Mazzaella and the post-recruitment stage of Fucus requires two assumptions to explain why the lifespan of Fucus within the Mazzaella turf was relatively short. The first assumption was that there was limited space for holdfast expansion as Fucus grew because the basal crusts of the Mazzaella turf preempted space. The second was that the relatively wider thallus of Fucus (compared to Pelvetiopsis) might be subjected to a greater drag force by wave action. Therefore, the outcome of interaction (J) is a combination of space preemption and of a physical factor. Both fucoids affected Mazzaella by forming a canopy (H) when the fucoids successfully settled and grew to an adult plant within or near Mazzaella beds. The shading effect of high intertidal species of fucoids on other neighbor species is rarely reported (but see Schonbeck and Norton 1980) in algal interactions. The interaction (N) has been observed later in the successional sequence, as the percent cover of Pelvetiopsis decreased as that of Fucus coincidentally increased in a few plots. The mechanism for this interaction is unknown. A similar pattern between these two morphologically similar species has been observed in Sousa’s (1984) work. Based on Sousa’s result, Chapman (1995) proposed a generalization for the common sequence of succession in high shores of the northeast Pacific, e.g., diatoms  ->  Ulva -> Pelvetiopsis  ->  Fucus. However, the mechanism of the  interaction between Fucus and Pelvetiopsis was not described in either Sousa’s research or  131 this thesis. My impression is that Pelvetiopsis is a better adapted species to the stressful high intertidal habitat than Fucus, since Pelvetiopsis grew independently of the presence of Mazzaella and barnacles. A difference, if present, between these two species would be that Fucus might have a more successful reproductive strategy (e.g., more output of propagules) and faster growth rate than Pelvetiopsis. To date, this is a pure speculation. In Chapman?s recent review (1995), there are no data available on these variables. With the information from this study, I suggest two general features of algal interactions: reversal of competitive dominance and the importance of morphological characters. First, reversal of competitive dominance can be a common phenomenon among benthic macroalgae. This was the case of interactions between ephemeral algae and fucoids (A and 0) and between Mazzaella and fucoids (G and H), hence in four cases out of six studied. Reversal occurred as succession proceeded or the life history stages of competitors changed, or sometimes due to facilitation by a herbivore (L). This bi directional competition was one of the reasons that the species maintains its diversity in the  system. Second, competitive dominance was related to morphological characteristics of plants. The common characteristics of species capable of space preemption were fast growth, spreading habit and/or turf-forming, whereas species capable of shading were large and/or of erect form. These results provide useful evidence for the relationship between plant traits and competitive dominance which has not been extensively investigated in algal competition (Olson and Lubchenco 1990). Herbivory by both limpets (B) and snails (L) had some similarities. First, both herbivores expressed selectivity for food; limpets preferred ephemeral algae and snails preferred Mazzaella. Second, herbivory was closely related to competitive interactions among the prey species. More importantly for their role at the community level, limpets facilitated algal succession and snails influenced in part the reversal of the competitive direction. Facilitations or positive interactions occurred in the system and these can be divided into two types depending on their mechanisms: direct and indirect interactions. Direct  132 facilitation was observed at three places. The interaction (K) was based on field observation and other studies (Boulding and Van Alstyne 1993) in which a dense turf of Mazzaella provided a refuge for snails against wave impact and desiccation. In interaction (1), Mazzaella enhanced Fucus survivorship for those individuals which grew at the edge of the Mazzaelkz turf. The preceeding two cases of direct facilitation were mechanically related to amelioration of harsh physical conditions. However, in the interaction (F), barnacles directly facilitated Fucus settlement; this was not related to physical stress, but rather to the presence of limpets. This facilitation was effective only when limpets reduced ephemeral algae which inhibited Fucus settlement. In other words, the intensity of the barnacle’s facilitation of Fucus (F) was less strong than the intensity of competition (A) between ephemerals and Fucus because the effect of the facilitation was not detectable in the absence of limpets. Both indirect facilitations (C) were coffelated with grazing. Limpets facilitated fucoid colonization by reducing the abundance of the competitors of Fucus, ephemeral algae. Snails also enhanced fucoid recruitment by reducing the percent cover of Mazaella, a species which also inhibited fucoid recruitment. Some evidence has been observed that algae had a deleterious effect on barnacles (D). Ephemeral algae (e.g., benthic diatoms, Enteromorpha spp.) were observed to overgrow, and eventually cause mortality of, barnacles, but this interaction required a precondition of no limpets existing in the system. However, the vegetatively propagating and laterally growing Mazzaella was often found to overgrow barnacles which grew in the vicinity of its turf, despite the presence of limpets. Barnacles also suffered from the foraging activity of limpets (E), which might bulldoze the propagules of barnacles. In conclusion, the overview of major interactions among all component species indicates that the current structure of the community is a consequence of a complex and non-hierarchical interaction network, which includes reversal of dominance and indirect interactions. The non-hierarchical interactions among algal species, and their well-balanced  133 responses to disturbance (e.g., different capabilities of recovery against the different sizes of disturbance), suggest a high likelihood of maintaining diversity in this algal community.  PREDICTIONS OF THE EFFECTS OF POTENTIAL ENVIRONMENTAL IMPACTS ON THE COMMUNITY STRUCTURE  A primary goal of empirical studies in ecology is to validate and increase the precision of predictive models for the structure of communities, so that potential environmental impacts might be assessed (Dayton 1971, Paine 1974, Menge 1976, Underwood et al. 1983). Environmental impacts which may come in a predictable way (e.g., seasonal desiccation or predator fluctuation) or stochastically (e.g., oil spills, storms) may result in a considerable reduction or depletion of certain species in the system. Such a change will influence other species which are directly or indirectly linked to that organism, and will eventually lead to reorganization of the community. However, predicting the structure of a community after a particular disturbance remains difficult, even if patterns of interactions among all major component species are known, because the outcome of biological interactions is variable in time and space, as demonstrated in many experimental results. Thus, it is difficult to examine the relative intensity of interactions in a nonhierarchical interaction network. Nonetheless, it may be instructive to assess how far the present experimental analysis of this community has allowed accurate predictions to be made about the fate of the community in response to diverse disturbances, or simply as to its persistence through time. Below, I finish my thesis with some predictive models about the effect of potential environmental impacts on community structure. Possible scenarios of community-level responses leading to reorganization of existing components were predicted in the light of the results of my in situ experiments. The predictions for cases 1 to 3 were made on the established (mature) community, while for cases 4 and 5 the predictions are for a newly formed patch.  134 CASE 1: Absence of limpets in the system. The first response of the community to the removal of limpets may be a dramatic increase of ephemeral algae. Subsequently, abundance of Pelvetiopsis, Fucus and barnacles will decrease due to the competitive superiority (possibly temporary) of ephemeral algae for space. The reduction by ephemeral algae may also be applicable to Mazzaella (Olson 1985, 1992). Competition for space between the ephemeral algae and other component species will be more intense during the growing season of ephemerals (October to April) than at other times of the year. Therefore, the recruitment of fucoids in October (the peak of recruitment for both fucoids) can be severely affected by ephemeral algae (especially benthic diatoms; Fig. B.2) rapidly preempting space. The maintenance of the fucoid population will be highly dependent upon the seasonal gap in abundance of the ephemeral algae (May to September) and upon at least some of the fucoid recruits germinating in this season. With the increasing abundance of ephemeral algae, Mazzaella would be more affected than any other dominant macroalgae because it is smaller than fucoids and reproduces largely vegetatively. The abundance of barnacles will gradually increase especially when ephemeral algae are low in summer. Therefore, the community will be dominated by Ephemerals-Fucus-Pelvetiopsis and the relative abundance of these three components will be influenced by the degree of reduction of limpet density. Barnacle abundance is expected to fluctuate with the seasonal cycle of ephemeral algae.  CASE 2: Absence of Mazzaella in the system. The existence of Mazzaella in the system has a negative effect on fucoid recruitment and a positive effect on Fucus survival. Unlike the effect of ‘total pruning’ of Mazzaella blades (Chapter 2), the complete removal (including basal crusts) of this alga will apparently cause increasing recruitment of both fucoids. Particularly, direct facilitation of Fucus recruitment by barnacles (F) becomes possible because limpets exist in the system (see further explanation in the previous section of this chapter). However, for post-  135 recruitment survivorship, Fucus is at a relative disadvantage compared to Pelvetiopsis because no habitat amelioration by Mazzaella is available. The relative abundance of Fucus and Pelvetiopsis in the community becomes a function of the relative intensity of the two interactions (F and I); I did not determine whether competition between these fucoids exists or not (except for the interaction (N)). Since there is evidence that the barnacle, Balanus glandula, enhanced the recruitment of Pelvetiopsis limitata (Farrell 1991), it is assumed that the interaction (F) may also apply to Pelvetiopsis at my study sites even though the data were not available from this study. If this assumption holds, Fucus abundance would be affected by the interaction (M) because of the absence of Mazzaella’ s protection; therefore, Pelvetiopsis wiil be relatively more abundant than Fucus in the system. The density of barnacles wifi increase because there is no Mazzaella overgrowth, but the density of snails will be much reduced in the habitat because no preferred food and no refuge exist. So, in the absence of Mazzaella the community will be reshaped into a Pelvetiopsis-barnacle dominated structure.  CASE 3: Absence of Fucus in the system. Mazzaelkj abundance will probably increase because of the reduced shading effect by Fucus. However, the abundance of Pelvetiopsis will be variable depending on its relationship with Fucus and the relative intensity of direct and indirect interactions. If Pelvetiopsis is competitively inferior to Fucus (the interaction N), the abundance of Pelvetiopsis will increase. If they are in a mutualistic relationship (mutually increased survivorship under strong desiccation and wave impact), Pelvetiopsis density will decrease when Fucus is absent. On the other hand, increased Mazzaella will negatively and directly affect Pelvetiopsis (G). Therefore, the possible outcomes in the case of Pelvetiopsis are unclear, indicating the need for information on this Fucus-Pelvetiopsis interaction. A similar uncertainty will appear in the case of the absence of Pelvetiopsis.  136 CASE 4: No Mazzaella when a small/medium (25-100 cm ) gap is formed in 2 summer. A possible scenario of succession is shown in Fig. 6.2. In the fall following the disturbance, limpets and ephemeral algae may be the first organisms appearing in the gap. Ephemeral algae (e.g., diatoms) will, however, be minimal due to grazing by limpets. In winter, limpets still control the abundance of ephemeral algae but Porphyra spp. may appear in small amounts because they seem to be a less prefered species than diatoms (Table 4.2). This negative effect of limpets on ephemerals may indirectly facilitate barnacles although limpets are also capable of grazing on young barnacles (Dayton 1971, Underwood et al. 1983, this study). As barnacles become larger in the spring, barnacles may enhance (facilitate) settlement of all algal species (especially fucoids). The indirect facilitation of fucoids by limpets may occur by limpets grazing on ephemeral algae which inhibit fucoid settlement. In summer months, ephemeral algae will disappear but perennial fucoids will remain. At this time limpets, excluded by barnacles in the patch, may migrate to a lower zone to forage for alternative foods (Olson 1992). A possible reduction of the Fucus population, due to both strong desiccation and the absence of Mazzaella, can be expected. The patch will ultimately be dominated by Pelvetiopsis-Fucus-barnacles. However, occasional inhibition of fucoids on barnacles can be expected as fucoids grow on the tests of barnacles; increased drag forces on adult fucoids (especially Fucus) may remove both plants and barnacles (Jernakoff 1985, Farrell 1991). Further invasion of ephemerals will be minimal because of fucoid preemption of space and a possible reintroduction of limpets. Thus, limpets may not recover their initial abundance in the plots because the patch is already full of perennial fucoids and adult barnacles.  CASE 5: No limpets when a large gap (e.g., 400 cnz ) is formed in winter. 2 A likely scenario of succession is shown in Fig. 6.3. In the spring following the disturbance, only a few ephemeral algae will appear in the plot along with a few juvenile  137 barnacles. Barnacle density will be slightly increased in the summer but no other organisms may be visible. However, in the fall, a rapid colonization of some ephemeral algae (probably diatoms) will occur along with a few recruits of fucoids. At this time, diatoms may effectively interfere with all other organisms in the plots by preventing new settlement of other algae and by killing existing barnacles (Castenholz 1961), In the late spring (April-June), as ephemeral algae decrease, barnacles may re-establish on the substratum and facilitate fucoid (maybe Fucus) colonization. Where space is limited, competition between fucoids and ephemeral algae may occur. In the summer, presumably desiccation-resistant Pelvetiopsis would be more abundant than Fucus; this pattern might vary depending on microhabitat conditions (e.g., slope, direction, wave-exposure). Barnacles may benefit from the fucoid canopy in that the latter reduces heat and desiccation (Dayton 1971, Farrell 1989). Ephemeral algae retain a low abundance. This summer structure will alternate with the other patterns which are typical in fall, winter and spring. The typical community organization in the latter seasons may be that Fucus, and ephemeral algae with similar abundance, compete for space. Fucus may inhibit both Pelvetiopsis (as in the interaction N in Fig 6.1) and also barnacles by pulling the latter off the rock substratum when it (Fucus) is removed by drag force (Jernakoff 1985, Farrell 1991). If Mazzaella invades the plot, this alga will be kept in check by both fucoids (the interaction H in Fig. 6.1). The accuracy of the above predictions on community reorganizations under various environmental conditions can be affected by at least the following three factors: 1) the unpredictability of the timing or intensity of recruitment of any component species, 2) the proximity of reproductive adult plants, and 3) the spatial variability of the type and intensity of biological interactions. The unpredictability of recruitment events has been pointed out as a factor influencing spatial pattern in marine community structure (Paine 1979, Underwood et al. 1983, Sousa 1984, Reed et al. 1988). In the present study, the recruitment of Pelvetiopsis was unexpectively low in the October 1992, which was  138 considered as the time for recruitment of this alga (Fig. 5.1). The variation in patch colonization and early species composition may depend largely on the proximity of reproductive plants (Sousa 1984). It may be difficult to predict structure with any accuracy if the vegetation surrounding a gap created by a disturbance is monospecific, because the distance of spore dispersal of some species (e.g., Fucus, Pelvetiopsis) is known to be <1 m (Menge et al. 1993). Also, my data indicate that there were some possibilities that the type and intensity of biological interactions can be changed depending on wave-exposure gradients (e.g., Mazzaella-Fucus interaction). As suggested by Underwood et al. (1983), for future experimental manipulations in the study of intertidal communities, examination of the range of potential interactions at the various naturally occurring densities is necessary to construct more powerful predictive models. The interactions and mechanisms reported in this thesis improve our understanding of the functional relationship among the different species in this assemblage and enable such predictions to be made for a common high intertidal community.  139 REFERENCES Boulding, B. 0., Van Aistyne, K. L. (1993). Mechanisms of differential survival and growth of two species of Littorina on wave-exposed and on protected shores. I. Exp. Mar. Biol. Ecol. 169: 139-166. Castenholz, R. W. (1961). The effect of grazing on marine littoral diatom populations. Ecology 42: 783-794. Chapman, A. R. 0. (1995). Functional ecology of fucoid algae: twenty-three years of progress. Phycologia 34: 1-32. 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. Farrell, T. M. (1989). Succession in a rocky intertidal community: the importance of disturbance size and position within a disturbed patch. J. Exp. Mar. Biol. Ecol. 128: 57-78. Farrell, T. M. (1991). Models and mechanisms of succession: an example from a rocky intertidal community. Ecol. Monogr. 61: 95-113. Jernakoff, P. (1985). The effect of overgrowth by algae on the survival of the intertidal barnacle Tesseropora rosea Krauss. J. Exp. Mar. Biol. Ecol. 94: 89-97. Menge, B. A. (1976). Organization of the New England rocky intertidal community: role of predation, competition, and environmental heterogeniety. Ecol. Monogr. 46: 355-393. Menge, B. A., Farrell, T. M., Olson, A. M., Van Tamelen, P., Turner, T. (1993). Algal recruitment and the maintenance of a plant mosaic in the low intertidal region on the Oregon coast. J. Exp. Mar, Biol. Ecol. 170: 91-116. Olson, A. M. (1985). Early succession in beds of the red alga, Iridaea cornucopiae Post. & Rupr. (Gigartinaceae): Alternate pathways. M.S. Thesis. Oregon State University. Olson, A. M. (1992). Evolutionary and ecological interactions affecting seaweeds. Ph. D Thesis. Oregon State University. Corvallis, Oregon Olson, A. M., Lubchenco, J. (1990). Competition in seaweeds: linking plant traits to competitive outcomes. 3. Phycol. 26: 1-6. Paine, R. T. (1974). Intertidal community structure. Experimental studies on the relationship between a dominant competitor and its principal predator. Oecologia 15: 93-120. Paine, R. T. (1979). Disaster, catastrophe and local persistence of the sea palm Postelsia palmaeformis. Science 205: 685-687. Reed, D. C., Laur, D. R., Ebeling, A. W. (1988). Variation in algal dispersal and recruitment: the importance of episodic events. Ecol. Monogr. 58: 32 1-335.  140 Schonbeck, M. W., Norton, T. A. (1980). Factors controlling the lower limits of fucoid algae on the shore. J. Exp. Mar. Biol. Ecol. 43: 13 1-150. Sousa, W. P. (1984). Intertidal mosaics: patch size, propagule availability, and spatially variable patterns of succession. Ecology 65: 1918-1935. Underwood, A. J., Denley, E. J., Moran, M. J. (1983). Experimental analyses of the structure and dynamics of mid-shore rocky intertidal communities in New South Wales. Oecologia 56: 202-2 19.  .  5 Oco  55 S  PELf  EPH  EPHf  A-  ———  N co —  _&co  FUC  Al  f  M  ..-  Solid line Dashed line Arrow head Round head Clear point Dark point  strong interaction weak interaction negative interaction positive interaction indirect interaction direct interaction  LEGEND  WAVE IMPACT DESICCATION  Fig. 6.1 The summary of interactions among the major component organisms shown in this study. PEL and FUC in the upper diagram indicate Pelvetiopsis and Fucus at the recruitment stage, while those in the lower indicate the fucoids after recruitment stage. Small letters are the abbreviations for types of interaction: (co) for competition, (fa) for facilitation, (gr) for grazing, (ov) for overgrowth, (pr) for preemption, (pt) for protection, (re) for refuge, (sh) for shading.  I  I  I  ) WINTER ‘ ‘  : strong interaction : weak interaction : negative interaction : positive interaction : indirect interaction : direct interaction  LEGEND  SPRING  Solid line Dashed line Arrow head Round head Clear point Dark point  >  SUMMER  4? FALL  Fig. 6.2 A predictive model for community structure, in the absence of Mazzaella, when a small/medium gap (e.g., 5x5 to lOx 10 cm) is created in summer. The size of the circles represents the relative abundance of component species.  FALL  4  FALL WINTER  > ‘  SUMMER  LEGEND  >  Solid line strong interaction Dashed line weak interaction Arrow head negative interaction Round head positive interaction Clear point indirect interaction Dark point direct interaction  SPRING  FALL WINTER SPRING  Fig. 6.3 A predictive model for the community structure, in the absence of limpets, when a large gap (e.g., 20x20 cm) is created in winter. The size of circle represents the relative abundance of component species.  SPRING ESUMMER  .  I I S S I * S S  S  (J  144 APPENDIX A ANOVA tables for the natural densities of snails from Chapter 3 Table Al Summaries of ANOVAs on the natural density of snails at Prasiola Point. A. densities of snails in monospecific patches of the three dominant algae. B. densities of snails in general habitats (3 transect lines). A.  ANOVA table  Source  -  monospecific patches  MS.  Species  2  151.49  163.79  0.0000  Time Species*Time  8  30.83  33.34  0.0000  16  4.35  4.71  0.0000  108  0.92  Error  B. ANOVA table Source  -  general habitats  MS.  Transect  2  13.75  7.81  0.0006  Time Transect*Time  8  15.81  8.98  0.0000  16  3.31  1.88  0.0278  132  1.76  Error  145 APPENDIX B Supplementary data and results from other studies in Chapter 4  MODEL OF SUCCESSION C C  C.)  .2 ci) .  Ui  0 I-  CU  U-  Early Successional Species  >  0 Ui  Cl) Ui  Equivalent  +  0  C)  Iii 0 Cl)  Later Successional Species  Figure B.1 Farrell’s model for the effect of consumers on the rate of succession ( consumers increase the rate of succession, 0 no effect, + consumers decrease the rate of succession) cited from Farrell (1991).  -  -  -  146  A. +B-L 80-------0 -0 ----h----  60-  -  -  --  -  —D-—— (I) I—  Porphyra Urospora Enteromorpha Mastocarpus Diatoms  40-  0 -J 0  E  20  0 C”  0-  0 0  z  I  I  I  w  >  0  C-)  B. -B-L 80-  z  w 0 w  60-  0----0-------h---—  —  ——  — -  —C--—--  Porphyra Urospora Scytosiphon Callithamnion Diatoms  40-  20-  0.  ç.  ob  Fig. B.2 Dynamics of ephemeral algae in limpet-excluded piots (-L), with (+B) and without (-B) barnacles.  147  Dunaan (1986)  V  N  Van Tamelen (1987)  Dethier & Duaciins (1984)  LEGEND Solid line : direct interaction Dashed line : indirect interaction Arrow head : negative interaction Round head : positive interaction ALG1 : microalgae ALG2 : macroalgae CHI : chiton BAR : barnacle LIM : limpet  Fig. B.3 Direct and indirect interactions shown in other studies.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088097/manifest

Comment

Related Items